Youtube end to end and each platform playbook

Dennis Ksendzov

Founder · September 30, 2026 · 70 min read

Ridge has spent about 14 million dollars on YouTube partnerships and thinks it could have gone harder.
William Painter scaled to 15 million a year, almost entirely on YouTube.
A Rogan sponsorship that went live last November was still producing purchases eight months later
The Smarter Every Day video last June brought almost half a million dollars, and topline revenue visibly jumped about 100 grand the day it went live.
For every incremental Shopify order YouTube produced 2.8 incremental Amazon orders.
1b · What you'll learn
- All aobu tyoutube sponsorshps minimize risk increase consistency - few brands crack the code
1c · The problem: the click number says YouTube does not work
- At Google conferences Cody hears buyers say they want to run YouTube upper funnel stuff but cannot get CFO buy-in because the ROAS looks worse than Meta.
- HexClad has kept YouTube at 1 to 1.5% of budget because it could not validate the channel.
- HexClad could not get YouTube creator CPMs under 40 to 50 dollars and often paid 75 to 100.

Part 2 · Who we are and why we dug in

2a · Who we are
- A health & wellness influencer marketing agency.
- We help wellness & health brands scale their existing influencer programs 3-10x in under 18 months.
2b · Origin story: the good creators kept getting more expensive
- As an influencer agency, the good creators kept getting more expensive.
- Dennis sold sponsorships for creators before he bought them for brands, so he knows where the padding hides in a rate
2c · So we went looking for the real playbook
- To find the playbook in the weeds, we talked with the CMO of AG1, the CEO of Onnit & the CMO of Feastables.

Part 3 · Two YouTube mistakes and the audit checks

3a · Mistake 1: YouTube has to pay back on the click, the way Meta does
- Because of the long tail and the brand value, Ridge gives YouTube creator partnerships a materially lower target ROAS than paid social.
- Jones Road set its YouTube target 40% below its Meta target on that reasoning and scaled from there.
- Precedent's model splits HexClad's YouTube prospecting revenue as 0.41% direct, 34% Shopify halo, and 65% into Amazon.
- Measure it in weeks, not days.
3b · Mistake 2: the cheapest views are the best deal
- Ridge pays a much higher CPM for Smarter Every Day because it skews older and more affluent and sometimes outperforms everything.

Part 4 · The four steps for YouTube sponsorships

4a · The four steps on this page

Part 5 · How to buy and run YouTube sponsorships

Step 1: -how to more effectively use influencer marketing on youtube
Step 2: platform tactics and strategies
Step 3: behcnmarks u can compare yourself too
Step 4: TikTok and LinkedIn next to YouTube

Part 6 · Testing YouTube first, then winning more budget

6a · Testing YouTube for the first time
- Connor MacDonald's advice for a first creator test is to concentrate the whole budget on one launch so you can read it.
- The agency recommended a 50 thousand dollar minimum to test the channel.
- The cheap validation before an incrementality tool: as YouTube spend scales, the share of post-purchase survey respondents who say they found you on YouTube should grow too.
6b · Getting more budget for YouTube
- The one thing every brand profitable on YouTube shares is a way to measure halo.
- Percent of new visitors in Northbeam justifies accepting a lower one-day-click ROAS on an upper funnel channel.
- Ridge shares YouTube viewer audiences back into search so it can see which branded queries came from YouTube.

Part 7 · When YouTube sponsorships do not pay back

7a · Where YouTube does not work
- Cody does not think YouTube works for every brand, and specifically doubts it at low AOV.
- HexClad copied Ridge's flat fee YouTube creator model and it did not produce performance.
- HexClad concentrated 85% of its flat fee creator deals in cooking, the most expensive possible category for it.
7b · What stays hard
- HexClad cannot stop discount code leakage regardless of the mechanic it uses: CleanCart shows around 500 attempted injections.
- Influencer attribution is by far their hardest, partly because YouTube changed the UI and clicking through got harder
- If every dollar must be attributed to an immediate click, the model will not suit your organization.
Ridge has spent about 14 million dollars on YouTube partnerships and thinks it could have gone harder.
William Painter scaled to 15 million a year, almost entirely on YouTube.
A Rogan sponsorship that went live last November was still producing purchases eight months later, all attributed through the UTM and that discount code.
The Smarter Every Day video last June brought almost half a million dollars, and topline revenue visibly jumped about 100 grand the day it went live.
For every incremental Shopify order YouTube produced 2.8 incremental Amazon orders.

1b · What you'll learn

YouTube sponsorsihps in and out
What the best in the industry you're doing, how you should be pricing deals, CPM benchmarks that you can use for negotiation, as well as how cross-platform works and where this fits in with your marketing budget.
- All this is taken from brands who are active in the space, so it's not ChatGPT data, it's actual actionable advice that you can implement to see results.

The problem many attirubution models confuse youtube

At Google conferences Cody hears buyers say they want to run YouTube upper funnel stuff but cannot get CFO buy-in because the ROAS looks worse than Meta.
HexClad has kept YouTube at 1 to 1.5% of budget because it could not validate the channel.
HexClad could not get YouTube creator CPMs under 40 to 50 dollars and often paid 75 to 100.

Part 2 · Who we are and why we dug in

Before trusting any YouTube pricing advice, know who is giving it, a wellness influencer agency that watched good creators get more expensive every year.

2a · Who we are

We work on brands that already pay creators, so the advice comes from growing real programs rather than starting from zero.

A health & wellness influencer marketing agency.
We help wellness & health brands scale their existing influencer programs 3-10x in under 18 months.

2b · Origin story: the good creators kept getting more expensive

Rising creator prices are why pricing matters at all, and having sold sponsorships first means knowing which part of a rate is padding.

As an influencer agency, the good creators kept getting more expensive. Every year the CPMs went up, and now it's not uncommon to see $50 or $100 CPMs, especially in educational spaces.
Dennis sold sponsorships for creators before he bought them for brands, so he knows where the padding hides in a rate

2c · So we went looking for the real playbook

Talking to the leaders who approve the spend at large brands grounds the playbook in what operators actually do with their money.

To find the playbook in the weeds, we talked with the CMO of AG1, the CEO of Onnit & the CMO of Feastables.

Part 3 · Two YouTube mistakes and the audit checks

With the source clear, the two beliefs that make brands misjudge YouTube sponsorships come next, then the checks that show a deal earns back at the right price.

3a · mistkae 1: YouTube has to pay back on the click, the way Meta does

A click only read sees a tiny slice of the revenue YouTube brought HexClad, with most of it landing later on Shopify and Amazon.

Because of the long tail and the brand value, Ridge gives YouTube creator partnerships a materially lower target ROAS than paid social.
Jones Road set its YouTube target 40% below its Meta target on that reasoning and scaled from there.
For example: Precedent's model splits HexClad's YouTube prospecting revenue as 0.41% direct, 34% Shopify halo, and 65% into Amazon.

3b · mistake 2: the cheapest views are the best deal

The price of a view says little on its own, because who watches and how ready they are to buy decide what the view returns.

He argues a double CPM on YouTube can still win because of the intent that Instagram and TikTok lack.
Ridge pays a much higher CPM for Smarter Every Day because it skews older and more affluent and sometimes outperforms everything. this video actually brought in 500,000 views. Definitely outperformed their investment by quite a bit.

3c · Main points from the audit

Eight checks give a brand a pass or fail on its own YouTube creator spend, from earning back and price to vetting and budget size.

Break even. Your influencer sales earn back what you spent on creators (you break even on CAC). Once a creator breaks even, you scale their content with Meta ads (new hooks and cta’s added).
CPM. Your creator CPM is no higher than 3x your Meta ads CPM.
Post attribution. You ask every buyer after checkout where they first heard of you, with your creators' names listed.
Vetting. Ultra important: You get each creator's YouTube audience screenshots before you pay. (or live recording even better*) Every creator you pay has regular viewers above 1 to 3% of their YouTube audience.
Creator growth and retention. At least 30% of your YouTube attributed sales come a month or more after the video goes up (videos should be evergreen).
Running winners as ads. Organic posts that breakeven or perform better have 10x the spend put into ads
Multi platform. You mark the days a big creator video took off and check your Amazon and store sales on those same days.
Budget size. Your creator spend over the last 12 months, counting fees, free product and the ads you ran on their posts, is at least 1% of your revenue.

Part 4 · The four steps for YouTube sponsorships

Once the mistakes and checks are clear, four steps lay out how to buy YouTube sponsorships that earn back, priced and measured against other platforms.

4a · The four steps on this page

Each step builds on the last, so the strengths come before the tactics, and the benchmarks come before any comparison with TikTok and LinkedIn.

-how to more effectively use influencer marketing on youtube
platform tactics and strategies
behcnmarks u can compare yourself too
TikTok and LinkedIn next to YouTube

Part 5 · How to buy and run YouTube sponsorships

With the four steps named, each one is taught in turn, starting with where YouTube sponsorships win and fail, then the deals and tactics brands use.

Step 1: -how to more effectively use influencer marketing on youtube

Knowing where YouTube wins and where it fails tells a brand whether a sponsorship can earn back before any money goes into a deal.

Step 1 · YouTube strenghts

Trust in the creator and videos that keep getting found let a YouTube sponsorship sell for years, while a TikTok post fades after a day.

HexClad rates YouTube as the best social channel for building a brand relationship because the content demands commitment. Best way to build a connection between the audience and the brand. The YouTuber typically has the most trust.
YouTube SEO content compounds where TikTok virality has to be repeated weekly. TikTok, every video gets views in the first 24 hours, then dies down, whereas YouTube is evergreen.
The thing brands miss about YouTube sponsorship is that it is evergreen.
The speaker believes that YouTube is where brands can drive the most meaningful connections with consumers, despite the high production value required.
Influencer was their most efficient channel and their best brand building channel at the same time

Cheap CPMs back then, little other advertiser demand on YouTube
Easy to work with, paid on time, loose brand guidelines
For a while the influencer team was their biggest team, doing mass outreach
Far more demand for YouTube now, so it is different [fix the ultimate confusion here]

He argues a much bigger CPM on YouTube can still win because of the intent that Instagram and TikTok lack.

He doubts it will work, but says test it with your money
Split 100k across core social channels, keep the best attribution
Double the CPM on YouTube could still perform better, who knows

Sources: Marketing Operators, E008: Your Questions Answered! Marketing Channel Expansion, Team Building & Influencer Marketing!, at 1:01:01 · Marketing Operators, E014: YouTube Deep Dive: Ads, Organic Growth, Partnerships & More, at 1:19:53

The thing brands miss about YouTube sponsorship is that it is evergreen.

Coupon codes and links tracked clicks, direct revenue and code usage
A creator talking to camera gave a longer ad read than Facebook
YouTube is evergreen, old videos still bring sales
Codes may have leaked to coupon sites, he admits

Example: Theo Von episodes from 2017 still produce sales today.

Worked with over 5,000 YouTubers
Peak year probably 2020, signed long deals when the world shut down
Creators needed revenue and more people were watching YouTube
Two full time US people plus VAs and scrapers, cold blasting lists

Step 1 · Where YouTube does not work

Knowing the limits stops a brand from spending on YouTube at a low order value or with a setup that returns nothing.

Cody does not think YouTube works for every brand, and specifically doubts it at low AOV.

Creative is probably harder than Meta, just a different kind
YouTube sits a little higher in the funnel
Probably needs a higher AOV, and measurement is the challenge

FIY for youtube ads - His warning is that YouTube will happily spend your money at zero return if the setup is wrong.

Less signal, more noise, so start with brand queries and post purchase surveys
You should see some people click through and purchase
Don't sit on a zero ROAS hoping it works somehow

At Google conferences Cody hears buyers say they want to run YouTube upper funnel stuff but cannot get CFO buy-in because the ROAS looks worse than Meta.

YouTube gives a halo effect that teams have to find and validate
there are some ways that you can do this, including testing through organic and then boosting the ads that work, will dive int o that later.

Sources: Marketing Operators, E014: YouTube Deep Dive: Ads, Organic Growth, Partnerships & More, at 39:32, 51:02

Step 1 · Brands that scaled on YouTube -

Ridge and William Painter show how far YouTube can go when a brand commits, and that a small team hunting unsponsored creators can run it.

Ridge has spent about 14 million dollars on YouTube partnerships and thinks it could have gone harder.

Ridge was early to YouTube partnerships and influencer
Pretty big team and a decent share of early budget
Wishes they had gone harder, like brands defined by podcasts

William Painter scaled to 15 million a year, almost entirely on YouTube.

Spent about 80% of their budget on YouTube
Eight figures exclusively with YouTube, around 2018 he thinks

They have worked with about 5,000 YouTube creators in total.

Started sponsoring YouTubers directly in 2016, alongside Squarespace
Crypto messed up the YouTube sponsorship market around 2021
They paid about $10 CPMs, afterwards quoted $3,000 CPMs
They could not compete in that market

For Ridge the company goal was to be every creator's first sponsor: if you had 10,000 subscribers and never done a brand deal, they would give you $50 and teach you how to work with brands.

Best accounts are usually the ones rising right now
They hunted new YouTube accounts that had never had a sponsor - still a good strategy today

Two full time people plus scrapers and VAs ran the entire sponsorship machine.

Over 5,000 YouTubers worked with so far
Peak year probably 2020, long deals signed when the world shut down
VAs and scrapers built lists, then cold blasted with clear instructions

Lower cost, higher risk, more arbitrage

TV needs $30 to 50K creative, then a $150K pilot
Most arbitrage, since product costs less than its value to creators
His vote for top of funnel off Meta is influencer, especially early

Sources: Marketing Operators, Channel Diversification in Ecommerce: Biggest Mistakes & Wins, at 34:34 · Marketing Operators, E014: YouTube Deep Dive: Ads, Organic Growth, Partnerships & More, at 24:59 · Open Residency, $200M CEO: How To Build and Scale Your Online Business in 2026 (Masterclass), at 3:22, 50:34, 53:03, 58:59

Celeb creator endorsement to amplify all other efforts:

Ridge picked a creator rather than a celebrity as the face of the brand because a celebrity is more disconnected.

Ridge was lacking a face for the brand
Searched for a flagship partnership to serve as that face
MKBHD, Marques Brownlee, became one of six people on Ridge's cap table alongside Connor and the three original founders, without the business ever selling equity.
A celebrity is the more traditional route, he notes

He values the deal in views: probably 500 million long form YouTube views this year alone, over a billion across the life of the deal.

Distribution is the asymmetric side of the deal
Views land in his content, with their products and ads
He puts that distribution at tens of millions of dollars

He puts the distribution alone at tens of millions of dollars of value.

Asks what a brand would pay for 500 million long form views
Probably won't reach 500 million this year
Over the deal, over a billion views with their products and ads

Sources: Marketing Operators, E014: YouTube Deep Dive: Ads, Organic Growth, Partnerships & More, at 1:25:17

Step 1 · More points

Who watches decides which brands YouTube suits, and an early sponsor like Ridge faced few rivals doing native integrations at the time.

YouTube over indexes in men. Ridge was really early to sponsoring YouTube creators. YouTube over indexes in men, and very few people were doing native integrations at the time.

Sources: Marketing Operators, May 2024, at 1:00:39

Up next: platform tactics and strategies.

Step 2: platform tactics and strategies

The tactics decide whether a sponsorship pays back, from how a deal is priced and timed to the hook, the discount code and the brand's own channel.

Step 2 · Your own YouTube channel

A brand channel can make videos that later become its best ads, as HexClad found with Gordon Ramsay, so paying creators is not the only route.

HexClad's organic strategy is recipe content, and it is modelled on Huckberry's Dirt and Yeti Presents.

How to cook a steak, perfect eggs, scallops
Long form series in production to rival those brands' series

Matt, HexClad's head of content out of food media, sits under brand, while community sits under growth.

Matt is leading organic YouTube strategy
More internal content resources will report into him
Lots of overlap between brand, community and growth
Building programs to merge loyalty and content

Jones Road started in November shooting educational videos per hero product, Miracle Balm, Foundation, face pencils, and they accumulate a few hundred thousand views each.

Not as far along as Hex Clad, but hiring and building it out
Shooting a lot in house since the brand launch
Bobby is very good at the education
Views keep racking up over time

celeb/creator endrosement terriotiry
The best sign of a successful organic YouTube video for HexClad is the algorithm picking it up, not the paid distribution.

Top recipe videos got picked up and hit millions of views
Then email and organic Instagram and TikTok posts send people to it
Paid media goes behind it if it could be a good ad

Gordon Ramsay's steak video has 4.5 million views, the scrambled eggs video 5.1 million, and the scallops one just under a million, none of it ever had paid behind it.

One of their top organic videos is Gordon cooking a steak
Other creators' recipe videos have also done well

HexClad's highest ROAS ad of the year is Gordon Ramsay's eight and a half minute organic YouTube recipe video, never made as an ad in the first place.

Made to live on organic YouTube as a fun recipe piece
Best one day click, one day view ROAS this year
Proof long form can work very well as an ad

Sources: Marketing Operators, E014: YouTube Deep Dive: Ads, Organic Growth, Partnerships & More, at 58:07, 25:57

Step 2 · The deal and what it costs

With a flat fee the real price per view is only known after the video runs, so the creator's category and audience decide whether it pays.

Typical deal structure sponsored videos 3-6 months, most brands forget they get 30 day ad rights free, they charge 20% of the influencer cost, 20k of the hundred, to the ad creative budget.

They decide this split somewhat arbitrarily
Coding 20k to ad creative makes influencer ROAS look better
Click ROAS is judged the same way as YouTube or Meta ads

Deals are priced as a flat fee back-solved from recent average views at an assumed CPM, call it $10.

Now sitting around $12 to $15 CPM, varies across the board
Average views taken from the last couple of videos

Because the fee is fixed, an under-delivering video raises the effective CPM and a viral one collapses it.

Fewer views than expected pushes the CPM higher
A trending viral video pushes the CPM way down
Same setup tried elsewhere got little performance out of YouTube

HexClad copied Ridge's flat fee YouTube creator model and it did not produce performance.

Setup was a vanity URL, discount code and UTM
Links usually went to a landing page matching the creator
Now a different YouTube strategy built on creators' organic content

HexClad concentrated 85% of its flat fee creator deals in cooking, the most expensive possible category for it.

Non cooking creators still reach engaged viewers who cook at home
Underpriced attention, even without an authority figure
New quarterly goal on top line impressions from non cooking influencers

Connor MacDonald estimates non-cookware creators would be about 90% cheaper.

Plan to test budget on non cookware influencers
His brand scaled YouTube by sponsoring almost any type of content
Skips politics and creators who get too controversial
Did not treat YouTube explicitly as a brand building channel

CPM depends entirely on the content: a Fortnite video skews younger and less engaged, so it's worth much less to Ridge than a science channel.

Creator ads lift answers to the post purchase survey
Creators say tell Ridge I sent you, a built in prompt
Fortnite creator is sometimes not even on camera
Sponsored Smarter Every Day a bunch of times

Ridge pays a much higher CPM for Smarter Every Day because it skews older and more affluent and sometimes outperforms everything.

Smarter Every Day runs deep, almost science experiment videos
CPM depends on the creator and the content
He thinks anything even remotely beauty will be very expensive

Sources: Marketing Operators, The Influencer Marketing Playbook - with Lily Comba, Founder & CEO of Superbloom, at 40:08 · Marketing Operators, E014: YouTube Deep Dive: Ads, Organic Growth, Partnerships & More, at 1:07:54, 1:08:01, 1:08:20, 1:12:05, 1:12:32, 1:05:42, 1:06:08

Step 2 · The first creator test

A first test only teaches something if the spend is concentrated enough to read, whether on one launch or across three platforms.

Connor MacDonald's advice for a first creator test is to concentrate the whole budget on one launch so you can read it.

Pick one campaign or product launch and over index on creators
Concentrate 50k or 100k in one period
Isolate other variables to see if it overperforms
Thinks it maybe pencils out and is totally worth trying

Connor Rolain's version is to deploy something like 100k split across YouTube, Instagram and TikTok creators and let the attribution pick the winner.

YouTube has a level of intent Instagram and TikTok lack
Narrow down once one channel shows the best attribution
Even double the CPM on YouTube could still perform better, who knows

The agency recommended a 50 thousand dollar minimum to test the channel.

Their buyers rarely follow beauty creators, maybe search occasionally
Would try health creators on the clean angle
Fashion or business creators are other options
Asking whether the channel is worth testing at all

Sources: Marketing Operators, E014: YouTube Deep Dive: Ads, Organic Growth, Partnerships & More, at 1:19:26, 1:19:53, 1:18:21

Step 2 · When the integrations go live

Landing every post during a sale and retargeting viewers later aims to turn creator views into orders, which HexClad struggled to do.

Ridge times all its influencer integrations to go live between its Black Friday sale launch on November 12 and Cyber Monday.

Every creator is told to post before Cyber Monday
Some procrastinators lag, posting as late as December 15
Goal is all impressions landing while the sale runs

About 10% of Ridge's partnership deals are sponsored YouTubers whose posts they also promote, closing the loop from podcast to feed.

YouTube viewers then get retargeted with the creator's partnership ads
Largely what they do with MKBHD
Also done with Steven Shapiro

HexClad blames its long consideration period for creator discovery not converting.

Same setup as last year, vanity URL, code, UTM, landing page
Probably a lot of discovery, not a lot of conversions
CPMs rarely under $40 to $50, often $75 to $100

Sources: Marketing Operators, E035: Black Friday Cyber Monday: How We'll Be Tracking, Analyzing & Maximizing Our Results, at 32:48 · Marketing Operators, How We're Reaching New Audiences on Meta with Partnership Ads, at 21:05 · Marketing Operators, E014: YouTube Deep Dive: Ads, Organic Growth, Partnerships & More, at 1:08:41

Step 2 · Discount code leakage

A leaked code hands the discount to coupon sites, which adds cost and blurs how much each creator really sold.

HexClad solved leakage on its Joe Rogan sponsorship using IntelliGems' checkout-level script discount, fired by the UTM.

IntelliGems applies discounts at the checkout script level
The UTM decides which discount gets applied
Rogan saw a ton of attribution and zero leakage
Downside, creator reads carry no code CTA, arguably a bad experience

A Rogan sponsorship that went live last November was still producing purchases eight months later, all attributed through the UTM and that discount code.

Told Rogan his CTA had to push people to the link
Purchases still came in over the last 30 days
Unsure if Rogan is even on YouTube anymore
Recommends checkout script discounts to fight code leakage

HexClad cannot stop discount code leakage regardless of the mechanic it uses: CleanCart shows around 500 attempted injections.

Leaks even when the code hides in a vanity URL
Creators told to say click to auto apply, still leaks
UTM data shows not many people clicked
Script level discounts worked really well on Joe Rogan

Ridge ran 10% off with creators for years, had plenty leak, and just ate the cost rather than fixing it.

Codes that clearly leaked get updated
Turning a code off breaks evergreen YouTube content
Viewers wanting the creator's code find it no longer works

Sources: Marketing Operators, E014: YouTube Deep Dive: Ads, Organic Growth, Partnerships & More, at 1:15:20, 1:16:12, 1:13:53, 1:14:26

Step 2 · The hook on YouTube

YouTube viewers watch long, often on a TV with sound on, so the first five seconds need a spoken story rather than a shock visual made for Meta.

The consumption context on YouTube is different enough that ad creative has to be different, with 40 minute average watch time and 39% of sessions on a TV.

One guest said 30% of their views are on connected TV
Viewers are immersed and consume content way differently
In stream creative that performs is far different
You still need a hook

On YouTube the audio hook matters more than the visual hook, the reverse of Meta, because everybody watching YouTube has sound on.

People skip ads without a good hook
On Meta a visual hook matters, not everyone has sound on
After the hook, no need to cut as quickly
Five seconds to earn their attention

A 40% thumb stop on Facebook buys three seconds, whereas YouTube gives you five seconds of everybody.

Attention is easier to keep once earned
Changes how you approach the hook
Hook becomes a lead into content they might like

Curiosity and story work as a YouTube hook where a shock visual works on Meta.

He's sure the Bobby ad hook is more founder led
Meta hook example, shredding wallets to stop the scroll
YouTube hook can open by starting an interesting story

Ridge sends vertical short-form to Shorts but edits longer content specifically for in-stream, where five seconds are non-skippable.

Shorts is undervalued ad inventory, still relatively undervalued
Shorts was a huge piece of the mix in Q4 2022
More editing time goes into slightly longer in stream cuts

YouTube creative lasts months, which he contrasts directly with TikTok.

YouTube is pretty set it and forget it
Creatives fatigue eventually, but no need for 50 a week
He manages it himself, doesn't take much time
Could see a Google specialist in the future

Sources: Marketing Operators, E014: YouTube Deep Dive: Ads, Organic Growth, Partnerships & More, at 14:21, 17:31

Step 2 · YouTube ads that worked

Long story ads that fail in the Instagram feed keep winning on YouTube, so one creative rarely serves both platforms.

HexClad's best YouTube ads teach a recipe and only subtly position the pan as the tool: two how-to-cook-steak videos are two of the top three performing ads.

Value beyond the product beats a straight pitch
Facebook hook shows old pans going into an industrial crusher
Crusher ad gets a 35 to 40% thumb stop
YouTube steak ads skip that visual hook

The second best HexClad ad opens with Chef Daphne, winner of that season of Hell's Kitchen, simply introducing herself rather than a visual gimmick.

Opens with I'm Chef Daphne, then how to cook a steak
Less direct response, more narrative storytelling
Gives value beyond the product being sold

HexClad's winning YouTube ads look nothing like its Facebook ads, they are studio-shot and high production value.

Now a good setup for producing that kind of content
Hero video has Gordon explaining what hybrid means
Very different from what goes in their Facebook accounts

Jones Road found the opposite on Meta, where the long form founder content works best and choppy short form took years to crack.

Narrative storytelling has worked best for them on Meta
Maybe because Bobby is really good on video
Struggled hard to get shorter, choppier ads working

Jones Road's YouTube winners are long studio pieces with founder Bobby on Meta, 60, 90 seconds or two minutes, that break direct response best practice.

Now a more diversified creative mix on Meta
Not traditional direct response
Search runs through an agency for now
Could see one Google person on search or YouTube plus another channel

Ridge's best performing YouTube ad is founder Daniel telling the Kickstarter origin story from 11 years ago.

Shorts and in stream can work together for discovery
They can be bought separately with different creative
YouTube content differs from what people flip through on Reels or TikTok

That ad, which also launched them on Kickstarter, was built as a sequence of 40 to 60 second vignettes, one of which was a Tim Ferriss read.

Spent about 80% of their whole budget on YouTube
Scaled to eight figures on YouTube alone
He thinks it was around 2018
Segments said who the brand is and what it does

Ridge built its Marques Brownlee ad as a story about Marques, opening on his own old clip saying he had 74 subscribers, before cutting to him becoming Ridge's chief creative partner.

Not their best performing content so far
Old clips start when Marques was about 13
Ridge calls the result a very full story

That long form storytelling does not work in the Instagram feed, only on YouTube.

Story qualifies Marques, then leads into him joining Ridge
Then covers his role and what Ridge is as a brand
Hedge: at least they have not found it working in the feed

Sources: Marketing Operators, E014: YouTube Deep Dive: Ads, Organic Growth, Partnerships & More, at 15:37, 18:06, 20:12, 25:26, 22:53, 23:14

Step 2 · The brand built into the video

When the product is part of the video itself, the sponsorship beats a plain ad read, as Ridge saw with Smarter Every Day.

The reference example is a ColorPop creator collaboration that built the world's biggest lipstick and got millions of views.

Old approach was spray and pray, cheap CPMs everywhere
That worked for a while, then the focus changed
Now the goal is ads people want to watch
Ashton gave the ColorPop example; he thinks the whole video was about it

A Smarter Every Day video Ridge sponsored performed three times better than expected because the brand was built into the experiment.

The ad read felt native to the content
It ended up around 8 minutes, he thinks
Ridge was an integral part of the video

Ridge scaled the idea by shipping its sweepstakes truck around the country so six creators could film with it.

Goal was to repeat that result more scalably
The truck was last year's sweepstakes giveaway

Every truck video outperformed expectations by multiples, and they had a clean comparison against creators without the truck.

Truck went into videos from six creators last year
Takeaway: deeper integration performs better
Now they seek rights to that content to run as ads
Paid placements named are programmatic YouTube and paid social

Ridge holds monthly calls with the MKBHD team on how to tell the Ridge story.

Calls go in depth on telling the Ridge story better
Also use new products or categories to build content programs

HexClad's allow-list test on a chef with little clout underperformed the Gordon, Chef Daphne and Chef Marcella ads run from the brand page.

Best content on their YouTube page is celebrity chef videos
Chefs with medium to high clout work best
Excited to whitelist ads from creators' own pages
He thinks allow list ads from YouTube will be huge

Sources: Marketing Operators, E014: YouTube Deep Dive: Ads, Organic Growth, Partnerships & More, at 1:21:11, 1:22:20, 1:22:34, 1:22:37, 1:24:47, 37:05

Step 2 · YouTube direct response

Direct response on YouTube still works when the first 3 seconds are strange, funny or shocking enough to stop a viewer who knows it is an ad.

Cody names V Shred as the best YouTube direct response he has seen, running a quiz funnel with outlandish hooks.

V Shred is a direct response supplement company
Hooks are outlandish but grab curiosity and attention
Another host watched a fitness ad for 4 and 1/2 minutes
He guessed it was probably V Shred

Javy is mentioned as doing a good job with creative, specifically with problem/solution ads that are entertaining.

They agree problem solution creative is the best
One Javy hook says don't buy Starbucks, on a dark screen
You can tell it is an ad, but people still click
Their ads pull viewers on to the next piece

The brand's creative strategy involves making the first 3 seconds of a video either 'really weird or really funny' to be entertaining and capture attention.

Viewers know it's an ad but find it entertaining
The same girl saying she is obsessed gets scrolled past fast
Goal is to interrupt the pattern of someone scrolling in bed

The brand uses a creative approach where the first 3 seconds of an ad show someone funneling capsules, which has performed well and increased engagement.

One of their best performing ads, engagement through the roof
High engagement helps their costs
Drew backlash; it can seem borderline offensive
Media buyer has to work out where the line is

Brent Umzu is a 'killer on YouTube' who uses shocking hooks for his ads.

Sells a gut cleanser using taboo wording
Hook says you carry 18 lbs of toxic waste
That first 3 second hook makes you ask what this is

The speaker mentions that 'this ugc needs to it needs to end,' implying a critique of certain user-generated content strategies.

He wants the generic UGC style to end
Shock stops people even though they know it's an ad
Example is Brent Umzu's 18 lbs of toxic waste hook

Step 2 · More points

Podcast buys now happen largely on YouTube, and Shorts limit outside links, so where a read appears changes how a brand tracks it.

Half of podcast views are on YouTube. Half of the views of podcasts come from YouTube at this point, says one operator on Marketing Operators. A podcast buy is very visual now.
Shorts take affiliate links only. A YouTube update removed hyperlinks in Shorts: no more links to external sites outside affiliate links. Shorts have their own linking feature instead.

Sources: Marketing Operators, September 2025, at 15:09 · Jon Youshaei, December 2023, at 10:43

Up next: behcnmarks u can compare yourself too.

Step 3: behcnmarks u can compare yourself too

Benchmarks let a brand check its own YouTube spend, prices and return targets against brands that already run YouTube sponsorships at scale.

Knowing what share of budget comparable brands give YouTube helps a brand judge whether it spends too little to read results or too much to defend.

Over the last 60 days, Ridge spent 5% of media budget on YouTube, and describes the channel as hit or miss.

Ridge crushed programmatic YouTube many years ago, when it was easier
Struggled after that, but saw success this year
He calls the 5% a little conservative

During Father's Day the YouTube share went to 6 to 8%, which is hundreds of thousands of dollars.

Base share over the last 60 days is 5%
Hit or miss overall, more success this year
Host calls it not insignificant at their spend

HexClad has kept YouTube at 1 to 1.5% of budget because it could not validate the channel.

Their YouTube has been mid roll and pre roll ads
Ridge spends more on YouTube as a percentage
He expects it to go way up very soon

Jones Road runs about 15% of media on YouTube, the highest of the three, ranging 12 to 20 across the year.

Calls YouTube pretty steady for them
Hard to measure, but he thinks it performs well

Connor MacDonald says he does not talk to many brands with YouTube anywhere near 25% of spend.

Triangulated data can support four to six, maybe 10% of spend
Without more clarity, 25% is hard to justify
Even testing into 25% is hard
MER can keep penciling out meanwhile

William Painter and, early on, Dr. Squatch ran 70% of budget on YouTube, where a failure would be obvious, and the hard part is the middle ground.

Without more clarity, 25% of spend is hard to justify
Scaling to a meaningful share is very tricky
He rarely talks to brands anywhere near 25%

Influencer is roughly 10% of the total marketing budget, and podcast sits inside that line.

Worked with about 5,000 YouTube creators in total
Typically paid about $10 CPMs
Since 2021, crypto names like FTS offered $3,000 CPMs
Now mostly normalized, so spending a lot more there

The rule of thumb he uses, on 5 million a month in digital ad spend, is 100k to 200k plus on influencer, about one percent.

Discussed earlier with Conor McDonald
He says they should probably spend it, not that they do

Ridge thinks it could spend 50 thousand dollars a day on YouTube fairly easily.

YouTube efforts feel easier to scale to him
Reddit could never take a fraction of that spend, he thinks
Platform has weird nuances, many of them benefits
Many advertisers have not cracked YouTube best practice yet

Sources: Marketing Operators, E014: YouTube Deep Dive: Ads, Organic Growth, Partnerships & More, at 13:07, 13:19, 14:56, 17:01, 48:59, 38:57 · Open Residency, $200M CEO: How To Build and Scale Your Online Business in 2026 (Masterclass), at 3:37 · Marketing Operators, The Influencer Marketing Playbook - with Lily Comba, Founder & CEO of Superbloom, at 40:53

Step 3 · What a YouTube view costs

Knowing the going rate per thousand views by niche gives a brand a starting number and room to push back in a negotiation.

The original economics were a five to ten dollar CPM for a 60 to 90 second native integration in front of an opted-in audience.

Viewers opted in, so there is some commitment
Performed efficiently and built the brand too
Anthony Fantano's fans joked about a wallet salesman reviewing music

Ridge averaged a seven dollar CPM in the early years because almost nobody was sponsoring creators.

Early sponsors included Manscaped and Dollar Shave Club
Rising CPMs shaped their second phase of partnerships
YouTube also skews male

Ridge now pays around 12 to 15 dollars CPM and has stopped scaling creator deals the way it used to.

He jokes CPMs track talk of the creator economy
Creators made it a profession and asked for more
They price off recent average views, offering about a $10 CPM

HexClad could not get YouTube creator CPMs under 40 to 50 dollars and often paid 75 to 100.

Lots of discovery, few conversions, he thinks
He ties it to their consideration period
Last year's worry was too few organic review videos
YouTube is top and bottom of funnel

Connor MacDonald says anything remotely beauty related is very expensive.

CPM depends on the creator and the content
Smarter Every Day skews older and more affluent, higher CPM
Those sometimes perform far better than anything else

The comparison CPMs named in the room are 28 to 30 on Meta against 15 on YouTube.

YouTube carries a level of intent Instagram and TikTok lack
Advice: put 100k across the core social channels, compare attribution
Then narrow down to whichever channel attributes best
Paying double the CPM on YouTube may still perform better, who knows

Ask a YouTuber for average watch time, click-through rate and subscriber versus non-subscriber views.

YouTube gives far more data than Instagram to predict performance
Average views are public, so compare them to subscribers
Lets you predict a CPA per YouTuber before you sign
The negotiation is where the fun part of YouTube is

Sources: Marketing Operators, E014: YouTube Deep Dive: Ads, Organic Growth, Partnerships & More, at 1:01:41, 1:06:37, 1:07:36, 1:09:04, 1:06:14, 1:20:02 · Marketing Operators, The Influencer Marketing Playbook - with Lily Comba, Founder & CEO of Superbloom, at 56:13

Step 3 · What YouTube counts as a view

If the platform undercounts what YouTube did, a brand that trusts the default numbers will undervalue the channel unless it builds its own signal.

Google counts a view at around 15 seconds, which means influence before that point is never attributed.

A click gets attribution, but needs a separate Google account
Google also credits post view conversions with no click
That shows a bit more than Northbeam would
He believes the view cutoff is 15 seconds

Ridge built a pseudo conversion metric for YouTube that discounts true view-through, cutting people who didn't watch enough seconds to count as a view.

Influence before the 15 second mark never gets attributed
The metric blends clicks and post view conversions
View through conversions are available inside Google
He calls the whole setup a total mess

Ridge shares YouTube viewer audiences back into search so it can see which branded queries came from YouTube.

Brand queries are a big piece of the journey from YouTube
They filter the branded campaign by the YouTube viewer audience
They look at impressions on branded search by query
He calls it geeky stuff he thinks is cool

That gives them a cost per brand query from YouTube.

Build audiences of people who viewed your YouTube ads
Share them to search and shopping, filter the branded campaign
See which branded queries came from YouTube
Used it a long time to judge whether YouTube worked

Sources: Marketing Operators, E014: YouTube Deep Dive: Ads, Organic Growth, Partnerships & More, at 42:08

Step 3 · The target ROAS on YouTube

A lower target on YouTube is only safe when a brand can show halo, through rising survey answers or a later conversion window.

The one thing every brand profitable on YouTube shares is a way to measure halo.

Successful YouTube brands otherwise do things very differently
He knows no one profitable on YouTube who skips validating halo
Geo incrementality is the part he calls a perfect science
Many growth leads can't prove it to their CFO, ROAS looks worse than Meta

The cheap validation before an incrementality tool: as YouTube spend scales, the share of post-purchase survey respondents who say they found you on YouTube should grow too.

Scale YouTube up while MER still looks good
Watch the trend line in the post purchase survey
No growth in YouTube answers means probably no incrementality
He calls this check number one

Percent of new visitors in Northbeam justifies accepting a lower one-day-click ROAS on an upper funnel channel.

Higher new visit share theoretically means upper funnel
39% of people watch YouTube on a TV and won't click
As a view platform, YouTube can theoretically take a lower target too
Together these set the target 40% below Meta

Jones Road set its YouTube target 40% below its Meta target on that reasoning and scaled from there.

Kept checking MER and the post purchase survey while scaling
Reached 10K a day, a few hundred thousand a month
At that point they wanted to scale it further

Because of the long tail and the brand value, Ridge gives YouTube creator partnerships a materially lower target ROAS than paid social.

Creator impressions are more memorable, which is worth something to Ridge
Give a deal time before fully calculating and concluding
Early reads are expected to back out over time

Northbeam's one-day, seven-day, 30-day and 90-day ROAS lag windows show YouTube converting significantly later than Meta and other channels.

You need value outside clicks, YouTube builds a lot of consideration
Find the measure that works for you
He admits it can get really complicated

On clicks and views, both HexClad and Jones Road say YouTube beats Meta.

One host always found Northbeam undervalued views
Adding one day view shows four or five times the attribution
On one day click and view, YouTube beats Meta for both
He wonders if Northbeam changed its model

Adding a one day view window multiplies attributed revenue four or five times.

They plan to judge MTA more on clicks and views
One host always found Northbeam undervalued views
On one day click and view, YouTube beats Meta for both

Sources: Marketing Operators, E014: YouTube Deep Dive: Ads, Organic Growth, Partnerships & More, at 1:03:56, 41:22, 41:12

Step 3 · Halo and the YouTube holdout

A holdout settles which attribution source to believe, and HexClad's showed much of its YouTube value landing on Amazon where clicks never see it.

Precedent AI splits revenue into direct and halo, separately for Shopify and Amazon.

Works on a channel and campaign basis
He did not see it broken out by YouTube yet
Lets them make an educated guess on channel percentages
For instance, 30 cents on Amazon for every YouTube dollar

HexClad ran its YouTube holdout because its four attribution sources disagreed with each other.

Clicks only MTA made YouTube look bad
Clicks and views looked a lot better
Post purchase survey looked really good
Organic videos taking off could skew those reads

Precedent's model splits HexClad's YouTube prospecting revenue as 0.41% direct, 34% Shopify halo, and 65% into Amazon.

Halo means they converted without taking the intended action
He is unsure how Precedent defines base revenue, will confirm
Shows traditional attribution methods won't pick this up

For every incremental Shopify order YouTube produced 2.8 incremental Amazon orders.

HexClad's Facebook holdout spillover to Amazon was roughly one to one
The holdout's Amazon spillover almost matches Precedent's model
Midpoint read only, the test isn't over yet
He assumes final results won't be too different

At the midpoint the incremental cost per acquisition on YouTube was orders of magnitude better than Facebook, though at much lower spend.

Two cell test, 50% holdout, measuring lift in orders
30 day test with a 14 day lag
Main metric is cost per incremental acquisition
Better CPIA was expected at much lower spend

Cody's counter-proposal is to buy reach at a two dollar CPM rather than conversions at twenty, since Amazon buyers never send a purchase signal anyway.

Amazon purchases give no purchase signal to optimize on
HexClad agreed it could hold out a reach campaign too
Devil's advocate says reach may find different people than conversions
Neither gets optimization, but conversions campaigns at least get targeting

HexClad's YouTube holdout is hard to read partly because affiliates, organic and paid deals all run at once.

Clicks only MTA looked bad, clicks and views better
Post purchase survey looked really good
Organic videos taking off can skew those reads
So they needed a holdout test

That small share is still 250 to 300 thousand dollars a year, which they consider enough to read the channel.

Only 1.5%, but not nothing
They have a good sense of the creative that works there
He expects the share to go way up very soon
Their YouTube creative is very different from their Facebook ads

HexClad runs a view-optimized campaign at 10% of its YouTube budget and it is the best ROAS campaign it has.

Allow listing gets its own campaign, a newer tactic
Scaled prospecting and remarketing campaigns hold top performing creatives
Prospecting is broad, all ages, target CPA, optimized for conversions

Sources: Marketing Operators, E014: YouTube Deep Dive: Ads, Organic Growth, Partnerships & More, at 1:25:53, 1:26:34, 1:27:39, 1:27:07, 1:29:01, 1:26:06, 15:21, 32:58

Step 3 · Attribution on creator deals

Creator results swing so much by video that one multiplier misleads, so brands combine codes, surveys and revenue spikes to judge each deal.

The Smarter Every Day video last June brought almost half a million dollars, and topline revenue visibly jumped about 100 grand the day it went live.

Influencer is their hardest channel to attribute, unless a video hits big
They triangulate with a post purchase survey
Host guesses most value comes from seeing it and searching later

They deliberately have no attribution multiplier for influencer because results differ too much by creator

He doesn't feel confident in any multiplier
Philip DeFranco posts daily, views stop after 24 hours
History channels keep getting views forever

Influencer attribution is by far their hardest, partly because YouTube changed the UI and clicking through got harder

They triangulate with codes and a post purchase survey
Smarter Every Day, early June, almost half a million in revenue
Topline revenue jumped 100 grand the day it went live
Value likely comes from seeing it and searching later

They do see directly attributable revenue from influencer, just nowhere near what a conversion oriented Meta campaign returns.

A long consideration period means less direct ROI
Main goal is arbitraging eyeballs with super cheap CPMs
They still watch performance and have attribution methods
That may differ for a lot of brands, he says

Sources: Marketing Operators, E008: Your Questions Answered! Marketing Channel Expansion, Team Building & Influencer Marketing!, at 1:03:12, 1:03:44, Marketing Operators, E008: Your Questions Answered! Marketing Channel Expansion, Team Building & Influencer Marketing!, 53:35

Step 3 · More points

A creator's own ad income per thousand views is a benchmark a brand can hold next to the sponsorship rate it is quoted.

Long form AdSense runs $6 to $10. One creator on Jay Clouse's show gets between $6 and $10 RPM or CPM in long form AdSense revenue. Even after a 5 to 10% cut of the affiliates, the affiliate CPMs come out very close to that.

Sources: Jay Clouse, August 2025, at 15:54

Up next: TikTok and LinkedIn next to YouTube.

Step 4: TikTok and LinkedIn next to YouTube

Seeing TikTok, LinkedIn, Meta and podcasts beside YouTube shows where sponsorships fit in the marketing budget and when another platform is the better buy.

Investors are treating TikTok the same way they treated Instagram initially, unsure if it makes sense.

Investors weren't sure Instagram made sense at first
There are more precedents now than there were with Instagram
Facebook's rise came without a rise of influencers
He saw the creator community rise most with Vine

Kwibi's bet is that everything is moving toward mobile and streaming.

Ten minute episodes filmed from several perspectives
Flip the phone horizontal to see another character's view
Huge networks and TV content creators already onboard
They are raising money

The number of creators, especially on platforms like TikTok, is astounding.

Many people are famous on TikTok alone
The tail of creators keeps getting longer, democratizing fame
He guesses kids' top career answer now is TikTok star

Experimenting with TikTok is recommended because it's a fertile place to play.

Brand ambassadors give a steady source of content and awareness
They slowly convert their followers to consider your product
A friend's company, dScout, went viral and downloads crashed their server
For B2B, make content and target it with Facebook's ad tools

LinkedIn can drive a significant amount of traffic, surprisingly.

Hard to find a LinkedIn influencer to promote B2B software
He cites Gary Vee as a LinkedIn influencer
A soap brand might test TikToks on washing hands 20 seconds, not 5
Silly angles work, like happy birthday hand washing songs

Step 4 · YouTube next to Meta

Meta sets the bar most finance teams judge YouTube against, so a brand needs a fair comparison on price per view and a slower return.

Interesting note by industry

By platform

Because of the long tail and the brand value, Ridge gives YouTube creator partnerships a materially lower target ROAS than paid social.

Give it time before calculating results and drawing conclusions
1 week in, they expect results to back out over time
YouTube partnerships get a much lower target ROAS
Why: YouTube impressions are more memorable, which is worth something

For the audit ho wmuch is youtube spend per day

Cody does not think YouTube works for every brand, and specifically doubts it at low AOV.

He does not think every brand will do well on YouTube
He thinks it does not do super well at low AOV
Higher in the funnel, so probably takes a higher AOV
He finds YouTube creative harder than Meta creative

On clicks and views, both HexClad and Jones Road say YouTube beats Meta.

He never looks at Northbeam views, finds they undervalue views
Adding one day view shows 4 or 5x more attribution
On one day click and view, theirs beats Meta

Convince cfo u need more yt than meta

Compare channels on price per thousand views. Ridge's CEO calls it ad arbitrage. At this point he gives Facebook $15 per thousand views, so the test is whether newsletters, influencers or YouTube sell attention at a better price.
Lower CPMs & longer views on YouTube. One brand on Marketing Operators is going to spend more on YouTube: lower CPMs, longer video views & more brand awareness, in the operator's view. It is doing the same thing with podcasting.
Expect a lower return than Meta. One operator on Marketing Operators reads YouTube on a one day click in Northbeam & starts from a lower return than Meta, because YouTube is new audiences. Meta is way better than that.
YouTube creators cannot be geo targeted. For one brand going international, Facebook & Google are the lion's share of the budget. Its YouTube creators are not geo targeting their viewers, so that partnership spend affects different markets in different ways.
Do not diversify off Facebook too fast. Ridge's CEO says Ridge tried to diversify too fast. Newsletter sponsorships worked, but he would have put all of his dollars into Facebook up until 2020.
Ask for ad cuts instead of an integration. On Jay Clouse's show: when a brand's goal is ads, a creator can swap one integrated YouTube video for 5 videos of 30 seconds, each with a different hook & call to action, to run on Facebook & Instagram.

Sources: My First Million, March 2025, at 1:59, 5:36 · Marketing Operators, November 2025, at 24:32 · Marketing Operators, February 2025, at 9:18 · Marketing Operators, June 2024, at 57:03 · Jay Clouse, January 2024, at 33:48

Step 4 · What each platform is for

Matching the product and audience to each platform's strength keeps YouTube for the jobs it does best, like deep niche trust or worldwide reach.

Pick the platform by the product. An operator on Brydge Club says a dog food business may need to be visual, which means partnering with influencers on YouTube. A vitamin brand that is functional may go on TikTok Shop & build an army of advocates.
Buy long form for a small niche. In Colin and Samir's conversation with the CEO of Instagram: to reach a small niche audience & have a deep, stable connection, long form is probably better, whether it's podcasts or maybe YouTube videos.
Celebrities for TV, creators for social. On BevNET, 50 Cent was called far more effective in traditional media, an out of home ad or a TV ad, which is how Vitamin Water used him. On social, YouTubers & Instagrammers born on the platform are more effective.
Add YouTube for B2B products. Lovable's Head of Growth: influencer marketing used to be a B2C practice, meaning TikTok & Instagram. Now it is TikTok, Instagram & YouTube, because YouTube is so huge right now for B2B products too.
Sell worldwide products on YouTube. On My First Million, the reason VPNs advertise all over YouTube: it does not matter if the ad is served to someone in Japan, India or Europe, because the VPN is available there.
Buy every channel one creator owns. On Open Residency: the only way to win trust with a creator's audience is repeated exposure across different mediums. The brand is on their social, their podcast & their newsletter, & not just once but many times.

Sources: Brydge Club, July 2026, at 50:21 · Colin and Samir, June 2024, at 49:49 · BevNET.com, September 2021, at 9:13 · Product School, November 2025, at 29:43 · My First Million, September 2023, at 17:20 · Open Residency, August 2026, at 1:20:39

Step 4 · What each platform costs

Comparing platforms on price alone misses how much the audience and the work behind each post change what a brand actually gets.

A narrow audience charges more. A speaker on Open Residency says his podcast, YouTube channel & X account are not massive, but the CPM he can charge for that subset of people is much higher than the CPM on YouTube with a million views.
Same fee buys different work. On Colin and Samir: for the same $20,000 or $30,000 brand deal, an Instagram might be a story or a photo, a TikTok is filmed on a phone, & a YouTube integration takes a lot of revisions.
New creators can cost nothing. One bootstrapped brand did not have to pay its creators, because they were just starting out & wanted to build a portfolio. It called the price of half a million views on YouTube or Instagram enormous.
Price affiliates against your paid channels. On Marketing Operators: look at the CPA other top of funnel channels get in Northbeam, then set an affiliate kickback that motivates the affiliate & is marginally better than what is paid to Facebook, TikTok, YouTube or a creator.
Start organic without a testing budget. Obvi's CMO Ash Melwani: back in 2019 a brand could profit on Facebook, so Obvi went straight for ads. Now, without 10K a month for testing, it is organic: TikTok, seeding to influencers, Instagram Reels, Facebook Reels, YouTube Shorts.

Sources: Open Residency, March 2026, at 2:33:33 · Colin and Samir, January 2024, at 1:01:42 · Ecommerce Influence, April 2022, at 17:48 · Marketing Operators, February 2025, at 1:05:09 · New Money Talks, June 2023, at 24:48

Step 4 · YouTube next to TikTok

A budget split across many low priced TikTokers behaves differently from one or two YouTube deals, so the mix decides how many creators one budget reaches.

Low TikTok rates, high YouTube rates. On Colin and Samir: when a brand has money to distribute, it is easier to give it to 10 TikTokers at low rates than to one or two YouTubers, because the rate for a YouTube video is so high.
Partnerships spread past YouTube. One brand on Marketing Operators had a very particular approach to YouTube creators for a long time. Partnerships today are more distributed, closer to spray and pray, with TikTok Shop in the mix & affiliate looking way different.

Sources: Colin and Samir, January 2024, at 1:02:30 · Marketing Operators, May 2024, at 28:14

Step 4 · YouTube next to Instagram

A YouTube read holds a viewer far longer than a reel or story, so counting impressions alone undervalues it next to Instagram.

2 minutes on YouTube. Lily Comba of Superbloom: a YouTube integration books the influencer speaking for about 2 minutes about the brand, a lot longer than a reel or an Instagram story gives.
Count the quality of the impression. One brand on Marketing Operators focused on YouTube creators for the last couple years. It has no big impression number, but a 30, 60 or 90 second integration on a million view video is quite different from an Instagram impression.
Check a YouTuber's Instagram engagement. When one brand on Marketing Operators sponsors YouTube creators, it looks at their engagement on Instagram. It is not all that scientific, but creators with cross-platform appeal seem to have YouTube ads that do better.

Sources: Marketing Operators, December 2025, at 58:55 · Marketing Operators, July 2024, at 55:44 · Marketing Operators, April 2025, at 35:28

Step 4 · Instagram next to TikTok

Small creators break through more easily on TikTok than on Instagram, which matters when picking where a brand builds its base of affiliates and ambassadors.

TikTok Shop creators sell all day. One operator on Marketing Operators sorts creators into collections & calls the difference stark between what a 50,000 person Instagram food or recipe page posts & a TikTok Shop creator who is selling products all day long.
TikTok reaches the niche. One bootstrapped brand saw content with a large amount of exposure for its type of niche on TikTok. On Instagram it was just going after beauty influencers.
TikTok is best at small creators. In Colin and Samir's interview with the CEO of Instagram, TikTok is called best in class at discovering small creators & helping them pop.
Harder to break in on Instagram. The CEO of Instagram told Colin and Samir it is harder to break in on Instagram & on YouTube. Instagram is not as good at valuing smaller creators, partly for a technical reason: less data.
Keep a base of ambassadors. HealthyCell's CEO: a brand needs a base of ambassadors & strong followers, because nobody can guarantee a viral post that hooks into the interest graph algorithm on TikTok, & soon on Instagram as the others adopt it.
Affiliate posts answer the search. On Limited Supply: a creator program puts out a predictable number of posts once it gets rolling. When people find out about the brand & search Reddit, TikTok or Instagram, they find all of these posts by affiliates.

Sources: Marketing Operators, February 2026, at 54:57 · Ecommerce Influence, April 2022, at 19:54 · Colin and Samir, June 2024, at 45:09, 1:00:30 · Young With Solutions, May 2025, at 21:23 · Limited Supply, September 2026, at 7:15

Step 4 · Podcasts next to YouTube & TikTok

Long podcast and YouTube reads give a brand minutes of attention instead of seconds, and that time is where trust with an audience gets built.

A long ad read is a different impression. On Marketing Operators: people are listening to 20 minute YouTube videos or a 90 minute podcast, & the brand gets 60 or 120 second ad reads. It is a different sort of impression.
An hour a week builds trust. On Second Nature, a podcast host says there is no better way to build trust with an audience than an hour or two together every week, next to 30 seconds, or maybe 10 minutes for a longer YouTube video.
Rotate podcasts & newsletters. One brand on Marketing Operators runs newsletters & podcasts the same way: hold on to the ones that perform well consistently, & otherwise always new podcasts & new newsletters, because it wants new audiences.
Let podcasters say it their way. Neuro Gum lets a lot of its podcasters talk about the product in whatever way they want, for better or worse.
Run YouTube & podcast deals in house. Magic Mind stood up a team in Eastern Europe to do direct deals with influencers, mostly on YouTube & podcast. It had 30 employees in Romania the year before.

Sources: Marketing Operators, February 2026, at 56:46 · Second Nature, March 2026, at 32:40 · Marketing Operators, December 2025, at 1:11:10 · TNE with Louisa Nicola, May 2026, at 55:46 · Ecomm Cowboy, June 2026, at 16:23

Step 4 · How each platform is measured

Each platform shows results differently, so a brand needs tracked links, whitelisting or surveys on every channel to compare them fairly.

Ads show the return, Instagram does not. Adi Arezzini: Facebook ads or Amazon ads are more numerical & transparent, a brand sees the money invested & the money that came out. There is no dashboard in the back of Instagram for the return on an influencer.
Whitelist to get the ad data. On Net influencer: an organic post only gives the screenshots the influencer sends. Whitelisting through Meta ads or TikTok ads gives video completion times, CPM & CPA, which decide where in the funnel each influencer goes.
Use the links each platform allows. On eCommerce Fastlane: TikTok can enable clickable posts, & trackable links go in bios & in Instagram stories. The key is attribution links in every promotion, pushed by the influencer across multiple channels.
Ad read buyers over-answer the survey. Ridge's ad reads say "tell Ridge we sent you", so the call to action is the post-purchase survey. One operator thinks those buyers fill it out at a much higher rate than some random person coming from Snapchat.
One code per platform. A creator on SnapStream says one sponsor uses two completely different departments for Instagram ads & podcast ads, an outside agency for one & internal for the other. They do not talk, & each gave a different discount code.

Sources: Adi Teami blends, November 2021, at 15:30 · Net influencer, August 2026, at 9:27 · eCommerce Fastlane, May 2026, at 39:52 · Marketing Operators, October 2025, at 55:46 · SnapStream, August 2023, at 46:54

Step 4 · Podcast influencer marketing playbook

Host reads built on personal use, codes and vanity links show how a sponsorship comes across as a recommendation rather than an ad.

Source: Dr. Chad Walding: Native Path Supplements, Training, Optimizing Health · Co-Movement Gym · Dr. Chad Walding · 19 Mar 2024

The interview itself is a sponsorship activation: NativePath sponsors this gym's podcast and has done since the first episode.

Why: clients feel amazing using NativePath supplements
The host frames the sponsor through his own clients' experience

The host tells a story from Wednesday night about a client who said he was peeing less, absorbing the water and felt fuller with more energy, proof a brand cannot write for itself.

He calls it the same story over and over
The client's words are quoted on air, not a brand claim

Source: Peter Gillooly: TWC Invents The First Ever Cancer Prevention Supplement to Boost Immune System · John Fredericks Media · Peter Gillooly · 16 Feb 2026

This is a host-read endorsement running on a personalised vanity URL and a named promo code, which is the whole media model on display.

URL twwc.health/godzilla with promo code Godzilla
10% off, pitched as get it now before it's too late
The read leads straight into CEO Peter Galuli of the wellness company

The host is a Mars subscriber on auto-refill and shows two jars on camera during the read, which makes it land as personal use rather than an ad.

Two jars of Mars came in the mail today
He was running out but gets an automatic refill

A second product is sold through a bus tour story: he popped it out and got Zpac and antibiotics, so the read carries a use case rather than a claim.

Product is a medical emergency wellness kit
Also lists ivermectin and hydroxychloroquine, plus a guide book
Calls it the greatest invention ever, every family should have one

The host converts live on air, which is the strongest form of the endorsement and is explicitly staged as a decision made during the interview.

He will order it today when he gets off the air
Tells the guest you kind of sold me on it

Source: Fixing America’s Health Crisis with Dr. Chad Walding of NativePath · P.A.S. Report | American Politics · Dr. Chad Walding · 15 Oct 2025

This whole interview is a sponsorship: the host introduces Dr Walding, co-founder of NativePath, declares it an advertiser, then frames the episode around fixing America's healthcare.

Guest is Dr. Chad Walding, NativePath cofounder
NativePath disclosed on air as a show advertiser
Disclosure first, then more importantly the health topic

The format is the interview as the ad: two identical scripted ad breaks read by the host tout NativePath's type one and three collagen, 90% of the body's total.

Only type 1 and 3 collagen fibers
Called the two most important collagen types
Those make up 90% of collagen in cartilage, bone, skin, hair, nails, gut

NativePath's price positioning in the read is bundle deals at a fraction of the price of other brands, so the discount is built into the pack size rather than into a creator code.

Doesn't clump, absorbs fast thanks to its peptide formula
Over 4 million jars sold, 1 million happy customers
8,000 five star reviews and a 365 day no risk guarantee

Source: Supplements Demystified: What Works, What’s Hype, and More with Cymbiotika Founder, Chervin Jafarieh · Molly Sims · 06 May 2025

The episode itself is a paid style placement: the host closes with Symbiotica.com/lipstick, 20% off and free shipping. 41:56

He credits Sharine for the discount and calls it family
The URL is said again, spelled letter by letter

The host also spells out his personal handle, CHERVIN333, so the founder's own account is part of the placement. 42:17

Host spells the brand name letter by letter again
Tells listeners to follow Sherine
Praises her for not overcomplicating it, which can overwhelm people

She names specific products she personally takes, which is the strongest part of the endorsement. 6:14

Says people are telling other people about it
Vitamin C liposome is the only one she can swallow
Likes its slight vanilla taste and how easy it is

She admits on air that she took a sachet that morning and another one just before recording, which is the demonstration. 6:28

She is not great with texture but loves the packets
Already took one, jokes she could take another

She frames the discount as costly, from the position of someone who also owns a brand, which raises its perceived value. 45:04

Repeats Cymbiotika.com lipstick for 20% off
Thanks the guest on behalf of listeners
As company owners, they know these discounts are big

He uses the interview to break unreleased product news, which gives the host an exclusive and gives the launch a hook. 22:15

Glycine is an amino acid, sedative and calming
New magnesium coming with 3 forms in it
He says I'll just leak it out here

Step 4 · Playbook to livestreams

Live streams cost enough to run that many brands cut them, so the ones that keep going lean on guests, giveaways and TikTok retargeting.

Going live is important because it's the most authentic way to interact with consumers, allowing for real-time conversation and engagement.

Lives are expensive to run, so many brands cut them
They doubled down and built a live studio in Las Vegas
They are trying to run longer lives
On TikTok, live viewers join the brand's retargeting algorithm

The key to a good live is being interactive and offering something special beyond just the product, such as giveaways or guest appearances.

People can buy products on their own, so add giveaways
They bring guests like Steve Aoki and Mitchell Hooper
They run founder lives all the time
Hooper, world's strongest man, fills the whole screen visually

The applause for shorts creators was bigger than for A-list celebrities, and even bigger for long-form creators and live content creators.

He heard it secondhand and hedges on the source
At a game, celebrities came out, then shorts, long form, live creators
Long form here means podcasters

Up next: what you are running this for.

Part 6 · Testing YouTube first, then winning more budget

With the steps taught, aim them at one of two goals, a first test that shows whether YouTube sponsorships earn back, or a measured case for more budget.

6a · Testing YouTube for the first time

A first test teaches nothing if it is too small or spread too thin to show whether YouTube works for the brand.

Connor MacDonald's advice for a first creator test is to concentrate the whole budget on one launch so you can read it.
The agency recommended a 50 thousand dollar minimum to test the channel.
The cheap validation before an incrementality tool: as YouTube spend scales, the share of post-purchase survey respondents who say they found you on YouTube should grow too.

6b · Getting more budget for YouTube

A bigger YouTube budget rests on proof of halo, so new visitor share and branded searches from viewers give finance something to check.

The one thing every brand profitable on YouTube shares is a way to measure halo.
Percent of new visitors in Northbeam justifies accepting a lower one-day-click ROAS on an upper funnel channel.
Ridge shares YouTube viewer audiences back into search so it can see which branded queries came from YouTube.

Part 7 · When YouTube sponsorships do not pay back

After the goals come the honest limits, where YouTube sponsorships fail on price or product, and what stays hard to measure even when they work.

7a · Where YouTube does not work

Low order values and overpriced creator categories are two ways a sponsorship fails to earn back, as HexClad found with cooking creators.

Cody does not think YouTube works for every brand, and specifically doubts it at low AOV.
HexClad copied Ridge's flat fee YouTube creator model and it did not produce performance.
HexClad concentrated 85% of its flat fee creator deals in cooking, the most expensive possible category for it.

7b · What stays hard

Leaked codes and harder click tracking mean some YouTube value never shows up cleanly, which a brand has to accept before it starts.

HexClad cannot stop discount code leakage regardless of the mechanic it uses: CleanCart shows around 500 attempted injections.
Influencer attribution is by far their hardest, partly because YouTube changed the UI and clicking through got harder
If every dollar must be attributed to an immediate click, the model will not suit your organization.

More points, not in a step yet

Find and vet a YouTube creator

Contact 150 YouTube creators a week. Grüns contacted 150 YouTube creators weekly, on top of 100 to 500 Instagram and TikTok creators contacted weekly.
Build a team for direct deals. Magic Mind stood up an entire team in Eastern Europe to do direct deals with influencers, mostly on YouTube and podcasts. It had 30 employees in Romania a year before the interview.
Hire a YouTuber to sign YouTubers. Ridge worked with Dan, a big YouTuber with about 400,000 subscribers, as a sponsor first. He is now on board signing people.
Ask for the full picture. Lily Comba's order: get the creator's media kit, get their rates, then ask for the full picture. What that holds varies by platform: Instagram, YouTube, podcast, Substack or newsletter, TikTok.
Refuse under 25% average watch time. Lily Comba's team will not work with a creator whose average watch time is less than 25%, because 75% of the video is not being watched.
Check views, not subscribers. On Colin and Samir's Ryan Trahan episode, the biggest running YouTuber at the time had about 100,000 subscribers and was getting about 10,000 views a video. The running community is inherently capped.
Ask for the by city screenshot. We ask influencers to share a screenshot of their by city analytics in YouTube and Instagram. If 20 percent of the people watching are in New York state, we can track which stores in New York state got an uptick.
Sponsor the channels the customer watches. One brand on Marketing Operators says its perfect customer is not watching a beauty YouTube channel, so sponsoring a beauty channel would not make sense. The people it is trying to reach would not be interested.

Sources: Zipchat AI | AI Agent for Ecommerce, February 2026, at 4:36 · Ecomm Cowboy, June 2026, at 16:23 · Alex Fedotoff, July 2021, at 23:54 · Marketing Operators, December 2025, at 48:13, 58:01 · Colin and Samir, June 2021, at 27:08 · Dennis Ksendzov, September 2026, at 2:13 · Marketing Operators, July 2024, at 1:18:06

Price and structure the deal

CPM is not the only way to buy. One brand on Marketing Operators typically buys distribution, on CPM. Lexus with Colin and Samir clearly is not thinking on a CPM basis: it is deeply integrating them into how it speaks to its customers.
60k plus for one YouTube integration. One brand on Marketing Operators had a couple of creators go live for 60k plus apiece for YouTube integrations, sometimes with Instagram as a piece of that. It buys them as big brand awareness: multi-million view videos.
A flat fee of 2500 to 7500. An operator on Limited Supply paid YouTubers a flat fee somewhere in the range of 2500 to 7500. There was no percent of spend, only maybe 500 bucks a month as a fee to run ads using the creator's page.
Creators remember their old CPM. An operator on Open Residency got back into sponsoring a lot of YouTubers & hit price memory. Creators said they used to get $1,000 CPMs, though their views had gone down.
Offer revenue share over a flat fee. The advice in Zipchat AI's AG1 breakdown: stop paying influencers flat fees for one post. Start with long-form creators within the niche whose audience trusts them &, if possible, give them revenue share.
Set the affiliate kickback against CPA. An operator on Marketing Operators sets the affiliate kickback off the Northbeam CPA of other top of funnel channels. It has to be lucrative to the affiliate & marginally better than what Facebook, TikTok, YouTube or a creator costs.
Fewer deals, more involved. An operator on Open Residency is now trying to do fewer YouTuber deals and be way more involved in each one. With MKBHD the brand will be on 20 of his videos this year, all very high touch.

Sources: Marketing Operators, August 2026, at 5:03 · Marketing Operators, September 2025, at 14:47 · Limited Supply, September 2026, at 19:16 · Open Residency, December 2025, at 54:21, 54:39 · Zipchat AI | AI Agent for Ecommerce, March 2026, at 10:56 · Marketing Operators, February 2025, at 1:05:09

The creator's side of a deal

Creators hold the line on price. A creator on Momentum Productions plays hardball with brands: it is fifteen hundred dollars, call back when the team has budget. YouTube is the full-time job, so a product alone is not okay for payment.
YouTube Shopping pays creators more. A creator on Jay Clouse's show says that instead of 3% on Amazon he may be getting 8%, 10% or 20% affiliate rates on YouTube Shopping, on very similar products at similar prices.
Small channels can add affiliate links. The same creator says a channel at the very first layer of monetization, the 500 subscriber and 3,000 watch hour requirement or a little higher, can add affiliates to its videos natively in the YouTube app.

Sources: Momentum Productions, February 2023, at 28:44 · Jay Clouse, August 2025, at 8:54, 7:24

What goes in the integration

Ask for a midroll, not a dedicated video. Ridge did not ask a gaming channel for a 30-minute video about why it loves Ridge Wallet. Nobody wants to make that & the opt-in rate is super low. It wanted integrated ads inside the content, like a midroll ad on YouTube.
One sponsor per video. Lily Comba thinks YouTubers are not going to agree to a dedicated video anymore, and are not going to have multiple sponsors within the same video either.
60 to 120 seconds inside 20 minutes. One brand on Marketing Operators gets 60 to 120 second ad reads inside 20 minute YouTube videos or 90 minute podcasts. It calls that a different sort of impression.
Suggest the video title. Lily Comba says a brand that understands YouTube influencer deals a little more deeply can suggest titles for the video as well.
Let the YouTuber shoot it. One brand on Marketing Operators is not bringing a creator into the studio. He is a YouTuber, and YouTube is a channel the brand is trying to show up on more, so he creates it for YouTube, super native.

Sources: Open Residency, December 2025, at 49:44 · Marketing Operators, December 2025, at 58:44, 59:38 · Marketing Operators, February 2026, at 56:46 · Marketing Operators, May 2026, at 35:53

How long a video keeps selling

Pay commission on a video that stays up. On Brando Monetti's channel, YouTube is called digital real estate: the videos stick on forever and keep getting views. A creator who posts for a brand gets views every single day, and can be given commission like 20%.
Searchable videos still earn years later. Genflow on the creator's side: a searchable idea that did well still makes money even years later, because the ads are simply still on the video.
Some creators keep using the product. One operator on Marketing Operators expects that sponsoring 40 YouTubers leaves probably 8 of them using the cookware for the rest of the year.

Sources: Brando Monetti, April 2025, at 18:33 · Genflow, May 2022, at 6:53 · Marketing Operators, February 2026, at 1:10:47

What did not work

50k a month was not enough. One brand on Marketing Operators went out with a 50k a month budget, the amount it was recommended. It was not enough, because that only tests a handful of creators, and there was no strong measurement strategy in place.

Sources: Marketing Operators, February 2026, at 1:13:32

On YouTube the audio hook matters more than the visual hook, the reverse of Meta, because everybody watching YouTube has sound on.

He thinks 40 minute watch time, 39% of sessions on a TV
You still need a hook or people skip ads
Audio hook matters more on YouTube since everyone has sound on
You have 5 seconds to earn attention, then cut slower

A 40% thumb stop on Facebook buys three seconds, whereas YouTube gives you five seconds of everybody.

Once you have attention it is easier to keep, cut slower
Hook becomes a lead into content viewers may like

Curiosity and story work as a YouTube hook where a shock visual works on Meta.

He is sure the Bobby ad hook is founder led, a story
On Meta they shred wallets to make people stop
On YouTube, open with let me tell you an interesting story

The consumption context on YouTube is different enough that ad creative has to be different, with 40 minute average watch time and 39% of sessions on a TV.

A cited brand had 30% of views on connected TV
Viewers are immersed, consuming content very differently
So in stream ad creative that performs is far different

On YouTube the audio hook matters more than the visual hook, the reverse of Meta, because everybody watching YouTube has sound on.

You still need a hook or people skip ads
On Meta a visual hook matters since not everyone has sound
After the hook you don't have to cut as quickly
You have 5 seconds to earn their attention

A 40% thumb stop on Facebook buys three seconds, whereas YouTube gives you five seconds of everybody.

No need to cut as quickly once attention is earned
Attention is easier to keep once you have it
He thinks the hook shifts toward content viewers might want

Curiosity and story work as a YouTube hook where a shock visual works on Meta.

Bobby ad hook is likely a founder story
Meta hook example, shredding wallets, purely visual
YouTube hook can open like an interesting story

HexClad's best YouTube ads teach a recipe and only subtly position the pan as the tool: two how-to-cook-steak videos are two of the top three performing ads.

Value beyond here is our product, go buy it
Facebook hook shows old pans going into an industrial crusher
That crusher hook gets 35, 40% thumb stop

The second best HexClad ad opens with Chef Daphne, winner of that season of Hell's Kitchen, simply introducing herself rather than a visual gimmick.

She offers to talk viewers through cooking a steak
Much less direct response, more narrative storytelling
Goal is value beyond the product being sold

Jones Road found the opposite on Meta, where the long form founder content works best and choppy short form took years to crack.

Narrative storytelling has worked best for them on Meta
Shorter, choppier ads were a long struggle on Meta
Maybe because Bobby is very good on video

Cody names V Shred as the best YouTube direct response he has seen, running a quiz funnel with outlandish hooks.

V Shred is a direct response supplement company
Hooks are outlandish but strong curiosity grabbers
Another host watched a fitness ad for 4 and 1/2 minutes
That ad probably was V Shred, he says

Javy is mentioned as doing a good job with creative, specifically with problem/solution ads that are entertaining.

Example: coffee ad telling viewers not to buy Starbucks
You can tell it is an ad
Still good at getting the click or the next piece

The brand's creative strategy involves making the first 3 seconds of a video either 'really weird or really funny' to be entertaining and capture attention.

Viewers know it is an ad, but it entertains
Same creator saying obsessed with this TikTok find gets skipped fast
Goal is to interrupt someone scrolling TikTok in bed

The brand uses a creative approach where the first 3 seconds of an ad show someone funneling capsules, which has performed well and increased engagement.

One of their best performers, engagement through the roof
Got backlash, can be borderline offensive around drug problems
Hard call for a media buyer when it performs best
The line to find, do people get the joke

Brent Umzu is a 'killer on YouTube' who uses shocking hooks for his ads.

Product is a gut cleanser sold with taboo wording
Hook claims you carry 18 lbs of toxic waste
That 3 second hook makes viewers ask what is this

The speaker mentions that 'this ugc needs to it needs to end,' implying a critique of certain user-generated content strategies.

Stopping someone mid scroll is the job, even knowing it's an ad
He says that UGC style needs to end
Points to Brent Umzu's taboo hooks instead