Return on ad spend is useful for measuring profitable demand capture, but it is not a complete measure of marketing effectiveness. Optimizing it in isolation can favor easily attributed conversions while neglecting the creation of new demand.

ROAS remains useful but should not be the sole marketing metric
Direct attribution primarily measures demand capture
Growth also requires investment in creating net-new demand
Highly precise metrics can encourage narrow optimization

“It's a metric to use, but it's not the end all and be all metric, which is where a lot of people think of it…”
03:30

“I think ROAS is a metric on how efficient you are at capturing existing demand. The other metric you need is how efficient you are…”
Kieran Flanagan · 05:00

#roas#attribution#growth#paid-media

Hot Take· 1

Hot Take22:00

Advertising Alone Leaves Your Growth in Someone Else's Hands

Paid advertising depends on platforms or publishers to assemble relevant audiences. A company with large ambitions can eventually outgrow those external audiences, so it must develop its own ability to attract attention and create demand rather than relying solely on media buying.

Relevant platform audiences have finite scale
Advertising rents access to audiences built by others
Rapid growth can exceed a platform's audience expansion
Owning audience growth gives the business more control

“What you're relying is on other people to attract and grow a relevant audience for you.”
22:30

“you want to put the power of growth in your control, not somebody else's control. And that's why you can't rely solely on advertising.”
23:00

#audience#advertising#demand-generation#growth-strategy

Explainer· 3

Explainer07:00

How Retargeting, Discounts, and Lower Spend Inflate ROAS

A marketer can improve reported ROAS without making the business healthier. Concentrating spend on existing customers, retargeting warm prospects, discounting products, or reducing scale can raise the metric while leaving top-of-funnel growth unsolved.

Retargeting concentrates spend on people already close to buying
Discounts can improve conversion while reducing economic quality
Reducing spend can preserve only the highest-return opportunities
A rising ROAS does not necessarily indicate expanding demand

“if I do more retargeting, if I run discounts, if I reduce spend, I can drive return on ad spend to the moon, right?”
07:30

“what it doesn't solve for, and then one of the points in this thread that he's making is like it doesn't solve for the hard…”
08:30

#roas#retargeting#discounting#metrics

Explainer10:00

How an Incrementality Model Measures Otherwise Invisible Growth

Incrementality estimates how many additional customers or how much additional revenue appeared because a campaign ran, even when conversions occurred through direct, organic, or other channels. The episode illustrates this by comparing regions exposed to social video with similar regions where campaigns remained off.

Incrementality measures additional demand caused by marketing
It captures conversions attributed to other channels
Regional holdouts can provide a comparison group
Observed lift is converted into an estimated contribution ratio

“An incrementality model is basically how many additional customers am I getting indirectly from this spend.”
Kieran Flanagan · 10:00

“How much incremental demand do you get from other channels because you are spending in brand and these other kind of paid campaigns in those…”
Kieran Flanagan · 10:30

#incrementality#measurement#experimentation#attribution

Explainer11:00

What a Direct ROAS Model Actually Measures

Direct ROAS follows a user from an ad click through the conversion funnel and compares resulting customer value with acquisition spend. SaaS companies may use lifetime value and first-click, last-click, or weighted attribution to estimate the return generated by each advertising dollar.

Direct ROAS depends on traceable ad interactions
Customer value is compared with acquisition cost
SaaS models often use lifetime value
Attribution may be first-click, last-click, or weighted

“ROAS is basically I spend a certain amount of dollars within these platforms like Google Search and Facebook, that someone clicks on the ad, someone…”
Kieran Flanagan · 11:00

“Usually a great SaaS brand is in like the three to five bucket, right? I get three to five dollars back for every dollar I…”
Kieran Flanagan · 11:30

#roas#saas#customer-acquisition#attribution

Story· 1

Story17:30

Nike's Shift Toward Measurable Marketing Became a Warning

A cited teardown argues that Nike shifted resources from demand creation toward programmatic and performance marketing aimed at serving existing demand. The critique was that billions went toward activity that was easier to measure but allegedly less effective than harder-to-attribute brand investment.

Nike reportedly increased programmatic and performance spending
The strategy emphasized existing customers and measurable traffic
Measurement convenience can distort capital allocation
Demand creation may be underfunded when attribution dominates decisions

“they shifted from a creating demand strategy, which would be like an incremental strategy, to a serve and retain demand, which would be like a…”
17:30

“Nike invested a material amount of dollars, billions, into something that was less effective but easier to be measured versus something that was way more…”
18:30

#nike#performance-marketing#brand#capital-allocation

Tool· 2

Tool12:00

Using Conversion-Lift Studies to Measure Video Advertising

Video ads often influence people who later convert through another channel, making click-based attribution incomplete. Conversion-lift studies address this by turning campaigns off in selected regions, leaving them on elsewhere, and comparing changes in demand between the groups.

Video exposure may influence later non-click conversions
Use geographic markets as test and control groups
Compare incremental demand while campaigns are on and off
Repeat studies as audience and platform conditions change

“No one's gonna click on these things, right? But they're gonna consume them and then at some point convert elsewhere.”
Kieran Flanagan · 12:30

“you go dark in some states, if you use the US, and you turn them on in other states, and then you look in those…”
Kieran Flanagan · 13:00

#conversion-lift#video-ads#experimentation#incrementality

Tool21:30

Use Burst Tests to Find a Platform's Efficient Spend Ceiling

Burst tests deliberately saturate an advertising platform to estimate the maximum budget it can absorb efficiently. Once that ceiling is known, marketers can forecast when demand capture will plateau and begin building incrementality and brand programs before reaching it.

Increase spend to probe the platform's saturation point
Identify the maximum budget that remains efficient
Estimate how long current demand can support growth
Start demand-creation investments before reaching the ceiling

“you can run these things called burst tests where you saturate the platform to figure out how much budget you can spend on that platform.”
Kieran Flanagan · 21:30

“So I better start figuring out the incrementality portion and the brand portion because I need to grow the amount of demand for this business.”
Kieran Flanagan · 22:00

#burst-tests#media-buying#saturation#budgeting

Takeaway· 3

Takeaway09:30

Why Every ROAS Curve Eventually Approaches One-to-One

Profitable demand on an advertising platform is finite. As a company spends further into the available audience, it eventually reaches less responsive prospects and approaches a one-to-one return, revealing that demand capture alone cannot sustain growth.

Every platform has a finite pool of profitable demand
Additional spend eventually reaches lower-intent audiences
Market saturation pushes ROAS toward one-to-one
Demand creation must begin before efficient demand capture is exhausted

“there is no world in which at some point someone's ROAS doesn't go to one to one.”
Kieran Flanagan · 09:30

“Which means that you have saturated all of the demand that you can acquire in a profitable way.”
Kieran Flanagan · 09:30

#saturation#roas#demand#unit-economics

Takeaway14:00

Why Incrementality Estimates Must Be Recalibrated

An incrementality ratio is not permanent because markets, audiences, platforms, and campaign effects change. Brands may need to repeat lift studies every few months, accepting a temporary decline in exposed markets to maintain a credible measurement model.

Incrementality assumptions decay over time
Repeat lift studies every few months
Holdout tests can temporarily reduce demand
The short-term cost supports better long-term allocation

“you have to update your numbers every so often. So like three to six months, maybe nine months.”
Kieran Flanagan · 14:00

“So I actually will see a decrease in demand in certain states to get my number, but it's worth doing because it's the way you…”
Kieran Flanagan · 14:00

#incrementality#calibration#lift-testing#analytics

Takeaway20:00

Three Warning Signs Your Advertising Is Over-Optimized

The hosts identify directness, artificial separation, and rising conversion costs as signs that a company may be optimizing for measurement rather than effectiveness. Channel overlap should not automatically be removed because interaction between activities can compound growth.

Treating direct measurement as inherently superior is risky
Evaluating every channel in isolation ignores interaction effects
Overlap can create compounding customer growth
Rising conversion costs may signal excessive dependence on measurable ads

“One, you care about the most direct measurement, two, you care about discrete measurement.”
20:00

“That overlap is magical. That overlap is what gives you growth and what gives you a flywheel of customers.”
20:30

#marketing-metrics#channel-mix#conversion-cost#growth

Frameworks in this episode

Burst Test Channel Capacity Model

Conversion Lift Incrementality Study

Demand Capture and Demand Creation Model

Marketing Measurement Trap Warning Signs

Three Advertising Buckets

Insights & moments

⚡Myth Buster· 1

Why ROAS Is Not the Whole Growth Picture

◆Hot Take· 1