The Efficient Spend Podcast

The episode delivers substantive frameworks and practical distinctions that a B2B operator would find valuable, particularly the taxonomy of transactional vs. lifetime-value marketing models, the concept of marginal ROAS as a north star metric, and the specific advice on when not to use MMM (new channels, strategic shifts). However, the conversation includes notable filler - the sabbatical digression, rapid-fire questions at the end, and some repetitive circling on the hybrid MMM+experimentation approach dilute the density somewhat.

“you're modeling those two events, acquisition and churn. And then you may, you know, done a lot of churn modeling in my career and done it with what are called survival models”

“The most naive models, well it, but it doesn't take much to kind of debunk one. Right. If you only spent on TV in your lowest part of the year and your model makes the conclusion that I can increase the accuracy of the model by saying TV made sales go down, that model should never be shared.”

Wallace presents genuinely useful distinctions (transactional vs. LTV marketing economics, the problem-formulation vs. regression-only approach, the idea of experiments as 'testimony' generation) that are not tired clichés. However, the core hybrid MMM+experimentation angle, while well-executed at Lift Lab, is now industry standard and the episode acknowledges other vendors have copied it. The thinking is sound but not especially counterintuitive or contrarian for an informed B2B audience.

“I see marketers conflate them quite often”

“everyone has fallen on their sword and kind of, you know, mimicked or copycatted some of the innovation”

Wallace is a genuine practitioner with deep domain expertise: founder of Data Song (multi-touch attribution), now CEO of Lift Lab (5+ years building MMM+experimentation), advisor-backed by Dr. Dominic Hansen (UCLA Anderson, textbook author on marketing response models). He has shipped products at scale, worked through the hard technical problems, and speaks from real operational experience rather than theory. This is exactly the caliber B2B operators benefit from hearing.

“my former company was called Data Song”

“our advisors, it was Dr. Dominic Hanson from the Business School Anderson School at UCLA”

Wallace provides concrete examples (branded search/affiliate economics, TV spend saturation scenarios, specific channel cascades like Instagram→Google→TikTok, cyber week impact modeling) and technical specifics (survival models, gas chromatography, incrementality test designs). However, he largely avoids naming client cases, specific ROI figures, or timestamped results. The discussion of predicted LTV model refreshes is generic; the Reddit/organic content example is hypothetical. More hard numbers and real case outcomes would strengthen this significantly.

“You can have low, low lift on something like branded search or affiliate or something like that at a high price point for your product. And the answer can come back by those clicks every time you can get a chance to.”

“if you start to land markedly different consumer behaviors that will have an impact on the consumer response auctions”

Paul demonstrates strong listening and follow-up: he challenges Wallace on the PLTV-model-on-model compounding error problem, pushes back on skepticism toward non-modeled channels, and asks probing questions about the tension between data and human conviction. Wallace responds substantively and doesn't dodge difficult questions. However, the host occasionally accepts Wallace's answers without pressing further (e.g., the AI/LLM discussion could probe deeper into real product roadmap specifics), and the rapid-fire format at the end undercuts momentum. The conversation is genuinely collaborative but not as sharp as it could be.

“I want to uh, challenge you a little bit on it because one of the things that we've struggled with, and I think many businesses struggle with, is that a change in spend, to equate that to a, uh, change in revenue, you need to make predictions around what that customer will be worth over the long time.”

“And one of the things that I've been challenged by is still believing that, like brand spend, for example, and certain executions need to be done, even though the model does not indicate that they should be done.”