# Paid acquisition: scaling spend and proving it works — Summary The Performance Marketing work group answers paid acquisition questions in group chats. They start with paid social that has stalled and ad costs that keep rising, then turn to proving what the spend drives. Updated: 2026-10-13 HTML page: [Paid acquisition: scaling spend and proving it works]() Transcript: [Full Chat Transcript]() Work group: [Performance Marketing]() ## Questions ### [Scaling paid social past a plateau]() Before adding budget, check how the channel is judged. Grading it on platform reporting or a blended CAC target can make a channel with room left look tapped out. - **Grade the channel on company numbers.** Mark Fiske sees channel teams graded only on platform reporting, which ignores incrementality and diminishing returns. He lines it up with company-level results, blends in post-purchase surveys, and moves the target from blended CAC to marginal payback. - **Push lookalikes wider.** Timothy Davis splits customer lookalikes into 1%, 2–4%, 5–7% and 8%+ bands and rebuilds them from the highest-value customers. Mark adds that the bid signal and efficiency target have to favor those customers too, or the platform drifts back to the cheapest converters. - **Concentrate or diversify.** Daniel Pearson would keep funding Meta while its marginal dollar still returns well. Mark would fund a fixed learning budget on a second channel early, judged on falling CPA and customer quality, because finding a working formula takes months. - **Rerun lift tests, then check them.** Timothy treats each platform lift result as a multiplier on reported numbers and reruns it, though below roughly $50K a month a test has too little signal. Once paid social is a large share of the mix, Mark checks platform lift against a matched-market test and trusts the geo test when the two disagree widely. ### [When paid social costs rise: the ads, the audience or the offer?]() Find the step where the cost rises, change one thing at a time, and judge the result on customers and revenue rather than platform CPA. - **Read CPM, then CTR.** Timothy Davis reads a higher CPM with steady CTR and conversion as a pricier auction. A CTR drop on the same audience points to tired creative, which a fresh ad confirms. Clicks that stop converting call for an audience test with the same ad and page. - **Separate the offer from checkout.** Mike Jacobs holds the ad and audience fixed and breaks out each step from product page to completed order, for direct and organic visitors too. A drop at shipping or payment is a checkout fix. A drop before add-to-cart across sources calls for testing the price or promise. - **Judge winners on payback.** Daniel Pearson compares the customers each variant brings in by cohort payback, since a cheaper first purchase can bring buyers who never return. A discount that lifts conversion can still weaken the offer's economics. - **Check the platform's numbers.** Mark Fiske reconciles platform conversions with new customers and revenue, and runs a holdout, volume permitting, if they diverge. Micah Moreau uses mature cohorts to estimate how many engaged visitors will buy within 60 to 90 days, and caps spend on a new audience while its cohort matures. ### [Measuring whether paid marketing adds customers]() Compare people or markets that see the ads with ones that don't, divide the lift by spend, then follow those customers until they repay what they cost. - **Start from a counterfactual.** Mark Fiske asks what sales would have been without the spend and uses a matched-market test or holdout to get close. Timothy Davis runs lift studies and reruns them, because the multiplier can move as audiences change. - **Platform reports can overclaim.** At Ancestry, Mark's quarterly mix model refresh could show a channel about 20% less efficient than its platform suggested, after spend grew 30 to 40%. He tests around the proposed spend increase, since incrementality can fall as a channel saturates. - **Then check payback.** Daniel Pearson tracks how long the incremental cohort takes to repay its cost, and checks LTV forecasts against realized revenue before raising the next budget. He tests the highest-spend channels first. - **Test channels that run together.** Mike Jacobs randomizes matched markets into a group that gets the extra spend and one that doesn't, with other channels unchanged, and reports a range when effects can't be separated. For lead gen, Timothy measures lift on qualified leads or paid accounts, rather than raw form fills. ### [When a marketing mix model earns its cost]() Build one when a wrong channel split would cost more than the model, the spend history can separate channels, and finance and channel owners will change budgets on what it says. - **Size the decision first.** Mark Fiske funds a model when the next budget call is large, for example when a consumer business adds TV or other offline media. If the history, the budget choice or the buy-in is weak, he starts with matched-market tests and post-purchase surveys. - **Check the spend history.** Daniel Pearson plots weekly spend by channel against customers and margin. If Meta and Google always rose together, the model's split between them is hard to trust. If its recommendation flips when a period is left out, he holds off and runs a geo holdout. - **Calibrate it with experiments.** Daniel, Timothy Davis and Mike Jacobs check the model against holdouts and lift tests, and feed results back in as spend and creative change. Mike notes that in one analysis about half of the purchases driven by social had no click. - **Tie it to cash.** Mark wants every scenario to show incremental customers, contribution margin and payback. Micah Moreau multiplies the low ends of the customer and margin estimates, and caps the increase if that case breaks the cash limit. ## Participants - [Timothy Davis]() — Growth at ElevenLabs. Former Senior Lead, Growth Marketing at Shopify - [Mark Fiske]() — Managing Director at H.I.G. Capital. Former VP of Growth and Marketing at Credit Karma - [Daniel Pearson]() — Co-founder of Bamboo. Advisor to HexClad Cookware - [Mike Jacobs]() — Chief Growth Officer at Consortium Brand Partners. Former VP of Growth at Etsy - [Micah Moreau]() — VP of Growth at OpenAI. Former VP of Growth Marketing at DoorDash --- Copyright © 2026 Super Me Network, Inc. All rights reserved. | [Terms of Service](https://www.superme.ai/terms) | [Privacy Policy](https://www.superme.ai/privacy)