Marketing runs on comfort blankets. The Sharp Cut cuts them up to see what's inside. This week, we did it as a murder mystery.

The victim: the click. Once upon a time a click was one honest fact, a machine noting that a file left a building. Over eleven years it got poisoned, dose by dose, until it was taking credit for sales it never caused.

The body: in 2012, eBay turned off roughly $51M in paid search across a third of the US, with matched control markets. Scored the way a dashboard scores it, the return was over 4,000%. Scored against the control, the only way that actually answers the question, it was minus 63%. Same company, same spend, same year.

Then we line up six suspects, the auction, the platforms, last click attribution, the CFO, the agency, and us, and ask each one the same two questions: did they have a reason, and did they have the access?

Marc and V get out of the costume for the part that stings: the doses we added ourselves. The Sport Chek war room. The machine V built to move money faster. The P&G cut everyone quotes that fails our own evidence test. And what happens to the click when the shopper is an AI.

We won't hand you a verdict, because we don't fully agree.

You're the jury.

Enjoy the show!

Chapters:

00:00 What the Sharp Cut is

00:49 A murder mystery: who killed the click?

01:30 Meet your six suspects

01:45 The body: eBay turns off $51M in paid search

03:35 4,000% ROI or minus 63%? Same campaign, two scores

03:57 What a click actually was (the first banner, 1994)

05:23 The 44% click rate nobody can verify

06:25 Cause of death: poisoned over eleven years

06:42 Three doses: price, bouncer, then credit

08:43 When a measure becomes a target (Goodhart's Law)

09:09 Whodunit: motive and opportunity

11:21 The evidence: Facebook's own 2016 deck

12:45 Out of the costume: the doses we added ourselves

15:28 The Sport Chek war room18:16 Building a machine to move money faster

19:41 The P&G $200M cut, and why it fails our own test

20:49 The alibi that holds: same test, opposite answer

21:27 The click's replacement: AI shoppers

22:39 The verdict is yours

24:06 One warning before you turn anything off

Sources:

Allouah, A., Besbes, O., Figueroa, J. D., Kanoria, Y., & Kumar, A. (2026). What is your AI agent buying? Evaluation, biases, model dependence, and emerging implications of agentic e-commerce. In Proceedings of the ACM Web Conference 2026 (pp. 8697–8700). https://doi.org/10.1145/3774904.3792943

Blake, T., Nosko, C., & Tadelis, S. (2015). Consumer heterogeneity and paid search effectiveness: A large-scale field experiment. Econometrica, 83(1), 155–174. https://doi.org/10.3982/ECTA12423

Chan, D. X., Yuan, Y., Koehler, J., & Kumar, D. (2011). Incremental clicks: The impact of search advertising. Journal of Advertising Research, 51(4), 643–647. https://doi.org/10.2501/JAR-51-4-643-647

Golden, J., & Horton, J. J. (2021). The effects of search advertising on competitors: An experiment before a merger. Management Science, 67(1), 342–362.

Gordon, B. R., Zettelmeyer, F., Bhargava, N., & Chapsky, D. (2019). A comparison of approaches to advertising measurement: Evidence from big field experiments at Facebook. Marketing Science, 38(2), 193–225. https://doi.org/10.1287/mksc.2018.1135

Lewis, R. A., & Rao, J. M. (2015). The unfavorable economics of measuring the returns to advertising. The Quarterly Journal of Economics, 130(4), 1941–1973. https://doi.org/10.1093/qje/qjv023

Lysen, S. (2013). Incremental clicks impact of mobile search advertising. Google. https://research.google/pubs/incremental-clicks-impact-of-mobile-search-advertising/

Simonov, A., Nosko, C., & Rao, J. M. (2018). Competition and crowd-out for brand keywords in sponsored search. Marketing Science, 37(2), 200–215. https://doi.org/10.1287/mksc.2017.1065

The Procter & Gamble Company. (2017). Additional definitive proxy soliciting materials (DEFA14A). U.S. Securities and Exchange Commission. https://www.sec.gov/Archives/edgar/data/0000080424/000119312517299631/d464866ddefa14a.htm

Facebook. (2016). Everything competes with everything [Slide presentation].

GoTo.com, Inc. (1999). Form S-1 and Form 10-Q. To be reformatted as SEC legal sources.

Nielsen. (2009). NetEffect CPG home scanner panel meta-analysis of 200+ online campaigns. As reported in Facebook (2016).

Nielsen. (2015). BrandEffect meta-analysis of 478 online global campaigns, October 2014 to April 2015. As reported in Facebook (2016).