Five receipts separate operators from performers: scope, basis, mechanism, a miss, and a referee. A real operator produces all five in conversation, because they are just an accurate description of work actually done. This page defines each one, then runs the test on me.
Let me open with the disclosure that should be on every page like this: I am exactly the kind of vendor you should be suspicious of. A marketer, marketing himself, to people who have been burned by marketers.
You have no way to know yet whether this site is a track record or a performance. The profession you are hiring from selects for persuasion, which means the interview is our home field and your away game.
So instead of asking you to trust the confident tone, I will hand you the test I would use in your seat. It works on candidates, agencies, consultants, and me.
What evidence should a marketer be able to show?
Scope comes first because everything else hangs on it. A named company, real dates, and the sentence "here is what I owned, and here is what I merely touched." Owned means the outcome was theirs to lose.
The distinction matters because growth wins are made by systems and teams, and most marketers will present collective team metrics as personal results. That warning has circulated among founders for years, and the ownership sentence is the antidote.
Basis is the receipt that dies fastest under questioning. A real number travels with three companions: a denominator, a timeframe, and a measurement method. "Grew revenue 300%" is noise until you know 300% of what, over how long, measured where.
A number that arrives alone is an anecdote wearing a suit.
The mechanism test
Ask them to explain why their best win worked. In causal terms. Slowly, if needed.
An operator who found a real lever can teach it to you on a whiteboard, because they had to understand it to pull it. When a plan-mix change beat every ad optimization at a client of mine, the explanation fit in two sentences about bid ceilings and customer lifetime, and any founder can follow it.
Someone who was standing near a win when it happened explains it differently: with adjectives. The campaign was innovative, the team was data-driven, the results speak for themselves. Mechanism is where borrowed wins go to die.
The same receipt protects you from a subtler failure: the win that was real but unrepeatable. A lever explained causally can be judged for whether it exists in your business too. Luck explained causally falls apart in the telling, usually by the second whiteboard question.
The miss and the referee
Receipt four is a specific failure, told unprompted, with what it changed. Real operators have an inventory of misses, because real operating produces them at industrial rates. A candidate whose history contains only wins is describing someone else's career or editing their own.
Receipt five is a referee: a person who held the revenue number above them, whom you may actually call. And when you call, one question outperforms the rest.
If they had left halfway through, what would have stopped happening?
References describing owners answer instantly, with a list. References describing observers go quiet, because honesty here requires admitting the machine would have kept running.
The red flags in any case study
The same test, inverted, reads any case study or portfolio in about a minute. Four flags account for most of the inflation you will meet.
A fifth flag lives off the page: polish that outruns specifics. Someone whose frameworks and posts are immaculate while their answers to the four questions above are vague has shown you which skill got the practice hours. Polish and capability are separate claims, and they require separate proof.
This checklist covers the people. The parallel exercise for their work product, reading the fingerprints inside your own ad accounts, takes 30 minutes and pairs well with it.
Run it on me
Scope: Rocket Mortgage, four and a half years in-house on the consumer funnel, where I owned conversion experimentation on the mortgage application flow. FX Replay, 18 months across consulting and CMO, where I owned the growth P&L end to end. What I merely touched at both is longer and less interesting, which is the point of the receipt.
Basis: the $2B figure is incremental loan volume from two A/B-tested changes, measured as lift against control on completion and close rates. The 2.6x is ARR over 18 months. Denominators and timeframes live in the linked case studies, which exist so you can check them.
The miss: most of my experiments fail, publicly and by design. The Rocket program won 42% of the time, against a 25% goal, which means the majority of what I shipped there did nothing or lost, and the operating system existed to make those misses cheap and instructive. Mechanism for any of it, I will happily whiteboard on a call.
The thin spot, named before you ask: my receipts concentrate in performance, funnel, and lifecycle at companies whose products people already wanted. If your problem is building a brand from nothing, my evidence there is adjacent rather than direct, and you should weigh it accordingly.
Referees who held the number above me are available on request, and the reference question above is the one I would encourage you to ask them.
Bring this checklist to the first call, with me or with whoever else you are evaluating. The gates I run in the other direction are also published. Let's talk.
