Nobody shops for a growth audit while the numbers look good. The call tends to come after one expensive quarter, when spend went up, results went sideways, and the moves that used to work stopped moving anything.
This is the buyer's side of that decision. How the audit itself gets performed I have already published: six lenses worked in sequence across the growth operating system a company already runs, the whole teardown given away in the growth audit playbook, free, down to a version you can run on yourself in about twenty minutes.
So set the how aside. Two questions decide the purchase: what actually lands on your desk when the work is done, and what it quietly costs to keep putting the decision off.
The signals you actually need one
Start with whether you need an audit at all, because plenty of good companies do not yet. The work earns its keep when a business has enough traffic and revenue for the leaks to be worth finding, and when the easy explanations have run out.
Before product-market fit, the right answer is almost always to keep hunting for the offer, and no audit will hand that to you.
The clearest trigger is an acquisition cost that keeps climbing while the team keeps insisting the channels are fine. The second is a channel that scaled beautifully and then went flat, with more budget buying the same number of customers month after month.
The third shows up live in a meeting: two people quote the cost of winning a customer, and the two figures are nowhere near each other.
If two or three of those describe your last quarter, the spend you are about to add on top is the most expensive way possible to learn what a month of diagnosis would tell you first.
The leaks are rarely visible from the top: the ANA had to trace open web programmatic dollars through the supply chain to find that of the $88 billion spent, $22 billion was wasteful or unproductive.
What a real audit hands you
Say you decide to buy. The fair question is what you are holding once it is over, because any founder has sat through enough decks that diagnose everything and change nothing.
The deliverable of a real audit is a ranked list of money: every leak priced, every fix ordered by leverage.
That list is only half of the deliverable. The other half is a calendar.
The findings get slotted into the first ninety days of work, with the items that unblock other items sitting at the top, so the sequence compounds instead of scattering your team across ten fronts at once. Each line has a named owner, so the plan survives contact with a busy quarter.
The best version of that roadmap is written so your people can execute it whether or not they ever work with the auditor again. If ongoing help turns out to be the right call, that is a separate decision with its own logic. The audit should stand on its own two feet.
What skipping one costs
The reason audits get postponed is that nothing on the calendar forces them. Your vendors have due dates. Taxes have a date.
The audit is the item with no deadline attached, so it slides quarter after quarter, behind every task that does have one. Meanwhile the budget it would protect is not growing: Gartner's 2025 CMO Spend Survey put marketing at 7.7% of total company revenue, the exact same share as last year.
Skipping an audit has no invoice, which is why it feels free. The cost is priced into every dollar you keep spending on the wrong fix.
The wrong fix is where the money actually goes. A team concludes that paid acquisition is broken and sinks a quarter into new campaigns, when the real defect is the page the traffic lands on.
At YouMail, paid looked dead because three of every four visitors bounced straight off the app store, and the channel was carrying the blame for a funnel problem. Sending that traffic into a proper web funnel instead pulled the bounce down to one in four and dropped the cost of a customer by 77%.
Every dollar poured into scaling the channel before that change was a dollar aimed at the wrong target.
None of that waste arrives as a line item you can point at in a board deck. It shows up as an acquisition cost everyone has quietly made peace with, a growth curve that flattened for reasons nobody can name, and twelve months of spend that bought a good deal less than it should have.
That is the invoice, and it is always larger than the price of the diagnosis.
What a good auditor asks of you
There is a test you can run before the work even begins: look hard at what the auditor asks you for.
A real one asks for access and very little else. Read-only entry to billing, analytics, the ad accounts, and whatever email platform you run. From there the data does the talking.
What a good auditor does not need is a standing meeting. The real findings are invisible from a conference room.
Whether your conversion events fire from the browser or the server, whether the platforms are claiming more revenue than your bank ever actually saw: that kind of defect only surfaces when someone opens the accounts and reconciles them against the money that landed.
So be wary of a proposal that is heavy on workshops and discovery sessions and thin on account access. That shape is selling you facilitation, and facilitation has never once found a broken tracking event.
The work happens heads-down in the numbers. Your calendar should barely feel it.
How to tell a real audit from a slide deck
Quality is genuinely hard to judge from the outside, especially before any money has changed hands. One tell cuts through all of it, and it lives in the recommendations. Read closely what the auditor is steering you toward.
Judge an auditor by what they tell you to stop doing. Anyone can find things to add.
The structural tells back that up. A slide deck leans on the platforms' own numbers, because reconciling billing data is real, unglamorous work.
It hands you a stack of problems with no dollar figures attached, because pricing a leak takes judgment and exposes the person who did the pricing. And it ends, reliably, in a proposal to keep the authors on.
The single most valuable line in a real audit is often the one that shrinks your budget: a campaign to kill, an agency retainer to end, a plan tier quietly putting a ceiling on customer value.
Those are the lines a firm selling its own follow-on work has every incentive to leave off the page. Economists running a large-scale experiment at eBay produced exactly that kind of line, finding the company's brand-keyword ads delivered no measurable short-term benefits.
When someone whose next invoice depends on your spending more tells you to spend less, that is when you know the analysis was honest.
If your last quarter had two or three of those signals in it, an audit is very likely the highest-return thing you can buy right now, well before you add another dollar of spend. That is the engagement I run: fixed in scope, built to be handed to your team, honest enough to talk you out of spending where you shouldn't. Tell me where it hurts.