Conversion Audit vs CRO Consulting: Which One First

Three clinical epidemiologists proposed a different way to read a drug trial in the New England Journal of Medicine in 1988. Instead of asking whether a treatment works, ask how many patients have to receive it before one of them benefits. They called it the number needed to treat, and it exposes what a significance test hides. A drug can clear every statistical bar and still need dozens of people treated to prevent one heart attack.

The conversion audit vs CRO consulting decision is that same question in a different setting. An audit is a one-off read of a single page that comes back with a ranked list of what is likely costing conversions on it. A consulting engagement is a standing team that scopes a funnel, runs tests across quarters, and owns the program end to end. Most teams compare the two on price, which sets up the wrong arithmetic. What matters is how much either of them has to move, and for how long, before it has covered what it costs.

A retainer is a treatment with a price attached to each outcome it produces. The trial result is not in dispute. Consulting engagements move conversion rates, and the good ones move them substantially. The number needed to treat asks the question that comes after. How much lift, across how many visitors, sustained for how many months, before the engagement covers what it costs? Most teams sign before anyone writes that sentence down, because the sentence requires a diagnosis, and the diagnosis was the thing they were buying.

What a CRO agency cost actually buys

A CRO consulting engagement is a named team taking responsibility for your conversion program. They scope the funnel, run a research phase that usually includes user interviews and recording review, propose a testing roadmap, and manage the execution of it over a period measured in quarters.

That is real work and it is worth real money. It is also the wrong instrument for the question "what should we look at first," in the same way that a surgical team is the wrong instrument for "is this worth operating on." The engagement assumes the diagnosis. Its first weeks are spent producing one, which you are paying for at program rates.

Our own positioning research puts the median CRO agency cost in five figures a month, with the low end at a few thousand and the top end more than twice the median. The same research puts onboarding an agency at around two months, and hiring the equivalent person in-house at six. I have no reason to think those numbers are unusual, and you can check them against any two proposals you have on your desk right now.

Set that against the commitment shape. You agree to the monthly figure in month zero and find out what it has to move in month three, if anyone ever computes it. Run the arithmetic the other way and it gets uncomfortable fast: take your monthly traffic to the page in question, your current conversion rate, and the revenue per converted visitor, and work out how many additional conversions a month the engagement has to produce before it has paid for itself. Then ask what percentage-point lift that is. For a lot of mid-market teams the honest answer is a lift large enough that no single page change will get there, which is useful information to have before the contract rather than after it.

When to hire a CRO consultant, and when not to

Hire one when the problem genuinely spans the organization. A pricing-model change, a self-serve to sales-assist handoff, an enterprise flow with procurement in the middle of it, a full-funnel rebuild where six teams have to move together. These need someone with standing to run a program, not a report. An audit cannot convene a meeting, cannot arbitrate between product and sales, and cannot sit in a room for six months making sure the roadmap survives contact with a reorg.

Hire one when you already know where the opportunity is and you lack the hands to work it. That is a capacity problem with a clear answer.

Do not hire one to find out where to start. That is the case this whole piece is about, and it is the most common one. A team has plateaued, the obvious fixes shipped last year, and the next move under consideration is a scoping call. What they actually need is a read on the page, and a read on the page does not cost a quarter of a program budget.

There is an honest middle, and it is where most of these decisions belong. A team runs the audit, ships the two or three fixes that turn out to be page-level, and takes the remaining findings into a scoping conversation as evidence rather than as a hunch. An agency given a ranked list of what is wrong scopes a tighter, cheaper engagement than an agency given a dashboard and a worry, which is a slightly awkward thing for the agency and a good thing for you.

What does an automated conversion audit do that a retainer doesn't?

An automated conversion audit reads the page itself and returns a ranked list of what is likely costing conversions on it, with the evidence pinned to the screenshot, in minutes, without a scoping call. It answers one question well: given what is actually on this screen, what would a reasonable visitor struggle with, and which of those problems is worth fixing first?

The way it gets there is worth a paragraph, because the difference from a consultant's first three weeks is smaller than you would expect. The page is captured, scored against named conversion health audit dimensions, and each finding is tied to a specific element with an estimated impact and a confidence level next to it. Our audit methodology writes out how the scoring works and where it is weakest. The output is closer to a consultant's findings deck than to an analytics report, which is the point.

Here is the part where I should be straight about what we cannot do. An audit measures the page, not your buyers. It does not know that your customers are procurement-heavy and will tolerate a longer form than the benchmark suggests. It does not know your sales cycle, your competitive position, or that the reason the pricing page is confusing is a compensation argument between two VPs. A good consultant learns all of that in week two, and the findings are better for it. When a Nudgent audit is not sure about a finding it says LOW confidence, and a LOW from us is genuinely weak evidence rather than a polite hedge. You should discount it accordingly.

The other limitation is that an audit predicts and does not verify. It cannot tell you whether the fix worked. That is what your analytics and your session recordings are for, and running the audit does not excuse you from measuring the result.

Running the number before the scoping call

Try this before the next proposal lands.

1. Pick the single page you would point an engagement at first.
2. Write down its monthly sessions, its current conversion rate, and the average revenue from a conversion.
3. Work out how many extra conversions a month would cover a five-figure monthly retainer, and turn that into a percentage-point lift.
4. Run an audit on that page and read the top three findings.
5. Ask whether those three findings, fixed, plausibly produce the lift from step three.

Step five is the whole exercise, and it has two honest outcomes. Sometimes the findings are small and the arithmetic says no single page will carry a retainer, which means the engagement needs a broader scope or a different page or should not happen yet. Sometimes the findings are large and specific, in which case you may find you can ship them yourself in a fortnight and put the retainer conversation off for a quarter.

Either way you have walked into the scoping call holding a number. The other side of that call has done this many more times than you have. Turning up with an estimate of what the work has to produce changes the shape of the conversation more than any negotiating tactic will.

What to do next

If the live question is "where do we even start," the cheapest correct move is to diagnose one page before committing to a program. Run the audit on the page where the money actually changes hands, read the ranked findings, and ship the ones your own team can ship. Keep the rest for the scoping conversation. You will know within an afternoon whether this is a page problem or a program problem, and those need very different amounts of money. More of how we think about this sits across the blog and the glossary, and we have drawn the same line against a named agency in the Conversion.com comparison.

One last thing, and it is the part of this I find uncomfortable.

Before the next scoping call, what does the engagement have to move, over how long, to cover what it costs? I mean a figure somebody worked out, not the one printed in the proposal. And if nobody on the team can produce one, what exactly is the proposal going to be compared against?