A Conversion Consultant Does Five Things. Testing Is Only One of Them.

A conversion rate optimization consultant checks that your numbers are real, studies how visitors behave, turns the findings into hypotheses, tests or stages changes, and reports what moved. Testing is one step of five, and for many sites it is not the right step yet. This guide walks through the work in order, shows how to tell whether your traffic can support an A/B test at all, and lists what to ask before you hire anyone. GrossiWeb's own approach is described on the CRO consulting page, and the steps below apply to any consultant you evaluate.

Step 1: A measurement check comes before everything else

The first job is to confirm that your conversions are counted correctly. If the form, call or purchase is not tracked, every later conclusion rests on a guess. A consultant checks that the actions you care about are recorded, that each one fires once per real action, and that your own team's visits are not mixed in with customer traffic.

In Google Analytics, a recorded action that matters to the business is called a key event, and Google's documentation says any event you collect can be marked as one (About key events). If tracking turns out to be broken, fix it first and write down the date, because conversion counts from before the fix cannot be compared with counts after it. The reporting side of this is covered under analytics and reporting.

Step 2: Research shows where visitors stop and why

Research finds the step where people drop out and gathers evidence about the reason. Quantitative work such as funnel analysis shows where visitors leave. Qualitative work such as heatmaps and session recordings shows where they click, scroll and hesitate.

Reading form submissions and sales call notes adds the customer's own words to the picture. GrossiWeb's own process starts with a funnel audit that quantifies drop-off at every step, then adds heatmap and session analysis. If you want a plain-language list of common causes before you start, read why a website gets traffic but no leads.

Step 3: Hypotheses turn findings into testable bets

A hypothesis states what you will change, why you expect it to help, and which number will show it. "Visitors from our ad stop at the form because the page never says what happens after they submit, so adding that explanation should raise form completions" is a hypothesis. "Make the button bigger" is not.

A good consultant ranks hypotheses by the strength of the evidence behind them and the effort needed to ship them. The work starts where the evidence is strongest, not where someone has an opinion.

Step 4: Testing or staged changes, depending on your traffic

Test when the page has enough visitors and conversions to finish an A/B test, and ship staged changes when it does not. An A/B test splits visitors between the current page and a variant and compares their conversion rates. The staged approach is simple: ship the best-supported fix, record the date, and compare before and after.

A before and after comparison is weaker evidence, because seasonality, ad changes and other shifts happen at the same time. It is still better than a test that can never finish. GrossiWeb's service copy reaches a similar conclusion: a site with too few visitors to reach significance is scoped as a research and redesign project rather than a testing program.

Step 5: Reporting ties changes to a business number

Good reporting says what changed, which metric moved, how confident the result is and what happens next. It should end in leads, calls or revenue, not only a conversion rate percentage. A variant that "won" but did not change the number of booked jobs has not helped you.

Reporting also keeps the work honest across months. The broader approach to turning visits into inquiries sits under lead generation and conversion.

How to tell whether you have enough traffic to A/B test

You have enough traffic when the page you want to test can reach a sample size you calculated in advance, within a period in which you can keep the page and the offer stable. The sample size comes from standard statistics. A sample size calculator built for A/B tests asks for four inputs: the baseline conversion rate, the minimum detectable effect, the significance level and the statistical power. It then returns the number of visitors each variation needs (Sample Size Calculator, Evan Miller).

1. Pick one page and one conversion, and read its current rate from your own analytics.
2. Decide the smallest improvement that would change a business decision.
3. Enter both into the calculator, set the effect to absolute or relative to match how you wrote it down, and choose a significance level and power.
4. Multiply the per-variation number by the number of versions, which is two for a simple A/B test, and compare the total with the visitors that page actually receives in a few weeks.

Small differences demand much larger samples, which you can see by lowering the minimum detectable effect in the calculator and watching the required number rise. A page that cannot reach the number in a reasonable time is a page for research and staged changes, not for a test. If the answer is unclear, a consultant should show you the calculation, not a promise.

Fix the sample size before you start and do not stop the moment a result looks significant. Evan Miller's article on the topic explains that checking a running test repeatedly and stopping at the first significant reading inflates false positives, and that committing to a sample size in advance avoids the problem (How Not To Run An A/B Test).

What to ask before you hire a CRO consultant

Ask questions that make the consultant show the method, not the pitch. The answers tell you whether you are buying a process or a promise.

• How will you verify that my conversions are tracked correctly before you start?
• What research will you run, and what will it show me that my analytics cannot?
• Given my traffic and this page, can you show the sample size calculation and how long a test would take?
• If my traffic is too low to test, what will you do instead?
• Who can edit the site, and how quickly? A test plan needs a site that can change fast.
• What will the report contain, and which business numbers will it use?
• Who owns the data, accounts and test setups when the engagement ends?
• Do you promise a specific lift? Be careful if the answer is yes, because no one can know in advance how much a change will improve conversion.

GrossiWeb, which serves businesses nationwide, quotes CRO consulting after a scoping call, because the work depends on traffic volume, the number of funnel steps and whether the site can be changed quickly enough to test. The quote lists the audit, research tools, planned tests and reporting cadence, so you can check it against the questions above.

What to do this week

1. Choose the one conversion that matters most, such as a call, a form or a purchase, and confirm it is recorded. Submit a test inquiry yourself and watch it appear in your reports.
2. Write down the current conversion rate for the single page that matters most.
3. Run a sample size calculation for that page and compare it with your real traffic.
4. Watch a handful of session recordings of that page before you change anything.
5. If you are about to raise ad spend, fix the page first. GrossiWeb recommends conversion work before increasing Google Ads spend, because conversion work produces more inquiries from the traffic you already pay for.