Lead generation website not converting: form problem or proof problem
Your sales happen on calls, so you assume a website audit cannot measure them. The page still owns one countable step. Two counts tell you whether you have a form problem or a proof problem, and what the gap is worth each month.
A lead generation website not converting has a problem you can measure, because the page's job ends at the booked call, and booked calls are countable.
"My sales happen on calls. A website audit cannot see any of that."
True about the calls. Wrong about the audit, and the distance between those two sentences is where your money sits. Nobody signs a 12,000 euro engagement off a web page. Nobody books the call that sells it without one.
What is your lead generation website responsible for?
One step. Turning a stranger into a booked call or a submitted enquiry. That is the entire job, and it is a countable event. Gartner puts the scale of it plainly: buyers considering a purchase spend only 17% of that time meeting with potential suppliers.
So the page is not a brochure supporting the sale. For most of the buying it is the only version of you in the room, and your close rate never gets tested on the people it loses.
Is it a form problem or a proof problem?
Two different failures that need two different weeks of work. A form problem means people wanted to contact you and the path broke. A proof problem means they never got far enough to want to. Separate them with two counts, and the larger loss names your problem.
Take last month. Total visits, visits to your contact or demo page, submitted forms.
- 1,400 visits last month
- 126 reached the demo page, so 1,274 never did
- 31 submitted, so 95 arrived and left
That site loses thirteen people before the form for every one it loses on it. The form is not the problem. The page in front of it never made the case. Shorten the form first and you spend a week recovering 95 people while 1,274 keep walking.
Want both numbers priced rather than counted by hand? Revslip reads your pages and ranks the losses.
Why does the page keep selling after your call ends?
Because you were never talking to the buyer on their own. Gartner finds a typical complex B2B purchase involves six to 10 decision makers. The person on your call has to re-argue your case to five or nine people who never met you, and what they forward is a link.
Gartner also measured what good information is worth. Customers who found the information they received helpful were 2.8 times more likely to experience a high degree of purchase ease, and three times more likely to buy a bigger deal with less regret.
That is the proof problem with a price on it. Your case study page is the argument your champion makes in a meeting you are not invited to.
What is one point of lead rate worth to you?
More than it is worth to a shop, because your conversion is not an order, it is an opportunity. Revslip's free estimate runs on traffic, conversion gap, order value and mobile weight. A consultative sale needs two numbers only your CRM has.
monthly visits × lead rate gap × qualified rate × close rate × deal value = monthly recovery
- Visits. 1,400 last month, from GA4.
- Current lead rate. 2.2%, so 31 enquiries.
- Target. 4.4%, taken from your best service page, never a published average.
- The gap. 31 extra enquiries a month.
- Qualified rate and close rate. 40% and 25%, so three extra deals.
- Deal value. 6,000 euros, so a bit over 18,000 euros a month.
Halve it before it goes near a plan. The first two inputs are measured, the last three are your estimates, and estimates multiplied together drift fast.
The part of this we cannot see Revslip reads your rendered public pages, roughly 200 conversion signals, and on a paid plan it tracks the changes you make through to revenue using your integrations. It cannot tell you whether the enquiries were any good, or what happened in the 48 hours after one landed. Harvard Business Review's 2011 audit by Oldroyd, McElheran and Elkington found most companies answer online leads far too slowly, which has never been a page problem. So connect HubSpot, compare one month of submissions against the ones your team called back, and if enquiries arrive and die, your money belongs in follow-up rather than a page audit.
Why do the published B2B benchmarks put you in two places at once?
Because the label you pick changes the number by nearly three times. Ruler Analytics reports B2B Services averaging 3.5% and Professional Services averaging 9.3%, in one report, on one method. A consultancy fits both labels honestly. Pick whichever one flatters the quarter.
The spread inside a single label is wider still. B2B Services averages 3.5% while top performers reach 15.9%, and B2B Tech averages 1.7% against 13.6% at the top. That is four and a half times, among businesses sorted into the same bucket, which tells you the bucket is not the thing explaining the difference.
Ruler says its database is mostly high-budget advertisers chasing higher value leads, which lifts every figure in it, and it sells attribution software. Take your target from your own best page.
What do you change first?
Whichever step loses more people, which in the example above is reach, not the form. Work in this order, recount after each one, and never change two things in the same week or you will not know which of them moved.
- Name the buyer and the problem above the fold. Not your category. A visitor is checking whether you have done this before, for someone like them.
- Put one named result on the page they land on. Company, number, timeframe. A wall of client logos is wallpaper.
- Say what the call actually is. How long, who turns up, what they leave with. A vague ask reads as a trap.
- Then shorten the form. Name, work email, one line about the problem. Every extra field has a measured cost.
- Settle the pricing question. Hiding it also filters out people who would have paid. Both sides of that, with the arithmetic.
Rather have the list ordered by euros, not effort? Run a free audit on one page.
How long before the new lead rate means anything?
Longer than you want. At 1,400 visits and 31 enquiries a month, a two-point move is roughly 30 extra enquiries, so you are comparing months, not weeks. Give it 30 days minimum, and compare against the same month last year if your pipeline is seasonal.
Your deal cycle hides the rest. A change made in March shows up in enquiries in March and in closed revenue in July, by which point four other things have changed. Read the enquiry count as the fast signal and revenue as the slow one. The 20-minute diagnostic is the version to run before any of this.
Who should not bother with this?
Anyone whose pipeline does not come through the website. If your deals arrive by referral, partner introduction or outbound, the site is a credibility check and nothing more, and auditing it is tidying an empty room. Your lead rate will still be low. It will also be irrelevant.
One thing about us. Revslip routes its audit logic by business type across eCom, SaaS, agency and other, so a consultancy lands in agency or other and gets no bespoke B2B services profile. Set your conversion goal to the enquiry and read the messaging and trust findings first. Our proof is 134 businesses audited, every one requested by an owner who already suspected something, which is a flattering way to pick a sample.
Committee size, supplier time share and purchase ease from Gartner on the B2B buying journey. Averages from Ruler Analytics, lead conversion rates by industry. Response time research from Oldroyd, McElheran and Elkington, Harvard Business Review, 2011.