Explore AI visibility with Big Bear

Article by: 
Dave Sheppard
Founder & Digital Strategist

A B2B website rebuild produces more qualified enquiries when the new site is planned around one question: which visitors should contact you, and what do they need to see before they do? More traffic and more form fills are side effects. The decisions that change enquiry quality are made before anyone opens a design file.

In short: Agree with sales what a qualified enquiry is, and measure your current rate. Structure the homepage so each buyer type can route itself to relevant proof. Write copy that helps the wrong buyers rule themselves out. Ask for the form fields sales needs, and judge the rebuild on qualified enquiries per month rather than conversion rate.

The shortest form is rarely the best one for quality. In Atticus Li's analysis of 500 A/B tests (a practitioner review, March 2026, method not disclosed), forms with two fields or fewer produced 34% fewer downstream qualified conversions.

Decide what a qualified enquiry is before anything is designed

A qualified enquiry is a contact from someone who fits your ideal customer profile (ICP) and has a need your sales team can act on.

Your ICP is the description of the organisations you serve best: sector, size, the problem they bring and what they typically spend.

A marketing-qualified lead (MQL) is a contact marketing judges likely to buy, based on fit and behaviour. A sales-qualified lead (SQL) is one sales has reviewed and accepted as worth a direct conversation. The gap between the two is where a rebuild earns or loses its value.

So write the definition down with your sales team. Then pull the last 12 months of website enquiries and tag each one as qualified or unqualified. That tagged list is your baseline, and without it nobody can say whether the new site did better.

This is also why we don't think website projects should start with design. In our Webflow rebuild process, sitemap and content planning, then wireframes and messaging, come before custom design. Each page gets a job before it gets a look.

Structure the homepage around buyer self-selection, not features

Information architecture (IA) is how a site's pages are organised, named and linked. Self-selection is a visitor choosing the path that matches their situation, without being asked.

A good B2B homepage uses IA to make self-selection easy. The visitor should know within seconds whether the site is for them, and where their proof lives.

How should a B2B homepage route different buyers?

  • Open with who it's for. State the client type and the problem in the buyer's words rather than describing your services.
  • Route by buyer type or problem. Short paths such as "For engineering firms" or "Replacing an outdated website" let each ICP find its own page.
  • Put matching proof on each route. A consultancy wants to see consultancy work. Sector-specific case studies do more than a generic logo wall.
  • Keep one primary call to action. Repeat it down the page, and add a lower-commitment option for buyers who are still researching.

What should you cut?

Feature grids that list everything you do. Navigation that mirrors your internal departments. Hero copy that would fit any competitor. Carousels that bury your strongest proof on the third slide.

Colleagues around a sticky-note covered whiteboard during a stakeholder workshop

The routes are only as good as the ICP list behind them. For Cambridge Design Partnership, we ran in-person stakeholder workshops to define ICPs and messaging before building LinkedIn campaigns for each ICP. A homepage rebuild needs the same input, and it is a short step from that list to your main routes. You can see how we work with B2B teams on that kind of positioning work.

Write messaging that pre-qualifies before the form

Getting more people to enquire is not the hard part. The hard part is helping the wrong people decide, politely, that you're not for them.

Your copy is not there to appeal to everyone. Not every visitor, not every company size. It is there to help the buyers your sales team can help recognise themselves.

In practice, pre-qualifying copy does four things:

  1. Names the client types and sizes you work with.
  2. Describes the problems you solve in the language buyers use on sales calls.
  3. Shows what an engagement involves, including a price range where you can publish one.
  4. Explains what happens after someone enquires.

A price range is a direct filter. Our Webflow Websites projects run from £5,500 to £35,000+, and a stated range like that lets a buyer with a much smaller budget rule themselves out before filling in a form.

Design the enquiry form to filter for quality, not just reduce fields

Form field friction is the effort each field adds. Every extra or ambiguous field lowers completion, but removing the fields sales needs lowers quality. The trade-off has to be made on purpose.

Close-up of a person's hands typing on a laptop keyboard at a desk

Cutting fields is not automatically a win. In an Unbounce test reported by CXL, removing three of nine fields cut conversions by 14%. Restoring them and rewriting the labels lifted conversions by 19.21%. Which fields you ask for, and how you word them, matters more than the count.

Count still matters at the extremes. The same 500-test analysis found forms with three to five fields averaged 25% higher completion than forms with eight or more. Combined with the two-field result above, that points to a short form built from high-value fields.

FieldWhat it tells salesFrictionOur recommendation
Name and work emailWho to reply toLowRequired
Company or websiteICP fit, at a glanceLowRequired
What do you need help with?Problem and service matchLow as a dropdownRequired, as a dropdown
Budget rangeWhether the work is viableMediumBands, optional or required by intent
TimelineUrgencyLowDropdown
Phone numberLittle at first contactHighOptional
MessageContextMediumOptional, with a prompt

What is progressive profiling?

Progressive profiling is asking for fewer fields at first contact and collecting more over later touches, such as a follow-up email or a second form. Atticus Li's analysis found progressive profiling beat a single long form in 78% of tested cases.

Do inline validation and clear labels reduce friction?

Cognitive load is the mental effort a form demands. Vague labels and open questions with no example raise it.

Inline validation checks each answer as the visitor completes the field, instead of after they press submit. In a 2009 usability study with 22 participants, published in A List Apart, inline validation raised success rates by 22% and cut completion times by 42%. Messages shown after a field was finished worked better than messages shown while people were still typing.

Measure enquiry quality after launch, not just conversion rate

Conversion rate can rise while enquiry quality falls. Shorten a form and you will often see more submissions and fewer sales conversations.

So track these instead:

  • Qualified enquiries per month, against the 12-month baseline.
  • Qualified rate: qualified enquiries divided by total enquiries.
  • SQL acceptance: how many enquiries sales takes forward.
  • Pipeline value from website enquiries.

Dave Sheppard, our founder and digital strategist, puts it plainly: more clicks, more traffic and better rankings "don't matter much" on their own. So report a metric only if it connects to a commercial outcome.

Which A/B tests measure enquiry quality?

An A/B test splits traffic between two versions of a page to see which performs better. Statistical significance is a measure of how unlikely the observed difference would be if the two versions actually performed the same.

Tests worth running after a rebuild:

  1. Current form against a shorter version, judged on qualified enquiries rather than submissions.
  2. Budget field required against optional.
  3. Homepage with buyer routes against a single generic call to action.
  4. Price range shown against hidden.

Hold traffic sources and your qualification definition steady for the whole test. Decide the sample size before you start. Evan Miller's analysis of A/B testing shows why: in his example, stopping a test as soon as it looks significant pushed the false positive rate to 26.1%, against a nominal 5%. His explanation of fixing the sample size in advance is worth reading before you run anything.

Many B2B sites won't reach that sample size on qualified enquiries alone. In that case, compare a full quarter after launch with your tagged baseline, and be clear internally that it is a before-and-after comparison, not a controlled test.

If you are planning a rebuild and want enquiry quality to be the measure, our Webflow website service sets out how we plan one, and our case studies cover the kind of work involved.

Frequently asked questions

Is Webflow a good platform for a B2B rebuild aimed at better enquiries?

Yes, if your marketing team needs to change pages and forms quickly. Webflow lets marketers edit pages and content without relying on developers. The platform helps through that speed. Enquiry quality still depends on page structure, copy and form design.

How long should you wait before judging whether a rebuild worked?

Wait at least one full sales cycle, and never less than a quarter. Enquiry volume shows up within weeks, but quality only shows once sales has qualified those enquiries. Judge against the same period in your baseline to allow for seasonality.