UX Optimization
User experience work on a B2B site is mostly detective work. Something is stopping a proportion of qualified visitors from getting to an enquiry, and the job is to find out what rather than to guess.
Why not just apply best practice?
Because best practice is a set of averages, and your site is a specific case. A rule that holds across consumer e-commerce may be irrelevant to an engineer looking for a compatibility chart.
Applying a checklist produces changes nobody can evaluate, and when the numbers move there is no way to know which item did it. Starting from observed behaviour produces a short list of specific problems instead.
What we actually look at
Four instruments, each answering a different question, none sufficient alone.
| Instrument | Question it answers | Its limit |
|---|---|---|
| Analytics | Where do people leave, and from which pages | Tells you where, never why |
| Session recordings | What were they doing when they left | Time-consuming, and easy to over-read a handful of sessions |
| Heat and scroll maps | What gets seen and clicked, and what never does | Aggregates dissimilar visitors into one picture |
| Form analytics | Which field they abandoned on | Only covers people who started |
Analytics narrows it to a page. Recordings and form data explain what happened on it. Used the other way round you spend hours watching sessions on pages that were never the problem.
What turns up most often on B2B sites
- The form asks too much too early. Company size, budget range and project timeline on a first enquiry, none of which a curious buyer will supply.
- The information a technical buyer needs is missing. No dimensions, no tolerances, no compatibility detail, so they leave to find it elsewhere and buy there.
- Navigation organised around the company. Structured by internal division rather than by what a visitor is trying to do.
- Critical content below a wall of introduction. Scroll maps routinely show most visitors never reach it.
- Mobile treated as an afterthought. Tables that overflow, forms that fight the keyboard, tap targets sized for a mouse.
- No obvious next step. The page ends and the visitor is left to work out what to do.
Asking users directly
Useful, with a caveat: people are reliable about what frustrated them and unreliable about what they would do. "I would definitely request a quote" predicts very little.
So take the complaints seriously and treat the predictions as hypotheses. A short prompt at the point of abandonment tends to produce more honest material than a general survey.
How we know a change worked
Where traffic allows, by testing. Where it does not, which is most B2B sites, by measuring carefully before and after and being honest that seasonality is a competing explanation.
Some changes do not need proving. A form that silently fails on mobile, a table unreadable on a phone, a page that takes eight seconds: repairing those is maintenance, not an experiment.
Does this include accessibility?
Yes, and they overlap more than people expect. Clear focus states, sensible heading structure, labelled form fields, adequate contrast and sane tap targets improve the experience for everyone, not only for people using assistive technology.
A form that a screen reader cannot navigate is a form that is confusing sighted users too, just less obviously.
Where this fits
Inside conversion optimization. It follows measurement, because behaviour data is only interpretable when conversions are recorded correctly, and it usually produces the hypotheses that testing then proves. Structural rebuilds are UI and UX design.
Frequently asked questions
Do you redesign the site?
Rarely, and not as the first move. Most findings are specific and local: a form field, a missing specification, an unclear next step. A redesign replaces everything at once, including the parts that were working, and makes it impossible to tell what helped.
How many session recordings do you need to watch?
Enough to see a pattern repeat, which is usually a few dozen on a specific page rather than hundreds across the site. One vivid recording is a story, not evidence, and it is very easy to redesign a page around a single confused visitor.
Will recording sessions create a privacy problem?
Not if configured properly. Form inputs should be masked so no personal data is captured, and the tool has to be disclosed in the privacy policy. Both are settings people forget rather than genuine obstacles.
See where you stand
We will review your current search visibility and show you what buyers find when they look for what you do.
Contact Us