Check the page before you touch the ads
If qualified traffic is arriving and converting at under 2%, more budget just buys more bounce. Conversion work is the cheaper lever: the same media spend produces materially more pipeline, and the gain persists.
Confirm the traffic really is qualified first — search terms, geography, job titles. If it isn't, that's a media problem and no amount of copy testing will rescue it.
Cut form friction first
This is almost always the largest single win in B2B. Every field you add is a reason to leave, and most of what sits on a B2B form exists for internal routing rather than for the visitor.
- Keep only what you need to make contact and route the enquiry.
- Move qualification questions to the follow-up call or a second step after submission.
- Never ask for information you can enrich automatically from a work email.
Rebuild above the fold around the problem
Most B2B heroes describe the company. Visitors are scanning for their own situation. Name the problem in the headline, state the outcome in the subhead, and put one unambiguous action in view — then support it with proof a sceptic accepts: named clients, numbers, certifications.
Test one variable at a time
Headline, hero value proposition, CTA copy, form length — one at a time, run to significance, measured on clean GA4 events for genuine submissions rather than platform-reported form fills. Bundled changes tell you something moved without telling you what.
Keep a written log of every test and its result. Six months in, that log is worth more than any individual winner because it tells you what this audience responds to.
Protect the gains through a redesign
If a redesign is coming, treat it as a migration: full 301 mapping, pre-launch crawls, on-page parity checks. I've shipped this exact sequence with zero rankings lost — but only because the migration had its own QA rather than riding along with the new design.
Frequently asked questions
What is a good B2B landing page conversion rate?
For high-intent paid traffic to a lead-gen page, 2–5% is a normal band and above 3% is strong. Compare against your own baseline rather than a public average — offer, price point and traffic source move it more than design does.
How long should an A/B test run?
Until it reaches significance on your traffic volume, and at least two full weeks so weekday and weekend behaviour are both represented. Stopping early on a promising lead is the most common CRO mistake.
How many form fields should a B2B form have?
Enough to contact and route the enquiry — usually name, work email, company and a short message. Everything else belongs on the follow-up call.
Want this applied to your account?
I run this playbook on live budgets. See the case studies or get in touch.
Start a conversation