A service page serving two buying intents ranks for neither and converts on neither. The diagnosis takes ten minutes in Search Console, but the repair can turn one approved page into two briefs, two review cycles and two URLs to maintain. B2B service page optimization starts with a constraint many agencies resist: one page, one job.
A shared page remains valid only when buyers use different words for the same service and take the same next action. Complexity is allowed; competing commercial decisions are not.
B2B service page optimisation is decided before layout
Design cannot settle a commercial argument the brief has avoided.
Use the single-intent test:
For [one query], this page helps [one buyer] choose [one service], then asks them to [one action].
If you cannot state the page’s one query and one next action in a sentence, the page is split and should be two pages. The query can include close wording variants, but every variant must describe the same purchase.
Lock four pre-build decisions:
- Primary keyword: the phrase representing the service decision.
- SEO title: the promise searchers should choose from the results page.
- H1: the on-page confirmation of that same promise.
- Primary action: the single commitment the page requests.
The pre-build rule is absolute: keyword, title, H1 and primary action are decided before layout, not retrofitted after design sign-off. Retrofitting them produces several of the redesign errors that blur the buying decision.
Our Lanteria work dealt with a broad Microsoft 365 HR platform serving multiple stakeholder audiences. The architecture routed those audiences instead of forcing every capability into one argument. The Lake Erie Shores project made the same structural decision for two distinct audiences: staying at the resort versus owning there.
Layout cannot rescue a page whose query and action disagree.
Diagnose split intent in ten minutes with Search Console
Open Performance, select Search results and apply an exact-page filter. The diagnosis then has four timed checks:
- Minutes 0–2 — isolate the evidence. Review the latest 90 days, then compare the latest 28 days versus the preceding 28 days. Filter to the country the service actually targets.
- Minutes 2–5 — group the queries. Write each substantial query group as a concrete pairing: “buyer wants X and should do Y”. Avoid abstract labels.
- Minutes 5–8 — calculate the split. Add the impressions for each group, then divide each total by the page’s impressions. Record clicks, click-through rate and average position alongside them.
- Minutes 8–10 — inspect the page. Compare each query group with the title, H1 and primary action. Choose Split, Retitle or Strengthen.
Apply three operating rules. These are editorial heuristics, not Google thresholds:
- Split: two query groups each contribute at least 25% of impressions in both date views and require different actions. Create one page for each job.
- Retitle: an aligned group has at least 100 impressions, an average position of 1–10 and click-through below 2%. Test the title and meta description before rebuilding the page.
- Strengthen: at least five aligned queries sit in positions 11–20. Rewrite the title and H1 first, then improve supporting proof and links.
Positions 1–10 trigger a snippet test versus positions 11–20 triggering an alignment rewrite. Average position remains an aggregate, not a fixed daily ranking.
What Search Console cannot tell you is whether the offer, price, proof or sales response caused weak conversion. That requires a focused audit of B2B conversion friction, not another interpretation of query data.
Our claim is falsifiable: if, after 90 days with stable offers and indexing, both successor pages still receive at least 25% of impressions from the other page’s query group and neither beats the combined page’s qualified-action rate, the diagnosis failed for that case.
Search Console exposes divided demand before design obscures the conflict.
The eight-block B2B service page optimisation template
Buyers need each block to remove a specific obstacle to the same decision. The template contains eight blocks, in this order:
- Query promise — “Have I reached the right service?” Confirm the buyer, service and outcome in the title, H1 and opening.
- Buying trigger — “Do you understand why action is needed?” Describe the operational problem that makes the service relevant now.
- Outcome — “What will be different?” State the expected business change without inventing certainty, percentages or deadlines.
- Scope — “What exactly are we buying?” Define deliverables, boundaries and exclusions clearly enough for an internal sponsor to explain.
- Delivery — “How will the work happen?” Explain the sequence, client responsibilities and likely demands on the buyer’s team.
- Fit — “Is this sensible for our organisation?” Name the conditions that make the engagement suitable and the conditions that do not.
- Proof — “Why should we believe the claim?” Match the buyer’s main risk with a relevant project, artefact or attributable client statement.
- Next action — “What happens after I respond?” Repeat one primary action and explain the immediate next step.
The structure is visible in our AfriCap Hub work: catalogue and filtering supported event discovery, while registration handled commitment. For Savgen, a technical multi-industry offer needed clear routing by application rather than one page carrying every industrial argument.
Tone comes after the job is fixed. A practical framework for choosing brand personality can shape how the argument sounds, but not which argument belongs on the page. Where the diagnosis requires new routes, restructuring and redesign support for B2B sites should begin with those routes rather than decorative layouts.
Every block earns space by resolving one buyer question.
Internal links for B2B service page optimisation need a numerical rule
A contextual link declares which destination matters and why. Navigation links alone cannot express that editorial priority.
Our rule is 3–5 deliberate contextual links into a service page from closely related content, not 30–50 incidental references scattered across the site. Global navigation and footer links do not count. A page with 100 internal links has no important ones.
The three strongest source types are a related problem article, a relevant case study and the parent capability page. A service page about redesign, for example, could receive a contextual route from a conversion-focused B2B redesign guide.
If fewer than three relevant pages can link in naturally, add the missing routes before publishing. If the service page contains more than seven contextual outbound links, remove destinations that do not answer an objection blocking the primary action.
Use descriptive anchors that predict the destination’s value. Repeating the target keyword as every anchor makes the site sound engineered rather than useful.
Internal links should declare priority rather than merely prove connectivity.
If this analysis exposes wider gaps in your site, we can turn the evidence into a focused redesign brief — book a call
What does not lift a B2B service page
The version that reaches us is usually carrying three failed fixes:
- Word-count padding. A 1,200-word page resolving all eight buyer questions is stronger than a 2,000-word page repeating four of them. Extra words do not repair a confused query, missing proof or weak action.
- Keyword density. Moving from 1% repetition to 3% repetition measures typing, not alignment. Searchers need the page to answer their service decision; no useful density target can confirm that.
- A second call to action. One action repeated three times reinforces the job. Two different actions—such as “book a consultation” versus “download a guide”—create two commitments and divide the page again.
Supporting links and ordinary navigation can remain. The failure begins when a secondary action competes for the same visual weight, copy emphasis or success measure as the primary action.
More copy cannot repair a divided commercial decision.
A UK B2B worked example in pounds
Take an illustrative UK cybersecurity consultancy deciding whether to replace one combined service page with two focused pages. The model uses six labelled inputs:
- One-off strategy, copy, design and build investment: £8,000 excluding VAT
- Gross contribution per won engagement: £10,000
- Qualified-enquiry-to-win rate: 20%
- Qualified service-page sessions: 250 per month across both successor pages
- Evaluation window: six months
- Current qualified-enquiry rate: 1.20%
The eight arithmetic lines are:
Contribution per qualified enquiry = £10,000 × 20% = £2,000
Extra qualified enquiries needed for break-even = £8,000 ÷ £2,000 = 4
Qualified sessions in six months = 250 × 6 = 1,500
Required rate increase for break-even = 4 ÷ 1,500 = 0.267 percentage points
Break-even enquiry rate = 1.20% + 0.267% = 1.467%, rounded to 1.47%
Extra enquiries required at a 2:1 contribution-to-cost threshold = (£8,000 × 2) ÷ £2,000 = 8
Approval enquiry rate = 1.20% + (8 ÷ 1,500) = 1.733%, rounded to 1.73%
Modelled contribution versus build cost = 8 × £2,000 = £16,000 versus £8,000
The commercial decision is explicit. Moving from one combined page at 1.20% to two focused pages requires 1.47% to break even or 1.73% to pass the 2:1 approval threshold.
The honest limit here is that the win rate and contribution are illustrative assumptions, not Actualyse results. The estimate is confounded by lead quality, sales execution, traffic redistribution and offer changes.
Commercial modelling turns a page split into a testable investment decision.
FAQ
Four operational questions settle common implementation choices.
How many proof examples should the page contain?
Start with one detailed example matching the buyer’s highest-risk constraint. Add a second only when it answers a materially different objection; ten unexplained logos should not replace one relevant case narrative.
Should a service page publish pricing?
Choose Price disclosure when three or fewer variables determine the fee. When four or more variables materially change scope, name those variables and make “request a scoped estimate” the primary action.
Should paid search and organic traffic use the same page?
Reuse the page only when three elements match: query, offer and primary action. If one differs, choose a campaign page built around one response instead of compromising the organic service page.
Where should an old URL point after a page split?
Use a 301 redirect to the successor inheriting the old page’s main query and add one contextual link to the second successor. Never redirect both buying jobs to an unrelated hub.
A 301 redirect must preserve the old page’s primary promise.
Summary
Use these six decision rules:
- Split any page whose query and next action cannot fit one sentence.
- Lock the keyword, title, H1 and primary action before layout approval.
- Split when two query groups each exceed 25% of impressions and require different actions.
- Build 3–5 relevant contextual links into every service page.
- Remove word padding, density targets and every rival primary action.
- Approve a page split only after defining its £ break-even threshold.
One measurable job should govern every service page.
Actualyse designs and rebuilds B2B websites that turn research visits into qualified pipeline. Book a call to talk through where yours stands.
