B2B Website Redesign: Strategy Before Templates

A B2B website redesign fails when it starts as template shopping. Someone finds a SaaS theme, a dark hero, a logo wall, and a fake product UI. Then the team tries to pour a six-month sales cycle into a homepage that was designed for a two-week trial.

B2B sites sell to a committee. There is a user, a budget owner, a technical reviewer, and often a founder who will google you at 11pm. The site has to serve all of them without pretending they are the same person. That is strategy and information architecture. It is not a color system.

I will not start a B2B redesign in Figma until the jobs, proof, and paths are written down. Templates come last. If that sounds slow, compare it to shipping a pretty site that sales still has to explain on every call.

Start with the buying committee, not a moodboard

List the humans who touch a deal. For each one, write what they need to believe, what they fear, and which URL should do that job. The economic buyer wants outcomes, risk, and commercial clarity. The user wants workflow fit. The technical reviewer wants architecture, security posture, and implementation reality. Procurement wants process.

Most B2B homepages try to speak to all four in one paragraph and end up speaking to none. The redesign should assign pages to people, then decide what the homepage is allowed to promise.

• Who lands from a sales email versus paid search versus a partner referral.
• Which page they need on the first visit versus the third.
• What they forward to a colleague, and whether that URL currently exists.
• Where deals stall, in the sales team's own words, not in marketing adjectives.

If you cannot get those notes from sales, the redesign is already guessing. Interview three closed-won and three closed-lost conversations. You are collecting objections, not testimonials.

Jobs the site has to do across a long cycle

A short-cycle site can live on a homepage and a pricing page. A long-cycle B2B site has to work for months. That means different jobs: category education, solution fit, proof for a skeptic, implementation detail, and a commercial next step that does not feel like a trap.

Write the jobs as sentences. 'Help a VP explain us to a CFO in five minutes.' 'Help a practitioner see the workflow without a demo.' 'Help a security reviewer find the facts without a sales call.' Each sentence deserves a URL, or a clearly named section on a URL. If two jobs share a page, say so. Do not hide a security page inside a blog tag.

If sales keeps sending PDFs because the site cannot be forwarded, the site is not in the deal. A redesign that ignores that will look better and still lose.

Map content you already have onto those jobs. Case studies, comparison pages, implementation guides, webinar recordings, and pricing explainers often exist as leftovers. The redesign's job is to put them in the architecture, not to commission 40 new essays before the nav is honest.

Information architecture before the visual system

IA for B2B is a hub and spoke model. Hubs are commercial: industry, solution, product, or service. Spokes are the questions the committee already asks. Internal links should make that relationship obvious. A solution page that does not point to proof is a brochure. Proof that does not point back to a commercial URL is a dead end.

1. Name the hubs. Usually fewer than you think: the offers you actually sell.
2. Assign existing URLs to hubs. Merge overlaps. Do not keep three 'platform' pages that fight.
3. Give each hub a proof object: a customer story, a method page, or a public artifact.
4. Give each hub one primary action. Demo, call, or application. Not all three as equals.
5. Only then draw the sitemap the designer will see.

This is also when you decide what dies. B2B sites accumulate campaign landing pages, old product names, and thought-leadership posts that cannibalize the hub. A redesign is a chance to pick winners. It is not a chance to preserve every URL out of fear. Retire on purpose, with redirects, after you know which URL is the winner.

Proof architecture, not a testimonials carousel

Committees do not convert because a carousel rotated. They convert when the risky claim has evidence next to it. If you say you implement in weeks, show the process. If you say enterprises use it, show a named environment they can recognize. If you cannot name metrics, do not invent them. Use specificity: stack, scope, constraint, outcome in words the customer actually used.

Place proof where the decision happens. Under the offer on the solution page. Next to pricing assumptions. On the implementation page, not only on /customers. A logo wall with no link to a story is decoration. A story with no path back to the offer is journalism.

Homepage, solution pages, and resource hubs

The homepage is a router with a point of view. It should name who it is for, what you do, and where to go next. It should not be a dump of every product surface. Solution or industry pages do the segmentation. Resource hubs do the education. Pricing or engagement pages do the commercial clarity.

B2B redesigns often invert this. The homepage becomes a magazine. Solution pages become feature grids. The blog becomes the only place with honest language. Then paid traffic hits the homepage and sales wonders why the calendar is random.

The conversion layer is offer, proof, and one action. I unpacked that in the pipeline rules I use on commercial pages. Use that on the money templates after the IA is set, not as a substitute for strategy.

What to put on a solution page

• The job in one sentence, in the buyer's words.
• Who it is for and who it is not for.
• How the work happens, at a level a practitioner can respect.
• Proof adjacent to the claim.
• One next step, with what happens after they click.

SEO constraints the strategy has to respect

Strategy that ignores search will still collide with it. You cannot launch six overlapping solution pages for the same job and hope Google picks the right one. You cannot rename every slug because the new IA used different product language. You cannot bury the only ranking guide because the new nav is 'simpler.'

Once the sitemap is honest, run the same-CMS SEO timeline: freeze, constrain templates, QA, launch, watch. That sequence is in what to fix before, during, and after a visual redesign. Strategy decides which URLs survive. The SEO timeline keeps them alive through the build.

Use Search Console to see which queries already map to which URLs. Those mappings are votes. If you want a new hub to win a query, plan the internal links and the title. Do not publish a second near-duplicate and wait.

Crawlability and unique URLs are still the baseline. Google's SEO starter guidance is boring on purpose. B2B redesigns get into trouble when they skip boring.

When you finally look at templates

Templates are a delivery mechanism for the jobs you already named. Shop them only after you know the page types: home, solution, industry, story, resource hub, article, pricing or engagement. If a theme cannot support those types without a hack, it is the wrong theme. Do not reverse the order because a Dribbble shot had nice type.

Component libraries should encode the strategy: offer block, proof block, objection block, next-step block. If the library is only hero, logos, and a three-column feature row, you bought a brochure kit. You will spend the project fighting it.

If the current site is technically messy, fix the system before you buy a content retainer to fill the new templates. That split is when a growing company needs a technical specialist versus an SEO agency.

B2B templates from marketplace kits usually encode a short-cycle SaaS story: hero, logo wall, three features, testimonial, demo form. A services firm, an industrial vendor, or a slow enterprise product will look wrong in that kit even if the type is nice. If your cycle includes a security review, you need a place for that packet. If your cycle includes a partner, you need a URL they can white-label in an email. Template shopping before those URLs exist is how you spend the budget on a homepage animation.

Comparison and 'versus' pages can be useful when the sales team already has a talk track. They are harmful when they are keyword sketches with no operational difference. Write them from win/loss notes. Link them from the hub that actually sells the offer. Keep one winner per competitor or category, not a cluster of near-duplicates.

A practical kickoff for this work is short. One workshop with sales and marketing. A crawl plus Search Console export. A list of URLs that already attract the queries you care about. Then a one-page IA: hubs, spokes, primary actions, proof objects. That packet is what a designer or an agency should receive. If they ask for brand colors first, they are not designing a B2B system. They are decorating.

Governance matters after launch. Who can add a new solution page? Who can spawn another 'platform' URL because a campaign needed a headline? Uncontrolled page creation is how B2B sites return to mush six months after a redesign. Put a rule in the CMS: new commercial URL requires a job sentence, a parent hub, and a proof block. No job sentence, no page.

I would rather own the product of a B2B site (the paths, the proof, the CMS fields) than restyle a maze. If you send the current IA and a typical deal cycle, I can tell you whether you need strategy, a visual refresh, or both, and in which order.