Website redesign prerequisites for B2B SaaS

Every design agency in the world will tell you they can start with your current positioning and improve the site around it. Very few will tell you that unclear positioning makes design more expensive, not less. The truth is that every round of design feedback that touches a strategic question - who is this really for, what are we actually claiming, why would they believe us - is a positioning round billed at design-round prices.

A homepage designed around a clear, specific, tested positioning statement ships faster, revises less, and earns the demo more reliably than a beautiful homepage designed around a positioning statement that the founder, the CMO, and the head of product all wrote differently on the same day.

The prerequisites checklist below is not bureaucratic gatekeeping. It is a time-and-money-saving filter. If your team passes every check in the table above, the sprint will stay inside its timeline and its quote. If your team is red on more than two rows, every additional week you spend on design instead of positioning will compound. The checklist exists because we have run sprints on unclear briefs and watched the revision count double. We stopped doing that. Now you know before you book.

Rob Kaminski, co-founder of Fletch PMM, shared three specific tests on Youroast Ep1 that we now use before every sprint kickoff. They are not complicated tests. They are uncomfortable ones, which is why most teams avoid them until a design agency forces the issue in round three of revisions.

The first test is the team survey. Ask every person who will have input on the design sprint to write down - separately, without discussing - what your product does, who it is for, and what problem it solves. Compare the answers. If you get different answers from the founder, the head of marketing, and the PMM, you do not have a positioning problem that design can solve. You have a strategic alignment problem that needs a positioning engagement first. Read Youroast Ep1 with Rob Kaminski for the full Fletch framework.

The second test is the stranger test. Find someone who does not know your company - a friend, a cold LinkedIn connection, a user researcher. Show them your homepage with the logo covered and give them thirty seconds to read. Ask them to describe what the company does and who it is for. If they cannot, the hero is not ready for a visual upgrade - it is ready for a rewrite.

The third test is the sales-call audit. Pull the last ten discovery calls. What were buyers actually shopping for when they found you? If the answer differs meaningfully from what your hero currently claims, the site is misaligned with the market's actual vocabulary for the problem you solve. Design cannot fix that misalignment. Positioning work can.

The most common prerequisite failure we see is teams arriving with multiple ICPs and calling them "segments." Segments are a legitimate growth strategy at Series B and beyond. They are not a homepage strategy at Seed to Series A.

A homepage designed for five ICPs is a homepage that speaks to none of them with enough specificity to earn the demo. Every section that tries to address multiple buyers simultaneously makes a compromise that weakens the message for all of them. The hero claim gets vaguer. The proof blocks get more generic. The CTA becomes less urgent.

Rob Kaminski's advice from Youroast Ep1: go uncomfortably narrow. If adjacent segments think "cool, but not for me" when they read your homepage, you are specific enough. If every segment thinks "this might be for me," you are not specific enough. The feeling of being too narrow is almost always a sign you are in the right place.

The practical rule: by Series A, put at least 80 percent of GTM on one ICP. Secondary segments can exist in sales. They should not exist on the homepage. Design the homepage for the best customer you have ever closed. Everyone adjacent to that profile will recognize themselves anyway - and the best customer will see themselves first.

The most expensive way to discover that your positioning is wrong is to run paid acquisition at scale into a homepage that cannot convert. The second most expensive way is to redesign the homepage first and discover the positioning was wrong during the first revision round.

Before burning budget on paid, outbound, launch video, or a design sprint, baseline how your team talks about the product today. Not how you want to talk about it - how you actually talk about it in sales calls, in emails, in LinkedIn posts. That baseline is your real positioning. The gap between it and the homepage copy is the first design brief.

Choose one positioning path before you choose a design partner. Existing category positioning - we are the X for Y - is simpler to design for because the buyer already has a mental model for the category. New category positioning - we are building something that did not exist before - requires more careful proof sequencing and a sharper 2D claim to earn attention before the buyer understands what they are looking at. Both are valid. Neither can be designed effectively without being chosen first.

Weroast partners with positioning specialists when teams need PMM depth before design. We do not treat positioning work as part of a design sprint - it is a prerequisite for one.

Rob Kaminski introduced a distinction on Youroast Ep1 that changed how we brief every sprint: the homepage carries a 2D message, and sales carries a 3D message. They are not the same message and they should not be written the same way.

The 2D homepage message is the simplest, most memorable version of your positioning. One clear claim. Three to five reasons to believe it. Enough specificity to earn a demo request from the right buyer. No more. The homepage is not trying to close the deal - it is trying to earn the conversation. Founders who try to put the sales conversation on the homepage over-explain the product in the hero, stack three different claims in the subheading, and add seventeen bullet points to the feature section because they cannot bear to leave anything out. That impulse kills homepage conversion. Depth belongs in demos, not above the fold.

The 3D sales message is dynamic, adaptive, and tailored to the specific buyer in the specific room. It handles objections, explores use cases, demonstrates depth, and proves ROI with specifics. It is a different artifact for a different moment. Trying to put the 3D message on a 2D surface produces a homepage that is thorough, comprehensive, and fails the stranger test every time.

You are ready for a design sprint when the checklist at the top of this page is green on every row, or when you have a specific plan to address the red rows before the sprint kickoff. Not after the sprint starts. Before.

You are ready for the homepage sprint specifically when one URL is the bottleneck - conversion on paid traffic, demo visits, or a launch moment. You are ready for Website + Design System + Claude Setup when the homepage message is proven in market and you know the next twelve months will require a pipeline of feature, use-case, and comparison pages. Read when to redesign a SaaS website for the stage-specific signals.

Sam Hogan at Hirebird said they tried half a dozen designers before Weroast and always come back. That consistency does not happen when positioning is unclear going into the sprint. It happens when the brief is sharp enough that every design decision follows naturally from it.

Seed to Series B SaaS and AI teams about to redesign, rebrand, or scale paid acquisition to a homepage that is not converting. Founders and PMMs who want to arrive at a sprint kickoff with everything locked so they can move at sprint speed instead of discovery speed. Teams working with positioning specialists like Fletch who want to connect positioning work directly to a design brief.

Teams looking for unlimited visual exploration without a conversion target. Pre-PMF teams who have not yet identified one dominant ICP - read the pre-PMF website design guide instead, which will tell you directly whether a design spend makes sense at your stage.