When positioning work should come before a redesign
Use a fictional before-and-after service offer to separate an unclear proposition from a usability defect, and decide what must be settled before redesigning pages.
Separate an offer decision from a broken interaction
A business can need clearer positioning and a working form at the same time. If the team keeps changing the intended buyer or the service being sold, a new layout cannot settle those choices. If an enquiry fails to submit, the failure needs investigation now. Agreement on the offer is not a prerequisite for repairing an observable defect.
I recommend resolving the audience and offer before a broad redesign when those decisions would change the page’s purpose. That does not mean every weak enquiry result is a positioning problem. Visitors may encounter a confusing route, a broken interaction or an offer unrelated to their need. Diagnose which boundary is failing before approving the larger project.
The example below is fictional. It contains invented offer copy and conditional observations, not participant quotations, a completed market study or an Ulties service promise. No interview, prototype session, form-delivery test or conversion improvement was observed.
Turn the hypothesis into an offer someone can examine
Assume a small consultancy has agreed that its first engagement is a review of one customer-onboarding workflow. Its intended buyer is an operations lead coordinating work across forms, spreadsheets and internal tools. For this exercise, the team can provide the review outputs described below, but software implementation is outside that engagement.
Fictional before copy: “We build digital solutions for ambitious businesses.” It gives the reader no specific task or deliverable to assess. A designer could create a polished page around it while the team still disagrees about whether the visitor is buying a review or a complete replacement system.
Fictional after copy: “Find where customer onboarding loses a handoff. For operations leads coordinating an existing workflow across forms and spreadsheets. We map one workflow and recommend the next change to investigate. You receive a handoff map with unresolved questions and a written recommendation. Software implementation is a separate decision.”
The revised specimen names the buyer’s situation and makes the first deliverable distinct from building software. It deliberately contains no saved-hours figure, delivery deadline or claim of market leadership. It is more specific against the stipulated scope, but that alone does not establish comprehension, demand or commercial value.
I recommend settling this narrower offer before redesigning the service-page structure. I would reject a full replacement-platform headline because the assumed first engagement does not deliver that platform. Keep limited visual exploration possible, but do not ask layout work to decide which service the business intends to provide.
Use research to challenge the specimen
For a real project, review suitably redacted sales questions and ask people who match the intended audience about a recent comparable decision. GOV.UK’s interview guidance recommends current or likely users, open and neutral questions, and concrete experiences. Ask what they were trying to achieve and what alternatives they considered. Liking a slogan is a different question from understanding whether the service fits their situation.
Next, put the specimen into a simple page prototype with a realistic contact route. GOV.UK describes moderated usability testing as observing people attempt relevant tasks. Use a task about deciding whether to enquire about an onboarding review. Ask participants to explain what they expect to receive before showing them an ideal answer. Observe where they look and what prevents completion.
Record the wording shown, the task context and the observations that support each interpretation. Recruit for the decision being studied and include relevant access needs. Stakeholder agreement can stabilize an internal brief; it cannot stand in for audience evidence.
Let different observations change different parts of the plan
Hypothetical branch A: if a participant expects the first engagement to include a new onboarding application, the review-versus-build distinction has not landed for that person. Revisit the offer wording and exclusions before investing in a detailed page system around the wrong expectation. Check the revised wording again. This would be a comprehension lead, not proof that an entire market misunderstands the service.
Hypothetical branch B: if the participant accurately describes the review deliverables but cannot find how to ask about suitability, investigate the page hierarchy and contact route. The observed obstacle would be finding the next step, so changing the audience definition would not directly address it. A navigation repair may be the narrower next experiment.
Hypothetical branch C: if a clearly marked synthetic enquiry is rejected unexpectedly or cannot be traced to the receiving system, investigate the form path immediately while message work continues. Distinguish browser acknowledgement, application acceptance and downstream receipt. A confirmation screen alone would not establish that the intended team received anything. This article has not tested a contact system.
Give positioning work an exit condition
For the fictional consultancy, the internal handover can state the chosen operations audience, the onboarding-review deliverable and the exclusion of implementation. It should also retain unresolved questions about suitability, evidence and commercial terms. The specimen is ready for audience testing under those assumptions; it is not approved public copy backed by completed research.
For a real redesign decision, agree which uncertainties the research must resolve and who can approve the offer. Proceed when the evidence is sufficient for the specific page decisions being funded, with remaining risks named. If observations challenge the offer, revise it. If they expose a local usability defect, repair that defect without reopening every strategic decision.
After release, define what counts as a useful enquiry before comparing outcomes. Note changes in traffic mix or measurement that could complicate interpretation. This article supplies no baseline, experiment size or statistical winner. A more specific headline is a hypothesis to examine, not a measured sales result.