Most website briefs begin with a look: “clean, modern, premium, easy to navigate.” These are reasonable preferences, but they are conclusions before the argument. A site can be all four and still fail to explain the business or produce an enquiry.
Before design starts, answer these seven questions. The answers will shape the sitemap, the copy, the build and the budget.
1. What job is the website doing?
A website can be a shop, a credibility document, a lead-generation tool, a booking system, a support library or a campaign landing page. These are different products.
If the job is lead generation, the site needs a clear offer and a short path to enquiry. If the job is credibility, it needs evidence: work, people, process, proof. If the job is e-commerce, it needs product data, inventory, payments and fulfilment. Trying to do all of these equally usually means none is done well.
2. Who is the primary visitor, and what do they already know?
A referral visitor may arrive knowing the brand and needing reassurance. A search visitor may know the problem but not the company. A paid visitor has seen a specific promise and expects the page to continue it.
Write down the two or three most important visitor situations. Then check whether the homepage serves them or only serves the company's desire to describe itself.
3. What is the one action you want?
Enquiry, call, booking, purchase, download, application, visit. One primary action is enough. Secondary actions can exist, but they should not compete visually with the main one.
“Contact us” is weak because it describes the mechanism, not the value. “Request a proposal”, “Book a site visit” or “Get the pricing guide” tells the visitor what will happen.
4. What does the visitor need to believe before they act?
Every purchase or enquiry involves a set of hesitations: Is this for me? Can they do what they claim? Is it worth the price? What happens if it goes wrong? Who is behind this?
The site should answer those questions in roughly the order they arise. That is what page structure is for. It is also why a gallery of pretty images with no explanation rarely converts.
5. Who will write and maintain the content?
The most common cause of website delay is not design. It is content that nobody has time to write, approve or gather.
Decide before the project starts:
- Who writes the first draft?
- Who approves the claims?
- Who gathers the images and permissions?
- Who updates the site after launch?
- How often, and with what technical comfort?
The answers determine whether you need a developer on retainer, a simple CMS or a static build with a documented update process.
6. What platform and constraints are real?
Platform choices should follow the requirement, not a trend. Consider:
- Do you need to publish frequently?
- Are there integrations: CRM, payments, booking, inventory, email?
- Who hosts it, and who is responsible for security and backups?
- Are there accessibility or procurement requirements?
- What is the budget for build and for ongoing maintenance?
A simpler build that your team can manage is often better than a powerful platform that becomes a dependency.
7. How will you know it worked?
Define the measures before launch. Depending on the site, they might include:
- Qualified enquiries per month.
- Enquiry-to-conversation rate.
- Booking completions.
- Time to first response, if the site feeds a sales team.
- Search visibility for priority pages.
- Page speed and mobile usability.
- Content updates completed without developer involvement.
Vanity metrics are easy to collect. Choose the ones that would cause a decision if they moved.
A website is not finished when it looks right. It is finished when a real visitor can understand the offer, trust it enough to act, and complete the action without help.
What this means for the build
If the objective is clear, the site can be scoped tightly: a small number of templates, a clear content model, a tested conversion path and sensible analytics. If the objective is unclear, the project will expand in every direction — more pages, more features, more opinions — because there is no criterion for deciding what belongs.
Spend the time on the seven questions. The design will be better for it, and the build will be shorter.