A discovery workshop is the one meeting in a website project where the decisions are still cheap. Everything built afterwards rests on what was agreed in that room, which is why a shallow session is the most expensive hour of the whole project. Growth Bully, a Malta performance marketing agency, inherits sites that were designed carefully against a brief that never said what the site was for, and the repair always costs more than the original page did.
The agenda matters less than the decisions. A workshop that ends in a sitemap and a mood board has not done its job. A workshop that settles what the site is measured on, where its visitors come from and how an enquiry becomes a conversation has done most of the project already.
What is a website discovery workshop?
A structured session before any design begins, where the business and the people building the site agree what problem the site solves, who it is for, what it must be measured on and what is in scope. It ends in written decisions rather than notes.
Most workshops are run by the people who will build the site, so they lean naturally towards build questions: structure, templates, content types, integrations. Those are real questions and they belong in the second half of the meeting. The first half decides whether the project as scoped is worth running at all.
What should a discovery workshop actually decide?
Six things, in this order: the commercial outcome the site is accountable for, the one or two visitor tasks that produce it, where the traffic comes from, how an enquiry is captured and routed, who owns the accounts and the content, and what is explicitly out of scope. Everything else follows.
- The outcome. One sentence naming what the site should increase, with a number attached to it before design starts, not after.
- The two priority tasks. What a visitor must be able to do in under a minute. Every third priority is a page nobody will maintain.
- The traffic. Which channels send the visitors, because the answer changes the pages.
- Capture and routing. What counts as an enquiry, where it lands, who owns it, and how fast it is answered.
- Ownership. Whose accounts the domain, hosting, analytics and ad platforms sit in, in writing.
- Out of scope. The list that protects the launch date, agreed while nobody is under pressure.
Those six fit on one page. If your discovery output runs to forty slides and does not contain them, the project has documentation rather than direction. The same test applies when you brief a marketing agency: specificity about the outcome beats volume about the deliverables.
Who needs to be in the room?
Whoever can say yes without checking, whoever answers the phone when an enquiry arrives, and whoever will publish on the site after launch. Three people with authority beat nine with opinions. If the person who signs off on copy is absent, the project has already acquired its first delay.
The person who handles enquiries is the one most often left out and the one with the most useful information in the building. They know which questions buyers ask first, which objections repeat, and which enquiries were never worth answering. That is the raw material for the pages that will carry your lead generation work, and it is free.
Why does the traffic question come before the sitemap?
Because a page built for someone arriving from a search result and a page built for someone arriving from a paid advert are not the same page. Decide the channels first and the structure follows from them. Decide the structure first and you will rebuild pages later to fit the traffic.
Search traffic wants depth, internal structure and pages that answer a specific question. Paid traffic wants a single offer, no navigation and one action. Social traffic wants proof before it wants detail. A site designed without that conversation ends up with a navigation tree that serves nobody in particular and a set of landing pages bolted on in month four.
What should discovery decide about lead capture?
Four things, before design: what counts as an enquiry, which form fields are genuinely required, where a submission lands and who owns it, and which events the site must fire so advertising can be measured. Agree them in the workshop and they get built. Raise them after launch and they get retrofitted.
Retrofitting is where the money goes. A form that posts to an inbox instead of reaching your CRM is a two hour job during the build and a week of untangling afterwards, because by then there are live campaigns, real enquiries and a count that does not match. The same is true of conversion tracking: it is a build decision wearing the costume of a marketing task.
What should you prepare before the session?
Bring evidence rather than opinions. Last year's enquiry numbers, the questions buyers actually ask, the pages that already convert, two or three rival sites with notes on what works, and a written sentence on what the business wants more of. An hour of preparation saves a fortnight of revisions.
- Enquiry history. How many came in, through which route, and how many turned into customers.
- The real questions. Twenty minutes of notes from whoever answers the phone.
- What already works. Your best performing page, email or advert, and your best guess at why.
- The platform position. What the site runs on now and whether that is staying, since a platform choice decides what you can publish later.
- The constraint. The date, the budget range or the launch event that is immovable.
What does a shallow discovery look like?
It produces a sitemap, a page count and a palette, and no agreed definition of success. The tell is that nobody in the room could state what the site is supposed to increase. The second tell is that every difficult question was recorded as something to confirm later.
Confirm later is where projects go to slow down. Each deferred decision becomes a blocker in build, and blockers arrive in the week when the design is signed off and the schedule has no slack left. A discovery session that ends with four open questions has not saved time, it has moved time to the most expensive part of the project.
What should you leave the workshop holding?
A one page document: the outcome the site is accountable for, the two priority visitor tasks, the traffic channels, the capture and routing rules, the measurement plan, the ownership list and the out of scope list. If it does not fit on a page, it has not been decided.
Write the measurement plan down in the same session, because that is the part everyone assumes is somebody else's. Four numbers is enough to start, and the method sits in our note on building a marketing measurement plan. Settle the account question in the same hour too: who owns your ad accounts is a cheap conversation before a build and an expensive one afterwards.
If you want the commercial half of discovery run properly before anyone opens a design file, that is the conversation we have at the start of every website engagement. Tell us what the site is meant to increase and we will tell you what has to be decided first.