What should actually happen in your first six weeks with an agency

Six weeks into working with a good agency, you should have something live or very close to it, and you shouldn't have been surprised once along the way. You should have clicked through the product before it was built, seen working software every week, and know exactly what you're getting at handover.

Here's what that looks like week by week, what you should be doing on your side, and the warning signs that things are drifting. Smaller projects move faster than this and bigger platforms take longer, but the shape holds.

Before week one

A lot of lost time happens before any work starts. Have these ready for kickoff:

• One decision-maker. Feedback from a committee, with nobody who can break a tie, is a reliable way to slow a project down.
• Access. Your domain registrar, hosting, analytics, social accounts, brand files, and any existing codebase. Chasing logins in week two is a waste of everyone's week two.
• A goal you can measure. "More leads from the website" is a start. "Twice as many qualified enquiries a month" gives everyone something to design for.

Week 1: discovery and a pilot sprint

The first week should be mostly listening. Expect workshops with the people who know your customers, a review of whatever exists today, and an agreement on scope and success metrics written in plain language. By the end of the week you should have a roadmap, a named team and a timeline you can hold people to.

We like to start with a short pilot sprint: a small, real piece of work with a clear deliverable. It lets you see how the agency communicates, how quickly they turn feedback around and whether their work matches what they promised in the pitch, before you commit to the whole engagement. It's a low-risk way for both sides to find out whether the fit is right, and it's how most of our own engagements begin.

Week 2: a prototype you can click

By the end of the second week you should be clicking through something. Not a PDF of static screens, but a prototype where you can tap a button and land on the next screen, so the main flows feel real.

This is the cheapest moment to change your mind. Moving a button in a prototype costs minutes. Moving it after it's built costs days. Put the prototype in front of a few real customers if you can, and watch where they hesitate. Their confusion is more useful than any opinion in the room, including yours.

If week two ends with a slide deck about "the design direction" and nothing to click, ask why.

Weeks 3 and 4: building, with a demo every week

This is where the product gets built, in weekly sprints. The ritual that matters most is the weekly demo: working software on a staging link, shown live, with time for questions.

A good demo shows real screens with real, or realistic, data. It covers what shipped this week, what's planned for next week and anything that's blocked. It's also where you say "that's not quite right" while it's still cheap to fix.

You should be able to open the staging site yourself between demos. QA should be happening on real phones and browsers, not only on the developer's laptop. And someone should be keeping the documentation up as they go, not saving it all for the end.

Week 5: analytics, content and a proper QA pass

Analytics goes in before launch, not after. If you launch without tracking, the first weeks, often the most informative ones, are gone for good. At minimum, set up:

• Page and screen views.
• Conversion events for whatever the goal from week one was: form submissions, bookings, purchases, sign-ups.
• A simple dashboard showing those numbers, which you can open without asking anyone.

This is also the week for loading final content, checking speed on mobile, fixing accessibility issues, and running through every form, payment and email with test data. If there's an SEO side to the project, titles, descriptions, structured data and redirects get checked now too.

Week 6: launch and handover

Launch should be dull. Deploy, check everything works in production, watch the monitoring and fix small issues quickly. The agency should be on hand for launch support, not moving on to the next client the moment the site is live.

Handover matters as much as launch. Before the project closes, you should have:

• Every account in your name: domain, hosting, analytics, app store listings, third-party tools. If the agency set them up under their own logins, move them now.
• Access to the code repository.
• Documentation that explains how the system is put together, how to deploy it, how your team edits content, and who to call when something breaks.
• A short list of known issues and next priorities, so nothing lives only in someone's head.

What six weeks looked like for Resync Spaces

Resync Spaces went from kickoff to deployment in six weeks. In that time we built a lead management platform with a custom lead qualification calculator, a real-time admin dashboard for tracking prospects, automated lead nurturing workflows and a design system built for fast iteration. The platform has since seen its lead conversion rate rise 214%, with leads converting 3.2 times faster, and it has scaled smoothly as lead volume grew.

The design system is worth a mention. It's a set of reusable components and rules, built early, so new screens and later changes are assembled from parts that already exist instead of being designed from scratch. On Resync Spaces, that was the whole point of it: fast iteration. The full story is in the Resync Spaces case study.

Warning signs

Some things should make you ask questions early:

• Nothing to click by the end of week two.
• Demos made of slides instead of working software.
• No staging link you can open yourself.
• Scope changes that arrive as surprise invoices, with no discussion first.
• Analytics described as "something we'll add after launch."
• Accounts registered in the agency's name with no plan to transfer them.
• Handover amounting to a zip file and a goodbye.

None of these mean the agency is bad. They do mean you should raise it now, while there's still time to correct course.

Your side of the deal

An agency can only move as fast as its feedback. The most useful things you can do are to turn feedback around within a day or two, keep one person as the final decision-maker, show up to the weekly demo, and tell the team early when priorities change. Delays tend to come from lots of small waits stacking up, rather than one big problem.

If you'd like to talk through your own first six weeks, you can book a free strategy session on our start a project page, or see what our web design & development team builds.