Two days is enough to launch a small business site. Not a rushed one — a properly finished one, with real content, working forms, decent performance, and a client who knows how to use it.
What makes it possible is not typing faster. It is sequencing: doing things in an order that prevents rework, and refusing to start the clock until the conditions that actually cause overruns are handled.
One clarification before the workflow. This is 48 working hours, not 48 elapsed hours. Spread it across a week if you like. The sequence is the point.
Hour zero: the gate
Almost every blown timeline traces back to something missing at the start. Before you begin, four things must be true. If any one is false, you do not have a two-day project — you have a two-week project with two days of design in it.
1. The content exists. Not “the client will send copy.” Actual text for every page, in a document, in your hands. Client content is the single largest cause of stalled builds, and no amount of speed on your end fixes it.
2. There is one decision-maker. Named, available, and empowered to approve. Two stakeholders with different tastes will cost you more time than the entire build.
3. Access is sorted. Hosting credentials, domain registrar, DNS access, email accounts, any existing analytics. Chasing a login on day two is a half-day gone.
4. Scope is written down. Page count, which forms, which integrations, how many rounds of revision. Write it in the proposal in plain language, including what is not included.
If content is the blocker — and it usually is — offer a paid copywriting add-on or supply a structured template with word counts per section. Either is better than waiting.
Hours 1–4: Decide, don’t explore
The trap here is browsing themes. You can lose a full day to a marketplace and arrive back where you started.
Write down your requirements first, in one short list: number of page types, whether it needs a shop or a booking flow, whether the client will edit it themselves, and any non-negotiable integration. Then shortlist three themes maximum, spend twenty minutes each, and pick one.
While evaluating, check the things that hurt later rather than the things that look good now: is there a documented override path so your customisations survive updates, is the changelog recent, and does the demo hold up on a phone. A theme that looks 10% better but fights you at every customisation is not the faster choice.
Then, before you go further: install it clean, and set up your child theme or override structure immediately. Doing this at the start costs fifteen minutes. Doing it retroactively costs an afternoon.
End of block: theme chosen, installed, override structure in place, staging site running.
Hours 5–12: Structure and content, before design
This is the block that most freelancers get backwards, and it is where the time is genuinely saved.
Build every page, in order, and pour the real content in. No design work yet. Do not open the colour settings. You are constructing the skeleton: navigation, page hierarchy, URL structure, headings, body copy, images placed at roughly the right sizes.
The reason is simple. Design decisions made against placeholder text are guesses, and half of them will be wrong once real content arrives — a headline that runs to three lines, a services list with seven items instead of the demo’s three, a testimonial four times longer than the demo one. Designing first means designing twice.
Set up redirects now too, if you are replacing an existing site. It is easy to forget under launch pressure and expensive in lost search traffic.
End of block: every page exists, with real content, looking plain.
Hours 13–24: The brand pass
Now design — and now it is fast, because you are making decisions against real content rather than imagining it.
Work in this order, because each step constrains the next:
- Typography. One display face, one body face. This transforms the site more than anything else you will do.
- Colour. One dominant neutral, one supporting neutral, one accent reserved strictly for actions.
- Spacing. Set the rhythm — section padding, card padding, heading offsets. This is where a template stops feeling like a template.
- Shape language. One corner radius, one border weight, one shadow depth, applied everywhere.
- Imagery. Client photos where they exist. Where they do not, one consistent treatment across all stock images.
Then delete. Every demo section you could not fill genuinely — the thin testimonial strip, the stats counter with invented numbers, the team grid with two people — comes out. Clients often want these because the demo had them. A shorter, denser page is better work and a faster site.
End of block: the site looks like the client’s, not the theme author’s.
Hours 25–36: The pages nobody demos
This block is what separates a professional handover from an amateur one, because it covers everything the client will discover in month two and blame you for.
Work through:
- The 404 page. Rewritten, on-brand, with a route back into the site.
- Search results, including zero results.
- Form states. Success message, validation errors, spam protection, and — critically — confirm the notification email actually arrives. Test it to an address on the client’s own domain.
- The legal pages. Privacy policy, terms, cookie notice, whatever applies locally.
- Metadata. Page titles, descriptions, and the social share preview. Post the URL into a private chat and look at what appears; that card is the site’s first impression on every share.
- Favicon. Still the theme author’s logo on more launched sites than anyone would guess.
- Mobile pass. Every page, on a real device, using your thumb.
- Performance check. Run the key pages on mobile. Compress oversized images, remove theme modules you are not using, and audit what third-party scripts the client has asked you to add.
- Analytics and search console, verified and reporting.
End of block: the unglamorous half is done, which is the half clients never see and always feel.
Hours 37–48: Handover
Reserve real time for this. Rushed handovers generate months of unpaid support requests.
- Record a screen walkthrough, ten to fifteen minutes, showing exactly how to do the three things this client will actually need: edit page text, add a post or product, and update the contact details. Send the video file, not just a live session — they will rewatch it.
- Write a one-page document listing where everything lives: hosting, registrar, theme licence and its renewal date, plugin licences, and who to contact for what.
- Transfer ownership properly. Accounts in the client’s name, with you added as a collaborator. Do not leave the business dependent on your personal logins.
- State what happens next. Maintenance terms, update responsibility, hourly rate for changes outside scope, and how support requests should reach you. Being explicit here is the difference between a maintenance retainer and an endless trickle of free favours.
Two things worth saying plainly
On charging for it. Some freelancers feel awkward billing a real fee for a two-day build. Don’t. The client is not buying your hours; they are buying the outcome and the judgment that produced it — which theme, which sections to cut, what to test, what to hand over. That judgment is why it took two days instead of two months. Price the result.
On when 48 hours is the wrong promise. Custom functionality, a data migration from an old platform, a large catalogue, multi-language, or a client who has not written the content yet. Say so before signing rather than after. Fast is a service; overcommitting is a liability.
At eThemeStudio, we build for this workflow specifically: documented override paths, exposed design tokens, recent changelogs, and demos that survive real content. If you are a freelancer or small studio launching regularly, get in touch about our extended licence — and tell us what slows you down most. It shapes what we build next.
Comments (0)