What Happens After Launch
Launch feels like the finish line and it is closer to the halfway point. Not because anything is wrong — because a website is a thing you own now, and owned things need attention.
Most of what follows takes minutes. The reason it goes wrong is that nobody said it out loud at handover. For teams that want to compare patterns across work rather than rely on impressions, workforce analytics software provides a more measurable view.
Week one
Check it on your own phone, on mobile data, not on your wifi. Then have two other people do the same on different phones. This catches more than any testing process, because real devices in real conditions are where sites actually break.
Send a test through every form. Then check it arrived, and check the spam folder. Contact forms failing silently is the single most expensive small fault there is: you never find out, and the enquiries simply stop.
Buy something, if you sell something. All the way through, with a real card, then refund yourself. Test the confirmation email, the receipt, and what a customer sees when a payment fails.
Read every page out loud. You will find typos your eye skipped past for weeks. For an independent reference beyond this site, Google Search Central is a useful place to compare approaches.
Confirm you can log in to everything — domain registrar, hosting, the site itself, analytics, payments. If you cannot, resolve it now while the developer is still engaged.
Month one
Tell search engines the site exists. Submit the sitemap in Google Search Console and Bing Webmaster Tools. Free, ten minutes, and it is how you find out about problems later.
Update every profile that points at your old site or at a platform: social bios, business listings, email signature, invoices, printed material.
Check what is being indexed. Search for your site and see what comes up. Staging pages left visible, or a live site accidentally blocked from indexing, both happen and both are quick to fix if noticed early.
Look at the first analytics. Not to optimise anything — to check the tracking works and that pages you expected to be visited are being visited.
Write down where everything is. One document: registrar, host, logins, who to call, when things renew. Store it where someone other than you could find it.
Every year
Renewals. Domain and hosting, and set a calendar reminder independent of any auto-renewal, because cards expire. A lapsed domain is the one failure that takes your address away, and it usually happens because a card expired, not because anyone decided anything.
Check the backups exist and restore. A backup nobody has restored is a hope, not a backup.
Review the content. Prices, services, team, the year in the footer. Out-of-date information does more quiet damage than an old design.
Run a speed test. Sites get slower over time as images and plugins accumulate. It is worth knowing before a visitor does.
What actually breaks, in practice
Forms stop sending. Hosting changes, a plugin update, an email provider tightening rules. Silent, and it stops your enquiries. Test quarterly.
Certificates expire, and browsers show a security warning. Usually automatic, occasionally not.
Plugins update and something moves. Which is why updates need a policy rather than a habit — and why the plugin list should be short.
Images pile up. Every uploaded photo stays, at full size, forever.
Somebody edits a page and breaks the layout. Ask to be shown how to edit safely at handover, and which parts not to touch.
What you do not need to do
Redesign every two years. A site that works does not need replacing because it is three years old.
Blog because you were told to. Publishing weakly and irregularly is worse than not publishing, because an abandoned news page dated eighteen months ago is a signal about the business.
Chase every plugin update the day it appears. Security updates promptly, feature updates when you have time to check.
Watch analytics daily. Monthly is plenty, and daily numbers on a small site are noise.
The honest bit about support plans
Most developers, myself included, offer ongoing maintenance. It is worth knowing when it is worth paying for.
Worth it if your site takes payments, if downtime costs you money, if you have no interest in learning any of this, or if the site is complex enough that a broken update is not obvious.
Not worth it for a small brochure site on decent managed hosting where you are comfortable clicking update and can wait a day if something breaks.
Ask what a plan actually includes. "Maintenance" ranges from a monthly automated report to someone genuinely watching your site. Those are different products at similar prices.
The short version
- Launch is the middle of the project; the site is a thing you own now
- Week one: test on real phones, send a test through every form, buy something yourself
- Month one: submit the sitemap, update every profile pointing elsewhere, write down where everything lives
- Every year: renewals, restore a backup, review content, run a speed test
- Forms failing silently is the most expensive small fault — test quarterly
- Support plans are worth it if downtime costs money; ask exactly what one includes