Here’s a thing that happens more often than it should. A business gets a new website. It’s better in every visible way. It’s faster, it looks current, everybody’s pleased with it.
And then the enquiries drop, and nobody can work out why, because the old site was worse.
The old site had something the new one doesn’t, and it usually isn’t anything you could see. It’s the years of being indexed, linked to and gradually trusted by Google, all of which is attached to things that quietly got thrown out during the rebuild.
What’s worth keeping
Your web addresses. Every page on your current site has an address. Some of those addresses are in Google’s index, in other people’s links, and in somebody’s bookmarks. If the new site puts the same content at a different address and doesn’t say where it went, all of that stops working at once.
This is the big one. It’s also the easiest to get right, and the easiest to forget.
The pages that already work. Usually two or three pages bring in most of the traffic, and they’re often not the ones you’d guess. A page somebody wrote in a hurry in 2019 answering one specific question can quietly be the best performing thing you own.
Content that looks like padding. Long pages get trimmed during redesigns because they look untidy next to a clean new layout. Sometimes the untidy detail was the reason the page ranked at all.
Everything around the site. Your Google Business Profile, your reviews, your analytics history, the email addresses on the domain. These aren’t part of the rebuild, so nobody thinks about them, right up until something goes wrong with one.
What to do before anybody starts designing
Take a list of every page on the current site. Not a mental note. An actual list.
Look at what your traffic does now, so that in three months you can tell whether the new site is doing better or worse. If you’ve never looked at your analytics, this is the moment, and if there aren’t any analytics, that’s worth fixing before the rebuild rather than after.
Note which pages bring people in, and make sure each one has somewhere to go on the new site. Not necessarily the same page. Just somewhere sensible.
That list, with each old address matched to a new one, is the whole trick. It has an unglamorous name, a redirect map, and it’s the difference between a rebuild that keeps its traffic and one that starts again from nothing.
The checkbox
One more, and it’s specific to WordPress.
There’s a setting that tells search engines to ignore the site. It exists for a good reason, because a site being built shouldn’t appear in Google half finished. It gets switched on at the start of most builds.
It also gets left on. A site launches, looks perfect, and is invisible to Google for a month because of a single tick box that nobody thought to untick. It has happened to better developers than me, and there’s no clever defence against it. It goes on a list, and the list gets worked through, because nobody’s memory is as good as they think it is at half past six on a launch day.
What I do on a rebuild
Redirects, first and every time. If a site is moving, every old address is pointed at its new home before anything goes live. I’ll come back to this properly in a piece of its own, because it’s the part most likely to cost somebody real money and it deserves more than a few paragraphs here.
Contact forms, actually tested. Not “the form submits and says thank you”. A real message, sent, and received at the address it’s supposed to reach. This is usually also where I set up proper email sending through SMTP rather than leaving it to whatever the server does by default, which is the arrangement most likely to deliver your enquiries straight into a spam folder, or into nothing at all.
Every device. Phones and tablets, and the enormous monitor somebody in the office insists on using. A layout that behaves on a laptop and comes apart on a phone isn’t finished, and the phone is where most of your visitors are.
None of that is glamorous. All of it is the difference between a launch and an incident.
If somebody else is building it
If you’re about to commission a rebuild from another developer, ask them one question: what are you doing about redirects?
It’s fair, it takes ten seconds, and the answer tells you a great deal. You’re listening for something specific and slightly boring. If the answer arrives quickly and mentions mapping the old addresses to the new ones, good. If it takes a while, and contains the word “basically”, make a note.