Export every URL you have
From Search Console, analytics and a full crawl. Use all three, because each misses things the others catch. This single step prevents the most expensive redesign failure there is.
Redesigns fail in predictable ways and almost all of them are preventable in an afternoon of planning before anyone designs anything.
Before a redesign, export every indexed URL and every page's traffic, identify what is ranking and must be preserved, and build a redirect map. On launch day check redirects, robots directives, tracking and forms. Skipping the URL export is the single most common cause of a post-redesign traffic collapse.
From Search Console, analytics and a full crawl. Use all three, because each misses things the others catch. This single step prevents the most expensive redesign failure there is.
Which pages earn traffic, which earn enquiries, which have inbound links. Those must survive the redesign, even if they are untidy. Designers reasonably want to trim, and the untidy pages are often the earning ones.
Traffic, rankings, conversion rate and page speed before you change anything. Without a baseline you cannot tell afterwards whether the redesign helped or hurt, and everyone will have an opinion.
Conversion, speed, mobile, maintainability, or just that it looks dated. If the answer is only the last one, be careful, because redesigns chasing appearance sometimes reduce conversion.
Every existing URL mapped to a destination, built as a spreadsheet before development finishes rather than during launch week.
Full backup, and keep the old hosting live for at least a month. If something turns out to have been missed, having the original saves considerable time.
| Check | Why |
|---|---|
| Every redirect tested individually | Syntax being valid is not the same as working |
| No noindex left from staging | The classic silent disaster |
| robots.txt allows crawling | Blanket disallow blocks everything |
| Canonicals point at live URLs | Not at staging |
| Forms submit and deliver | Test from a phone, check spam |
| Analytics and conversions fire | Otherwise you are blind |
| Sitemap updated and submitted | So the new structure is discovered |
| Coverage watched daily for two weeks | Catch problems while they are small |
Launch mid-week, never Friday, so the people who can fix problems are available for the next few days.
| Ask them | Good answer | Walk away if |
|---|---|---|
| Who owns the site and accounts? | You do, in your own name | Accounts in the agency's name |
| Can I open three live examples? | URLs you can check yourself | Screenshots and a PDF |
| What is explicitly not included? | A written list with change pricing | Everything is included |
| How will we know it worked? | A metric agreed before work starts | Activity reports |
The exclusions question decides whether the final invoice matches the quote.
Not exporting existing URLs before starting. Pages that ranked return 404s on the new site and the traffic goes to a competitor. It is entirely preventable and it is the most expensive failure in this category.
It can, and it does not have to. Damage comes from missing redirects, removed content and unnecessary URL changes rather than from new visuals. With a complete redirect map and preserved content, rankings usually hold or improve.
If it earns traffic or has inbound links, yes. The untidy pages are frequently the earning ones, and cutting them for design reasons is the second most common cause of post-redesign traffic loss.
With correct redirects, expect two to six weeks of fluctuation then a return to previous levels. If traffic has not recovered after two months, the usual causes are missed redirects or content that was removed.
Mid-week, in the morning, when whoever can fix problems is available for the next few days. Friday launches mean any issue lives all weekend, and something always comes up.
Search Console coverage daily, landing page traffic against the same period before launch, and any page that dropped to zero, which almost always means a missed redirect.
Send us what you are trying to fix and we will tell you what it takes, what it costs, and whether we are the right people for it. If we are not, we will say so.