Web Design
Website redesign without losing what works
A redesign should make your site faster, more credible and easier to use, without throwing away the rankings and content that already work. We audit first, keep what earns, and rebuild the rest.
Who it is for: Businesses whose site looks dated, loads slowly or no longer matches what they sell, but who cannot afford to lose existing search traffic.
Everything in this service
- Audit of current design, content and search performance
- Content mapping and full redirect plan for every URL
- New design built mobile-first on current best practice
- Speed and Core Web Vitals as launch requirements
- Pre-launch and post-launch ranking checks
What to expect
- A modern site your team is proud to send people to
- Rankings protected through the switch
- Faster pages and clearer paths to enquiry
How website redesign actually works
Audit what already works before you change anything
The riskiest thing about a redesign is that it usually starts from what the team dislikes rather than from what the site earns. Someone is tired of the look, and a project begins. Meanwhile a handful of pages are quietly bringing in most of the traffic and most of the enquiries, and nobody has identified them.
So the first job is an inventory. Pull every URL, then attach the numbers to each one: organic entries, conversions, backlinks, and which queries the page ranks for. That produces four groups. Pages that earn and must be protected, pages with links pointing at them that must keep their address, pages that get traffic but do not convert, and pages nobody visits. Only the last group is genuinely safe to delete.
This is also the moment to record why the current site performs the way it does. A page that ranks well often does so because of its depth, its specific wording or its internal links. A redesign that shortens it for visual balance can lose the ranking without anyone realising which change caused it. Write down what must survive before the design work starts, because once new layouts exist, content gets trimmed to fit them.
- Export every URL and attach traffic, conversions and link data
- Identify the small set of pages carrying most of the value
- Note which pages have external links pointing at them
- Record what makes a top page work before anyone redesigns it
- Only delete pages that have no traffic, no links and no conversions
Protecting rankings: URL mapping and redirects done properly
Most traffic losses after a redesign trace back to one thing. URLs changed and the old addresses were not pointed at the right new ones. Search engines and every external link still refer to the old addresses, and if those return a not found page, the accumulated value is discarded.
The safe default is not to change URLs at all. Changing them buys very little and risks a lot, so keep the existing addresses wherever the page still exists in some form. Where a change is genuinely necessary, build a mapping spreadsheet with every old URL and its single best new destination. The destination must be the closest equivalent page, not the homepage. Bulk redirecting a set of retired pages to the homepage is treated as a soft error and loses the value you were trying to keep.
Use permanent redirects, avoid chains where one redirect points at another that points at a third, and check the whole map on staging before launch rather than after. Remember the parts people forget: image files, PDFs, old campaign pages, the XML sitemap, canonical tags, and any hardcoded internal links inside content. Internal links should be updated to point straight at the new address rather than relying on the redirect, because a site running every internal link through a redirect is slower and harder to debug.
- Keep existing URLs unless there is a concrete reason to change them
- Map every old URL to its closest equivalent, never bulk to the homepage
- Use permanent redirects and eliminate redirect chains
- Include images, PDFs and old campaign pages in the map
- Update internal links to the new addresses rather than leaning on redirects
Content migration is where redesigns quietly lose value
Content migration gets treated as a copy and paste job, and it is where a surprising amount of damage happens. New templates have different content slots. Copy gets shortened to fit a layout, a section is dropped because there is nowhere obvious to put it, and a page that answered eight buyer questions now answers three.
The way to avoid it is to treat content as the input to design rather than the thing you pour into a finished design. Decide what each important page must say, then build a template that fits it. When it goes the other way, the template wins every argument and the page loses substance.
Practical checks during migration: headings should keep their meaning and their hierarchy, page titles and descriptions should be carried across deliberately rather than regenerated as defaults, structured data should be reapplied, and image alt text should survive. Alt text is lost more often than anything else, because images get re-uploaded through a new media library. Run a crawl of staging against a crawl of the live site and compare titles, headings and word counts page by page. The pages where word count dropped sharply are the ones to review by hand.
- Design templates around the content, not content around the templates
- Carry titles, descriptions and heading structure across on purpose
- Reapply structured data and check it validates on staging
- Preserve image alt text, it is the most commonly lost element
- Crawl staging and live, then compare word counts to spot silent losses
Phased launch or big bang, and how to choose
There are two honest approaches. Replace everything at once, or replace in phases starting with the pages that matter most. Both work. The wrong one for your situation is the one that fails expensively.
A single launch suits smaller sites, a full platform change, and any case where a hybrid old and new site would be visibly incoherent. It is faster and cheaper to coordinate. The cost is concentrated risk: everything changes on one day, so if traffic drops you have dozens of possible causes and no easy way to isolate one.
A phased launch suits large content sites, sites carrying a lot of search traffic, and organisations that cannot absorb a bad week. Templates are replaced in groups, each phase is watched for a few weeks, and problems are caught while the sample is small enough to diagnose. The costs are real: a period where two designs coexist, more coordination, and a longer overall timeline. A useful middle path is to launch the new design across the site but leave the highest-earning pages structurally alone in the first phase, so the biggest earners change last and under observation.
- Single launch: faster and cheaper, but risk lands all at once
- Phased launch: slower, but problems appear while they are small
- Platform migrations usually force a single cutover
- High-traffic content sites usually justify phasing
- Either way, launch mid-week and never before a holiday or a quiet week
Launch day and the monitoring window that follows
The first two weeks after launch decide whether a redesign holds. Problems in that window are usually cheap to fix and expensive to miss, because a page that has been returning errors for a month behaves differently from one fixed within a day.
Launch day itself has a fixed checklist. Confirm the staging site is blocked from indexing and the live site is not, because a stray directive blocking crawlers is the single most damaging launch error and it is entirely silent. Submit the new sitemap. Spot check redirects from a sample of important old URLs. Verify analytics and conversion tracking fire, since tracking codes are lost in migrations constantly and you will not notice from the traffic graph alone. Test the forms by actually submitting them and confirming the email arrives.
Then watch. Errors and crawl issues daily for the first week. Rankings for your priority queries weekly. Traffic compared against the same period last year rather than last month, so seasonality does not mislead you. Expect a small dip in the first fortnight as engines re-crawl and re-evaluate. What is normal is a shallow dip that recovers. What is not normal is a sharp drop on a specific set of pages, which almost always points at a redirect, a blocked resource or lost content on those pages.
- Confirm the live site is crawlable and staging is not, on day one
- Submit the new sitemap and check a sample of redirects by hand
- Test analytics, conversion tracking and every form by using them
- Watch errors daily for a week, rankings weekly for a month
- A shallow dip is normal, a sharp page-specific drop is a bug to find
Judging whether the redesign worked
A redesign should be measured against the goals set at the start, and those goals need to be written down before anyone sees a mock-up. Otherwise the assessment collapses into whether people like the look, which is the least reliable measure available.
Useful measures depend on the goal. If the goal was credibility and enquiry volume, watch enquiries per thousand visits rather than raw enquiries, because traffic moves for unrelated reasons. If the goal was speed, compare field data from real users before and after rather than a lab score. If the goal was to make content easier to find, look at whether more visits reach the pages that matter and whether internal search use falls.
Allow enough time. Search performance typically takes several weeks to settle after significant structural change, so a judgement made in week two is noise. Set a review at the one month and three month marks, compare like for like periods, and be willing to reverse specific changes. Reverting one page template that lost ground is a normal part of the process, not an admission of failure.
- Agree the measures before design starts, not after launch
- Use conversion rate rather than raw enquiry counts
- Compare against the same period last year to control for seasonality
- Wait a month before drawing conclusions on search performance
- Be ready to roll back a specific template that clearly lost ground
A clear path, step by step
- 01
Audit
What works, what does not, and which pages carry your traffic. Nothing valuable gets discarded.
- 02
Plan the migration
Every URL mapped, every redirect planned, content kept or improved page by page.
- 03
Redesign and rebuild
New design and build, tested on mobile first and against speed budgets.
- 04
Launch and verify
Staged launch, then we watch rankings and traffic closely and fix anything that moves.
Why choose us for this
We treat a redesign as a migration, not a repaint
Search performance is planned before pixels are drawn
Honest about which pages should stay untouched
Common questions
Will a redesign hurt my Google rankings?
It can if URLs change without redirects or content is thrown away. Done properly, with mapping, redirects and content preserved, rankings hold and usually improve as speed and structure get better.
How do I know if I need a redesign or just fixes?
If the structure and platform are sound, targeted fixes to speed, copy and key pages often deliver most of the value. We audit first and tell you which, with costs for both.
Can you redesign in stages?
Yes. Highest-impact pages first, the rest in planned phases. It spreads cost and lets each phase prove itself before the next.
Explore related work
Want this for your business?
Book a free visibility call and I will tell you honestly whether I can help.
How this is delivered
One person leads every project. Where a job genuinely needs a specialist, I bring in people I have worked with before and manage them, so you get one point of contact and one invoice rather than three suppliers blaming each other.
- You talk to the person responsible for the work, not an account manager
- Specialists are briefed and managed by me, and their work is checked before it reaches you
- One contract, one invoice, one place to chase