Website redesign
The site is dated, slow, and every small change turns into a ticket.
A redesign is usually sold as a new look. The expensive part is everything underneath — the URLs, the content, the search traffic you already earned, and whether your team can run the thing afterwards.
From $8,000 · per project
Signs
You know it is time when
- Every content change needs a developer, so simple edits wait weeks.
- The platform is out of support, or the licence keeps climbing and you cannot leave.
- It is slow on a phone, and you suspect that is costing you.
- It is not accessible, and that has stopped being theoretical.
- A previous redesign lost rankings or broke links, and nobody wants to repeat it.
- Nobody can find anything — including your own staff.
The risk
The most expensive redesign mistake is invisible for a month
A site launches, everyone is pleased with the design, and eight weeks later enquiries are down by a third. What happened is nearly always the same: URLs changed, redirects were partial, pages that earned traffic for years stopped existing, and nobody was watching the one number that would have shown it on day three.
So redirects and content parity are not a launch-week checklist item here. They are the work:
- Every existing URL inventoried and mapped to its replacement before the new site is built, not after.
- The pages that actually earn traffic identified first — a redesign that reorganises those without knowing they matter is where the loss comes from.
- Redirects tested as a list, not spot-checked by clicking around.
- Search performance watched for the weeks after launch, when a problem is still cheap to fix.
Method
From $8,000
How the work runs
- Audit. What you have, what earns its keep, what is broken, and what is worth carrying over. You get this as a document whether or not you continue.
- Map. URLs, redirects, content model, and the templates that will replace the pages. The decisions get made here, where changing your mind is free.
- Build. A fast, accessible site on a platform your team can maintain — with accessibility tested during the build, not discovered afterwards.
- Migrate and launch. Content moved, redirects live, search performance watched daily through the window where it matters.
- Hand over. You own the accounts, the repository, the domain and the knowledge. Documented, with a working session for whoever inherits it.
From $8,000. Fixed price once scoped — you should not be paying by the hour for my estimate being wrong.
Handover
Handover is the deliverable, not a courtesy
The most valuable thing I have seen a large organization do with its website was to stop outsourcing it. Not because agencies are bad, but because someone inside who cares gets wiser every cycle, while an external team starts from zero on every engagement and takes the knowledge with them when the contract ends.
So this work is built to be left. If you want me to keep running it afterwards, care plans exist — but that should be a choice you make, not a dependency you discover.
Limits
When a redesign is the wrong call
- A brand-new site with no history. That is simpler and cheaper — it is a build, not a redesign.
- Anyone who wants only a new look. If the structure and content stay as they are, you are buying a paint job and should pay paint-job prices.
- A budget under about $8,000. Below that I cannot do the migration properly, and doing it improperly is how sites lose traffic.
- Design-by-committee with no decision-maker. I will ask who signs off in the first conversation.
Evidence
Rebuilds like yours
Open one and check the things a redesign is supposed to get right — speed on a phone, navigation, whether the content is findable.
The estate moved onto modern rails without losing URLs or search visibility, and accessibility became something checked on every build.
Several thousand public pages, content owners across dozens of departments, a statutory accessibility obligation, a small team, and no option to take the site down while you fix something.
A bought-in product replaced by something the college owns — the same functionality, run by the same staff, now with no external cost and full control.
Orientation ran on a purchased platform. Staff already updated the scenarios themselves; what the subscription bought was the functionality underneath — renewed on the vendor's cycle, on a product the college did not control.
More leads than they could process — the owner asked to switch ads off for some services while the team caught up. And the site's simple structure has since become the standard competitors copy.
Licensing services are bought through a Google search, and every competitor bids on the same clicks. Whether the ad spend comes back as leads is decided by the page the click lands on.
People who would have picked up the phone started making contact through the site instead — and the build has since moved off a dying stack without losing any of that.
An established provider whose site was not being found by the people already searching for what they sold.
Shelter House
shelterhouse.on.ca (opens in a new window)
Emergency shelter and community services, Thunder Bay
Help is one tap from the homepage, and the site has kept up with them since.
People arrive at a site like this in a crisis, often on an old phone, needing one specific answer immediately. The site also has to hold up in front of funders.
A migration has a mapped URL inventory rather than a memory of one, and handover is a document rather than a phone call.
Most agency delivery lives in someone's head. Decisions get made in a call and never written down, the reason for a choice is lost by the time it matters, and the next person — including the client — inherits a site with no record of why it is the way it is.
Questions
Before you commit
Will we lose our search rankings?
Not if the migration is done properly, and that is most of what you are paying for. The honest version: expect a wobble for a couple of weeks while the new URLs are recrawled. What you should not see is a sustained drop, and if one starts, we will know early because it is being watched.
How long does it take?
Typically two to four months for a content-heavy site. The build is rarely the slow part — content decisions and internal approvals are. I will tell you at the start which parts depend on you, because those are the ones that move the date.
Can our team update it afterwards?
Yes, and that is a design constraint from the beginning rather than a feature added at the end. If your team cannot edit a page without me, I have built the wrong thing.
What about our old PDFs?
They usually come along, and they are usually inaccessible. Worth deciding deliberately: convert the important ones to real pages, remediate what must stay a PDF, and retire the rest. See AODA compliance for why this matters.
Do you work with our existing designer?
Happily. I would rather build someone else's good design than insist on owning that part.
What happens to our existing content?
It gets inventoried before anything is designed. Most sites carry a third more pages than anyone realises, and a meaningful share earn nothing and should be retired rather than migrated. You approve that list — I will recommend, but nothing gets deleted on my say-so.
Can we do it in phases?
Usually yes, and for large sites it is often the sane answer — move the templates that matter first, run the two systems briefly, then retire the old one. It costs a little more in total and removes the single biggest risk, which is a hard launch date nobody can move.
Start with the audit
Send me the URL and what is driving the timing — a platform going out of support, a compliance date, a rebrand, or just years of accumulated frustration. You will hear scope and whether a full redesign is even the right call.
Start with the URL
Send the address and what is driving the timing — a platform going out of support, a compliance date, a rebrand, or just years of accumulated frustration.
Goes straight to me. No list, no automated sequence — see the privacy policy.