Websites get sold as projects because projects are easy to sell. Fixed scope, fixed price, a launch date to aim at and a clear moment where everyone shakes hands and moves on. Clients like it because they can budget for it. Agencies like it because they can resource it. It is a tidy arrangement, and most of our industry is built around it.
The trouble is that the handshake happens at the wrong end. Launch day is not the finish line. It is the first day the thing is actually in use, and almost everything that determines whether a website earns its keep happens afterwards, in small increments, over years, long after the project that created it has been closed off and forgotten.
Which is roughly how buildings work. You commission one, you build it, and then you occupy it for the next twenty years, and the occupying is the entire point. The difference is that buildings come with someone whose job is to look after them. Websites usually do not.
What twenty years actually looks like
We have clients we have been looking after since the early 2000s. Two decades is long enough for a website to stop being a thing you built and start being a piece of infrastructure the business runs on.
In that time, a site will typically have been redesigned three or four times. It will have moved hosts, changed platforms, absorbed a rebrand, gained an online store and lost a section nobody could remember the purpose of. It will have survived a merger, a name change and at least one very memorable week.
More importantly, it will have been through a lot of people. The marketing manager who commissioned the original site has moved on. So has their replacement, and their replacement’s replacement. The general manager who insisted on the photo of the old building has retired. The staff member who quietly maintained the events calendar for nine years has left, and nobody is entirely sure what she used to do on the first Monday of the month.
The organisation persists. The knowledge does not. It walks out the door in exit interviews and half-finished handover documents.
The knowledge that never made it into the website
This is the part people underestimate. The valuable thing is not the code. Code can be rewritten, and eventually will be. The valuable thing is the accumulated understanding of why the site is the way it is.
Why that enquiry form sends to two inboxes instead of one, and what happened the year it only sent to one. Why a particular page has a redirect pointing at it from an address nobody uses any more, and the printed brochure from 2019 that is the reason it has to stay. Why the team never publishes anything on a Friday afternoon. Which product name is legally sensitive and which one just annoys the sales team. How the booking system talks to the finance system, and the one field that must never be renamed.
None of that is written on the website. Very little of it is written anywhere. It lives with whoever has been paying attention the longest.
When we say we have been the constant, that is what we mean. Not that we have been sitting there doing nothing for twenty years, but that we have been the continuous thread of memory across four marketing managers, three redesigns and one very complicated migration. A new marketing manager can start on a Monday and by Wednesday have someone on the phone who can explain how their own website works and, more usefully, why.
Why the project mindset gets expensive
Treat a website as a series of unrelated projects and you pay for the same knowledge over and over.
Every new supplier starts from zero. They do not know the business, they do not know the history, and they have no reason to trust anything they find in the existing site. So they do the rational thing and propose a rebuild, because rebuilding is genuinely faster than understanding. Three years later, the next supplier looks at their work and reaches exactly the same conclusion.
Meanwhile the things that quietly matter get lost in each transition. Redirects break, so years of accumulated search rankings evaporate. Integrations get rebuilt slightly differently and reports stop reconciling. A form that fed a database now just sends an email that nobody actions. None of it is dramatic. It is just erosion, and it costs real money.
Buildings avoid this because we accept that maintenance is normal. Nobody demolishes an office block because the carpet is dated. They replace the carpet, and the structure carries on.
What stewardship actually looks like
It is less exciting than a rebuild, which is probably why it gets sold so rarely.
It looks like knowing the site well enough to notice when something is off before the client does. Keeping a written record of decisions, so the reasoning survives the people. Doing the unglamorous upkeep on a schedule rather than in a panic. Making small improvements continuously, so the site never drifts far enough from current that only a full replacement will do. Being available to answer a question that takes ninety seconds, without turning it into a scoped piece of work.
And when a redesign genuinely is the right call, and sometimes it is, doing it as a renovation rather than a demolition. Keeping what works. Carrying the history across. Making sure the new site inherits the twenty years of hard-won knowledge instead of starting the clock again.
The point
A website is not a thing you finish. It is a thing you occupy. It has an ongoing relationship with the people who use it, the people who update it and the business it represents, and that relationship needs someone tending it.
Most organisations already understand this instinctively about their physical premises. They have a maintenance budget, a cleaner, someone to call when the heating fails. The website, which quite often does more business than the front door, gets nothing of the sort.
If you have inherited a site and you are not sure who has the memory of it, that is worth resolving before it is worth redesigning. It is the kind of thing our care plans exist for, and it is also just a good question to ask, whoever you end up asking it of.
