
The Cost to Maintain a Website Per Year, Over Three Years
A site is not a purchase, it is a subscription to its own continued existence. Here is every line that shows up between launch and year three.
Author

Founder & CEO
Burhan has spent ten years on both sides of a software project: writing the code, then running the delivery, then sitting across the table from the person paying for it. He founded Devfinix to close the gap he kept seeing between agencies that could build software and agencies that could market it, and the clients who paid twice because nobody did both.
Most of his week goes on the parts of agency work that decide whether a project succeeds before a line of code is written — scoping, estimating, choosing a contract model, and being straight with a client about what their budget will and will not buy.
On this blog he writes about estimation, delivery and the commercial side of building software: why two quotes for the same brief differ by four times, what a discovery phase is actually for, and when the honest answer is that the work should not go ahead.
Writes about

A site is not a purchase, it is a subscription to its own continued existence. Here is every line that shows up between launch and year three.

Refusing discovery does not save budget. It moves the cost to the phase of the project where changing your mind is most expensive.

A fixed price is not a safer price. It is a price with a risk premium and a change-request tax already inside it. Here is how to pick a model you can live with.

There are only four real options for an ageing system, and the one most teams reach for is the one that goes wrong most. Here are the criteria for each, including doing nothing.

The hourly rate is a price, not a cost. Rework, timezone overlap, seniority mix and management burden decide what you actually pay per shipped feature.

Eight questions that actually separate one agency from another, written by someone who answers them for a living, with the replies that should end a conversation.

Some technical debt is a sound business decision and some is negligence. Here is how to tell which one your team is carrying, and what to ask for so it stops compounding quietly.

Most accounts stop measuring at 'form submitted', so the bidding learns to buy cheap rubbish. Closing the loop to the CRM changes what the algorithm optimises for.

Two agencies price one brief very differently, and it is almost never the rate. It is what each of them put in the quote and what they left out.

The spread between app quotes is almost never greed. It is scope, the definition of done, team shape and who ends up carrying the risk.