No two projects cost the same, and any agency quoting you before understanding the scope is guessing. Here is what actually moves the number — so you can judge a quote instead of comparing them blindly.
Ten pages built from three templates cost far less than ten pages each designed on their own. What drives the price is the number of distinct layouts, not the page count.
Connecting a CRM, an ERP, a payment provider or a stock system is where budgets move most. Each connection carries its own edge cases, and those are the hours nobody estimates well.
A redesign that keeps its SEO needs a URL map, redirects and structured data checked before go-live. Skipping it is cheaper on the quote and far more expensive three months later.
Content is the most common reason a project slips. Editorial production, translation and photography are real line items — and a site waiting on text costs you in delay, not just in budget.
The fastest way to lower a quote is to ship less, sooner, and add what real usage proves you need. It also lowers the risk of paying for features nobody opens.
A handful of layouts, an editable CMS, sound SEO foundations and a contact path that works. This is the majority of projects, and the range is wide because "a few pages" covers very different things.
Catalogue, product pages, checkout and the plumbing behind them — payment, shipping rules, tax, stock. The catalogue size matters less than how many systems have to agree with each other.
A tool built around your own process, or a product you sell by subscription. Here the price follows the number of rules to encode and the cost of getting one of them wrong.
Thirty minutes on what the site has to achieve and what already exists. No quote is sent from this call alone — it exists to tell us whether the project is even the right one.
On anything but the simplest site, we scope before quoting: pages, integrations, content, migration. It is a paid, bounded deliverable — and it is yours, whether or not you continue with us.
A fixed number, the assumptions it rests on, and an explicit list of what is not in it. Most budget disputes come from that last list being missing.
A package is priced by deciding in advance what you get — which means the parts that do not fit your case are still in it, and the parts you actually need are not. We write a spec first, then price that. It takes longer to get a number, and the number survives contact with the project.
Because a price list for work this variable is either useless or misleading. The same "five-page site" costs three times more with a CRM connection and a migration behind it. A number given before scoping is a number that will change — we would rather be slower and firm than fast and wrong.
Read what each one excludes, not what it includes. Then check three things: who actually does the work, whether redirects and SEO continuity are named, and whether the code and accounts end up in your name. A cheap quote that leaves you unable to change supplier is not cheap.
Yes, and it is usually the right call. An audit, a scoping sprint or a first usable version reduces the uncertainty before a larger commitment. It also means you spend on what real usage proves you need.
It does, when nobody plans for it. Traffic drops after a redesign come from lost URLs, missing redirects and structure that changed without anyone mapping it. That work is part of the project here, not an option — losing the rankings you already paid for is the most expensive outcome of a redesign.
Maintenance is not one thing, so it is not one price. Hosting, a domain and TLS renew every year whatever you do. On top sit security and dependency updates, backups with a tested restore, and the small content or feature changes a live site always needs. A site built to be edited by your team costs far less to keep than one where every text change goes through a developer — which is why we build for the former.
By the project, once the scope is written. An hourly rate tells you what an hour costs, not what the result costs — and it puts the risk of a bad estimate on you. When the work genuinely cannot be scoped up front, such as ongoing growth or an evolving product, we use reserved capacity per cycle instead, so you know the monthly commitment even when the backlog moves.
The recurring costs of a site are the domain, the hosting, the TLS certificate and any third-party licences or paid integrations the project needs — a CRM seat, a search service, an email provider. We name every one of them in the quote, with who the contract sits with, before you commit. Recurring costs discovered after launch are the ones that sour a project.
Below a certain point we will say so rather than deliver something thin. When that happens we usually suggest a narrower first version, or point you to a format that fits better. A project that cannot be done properly at the budget available is not a project we take.
Thirty minutes on your goal, your existing site and your constraints — then a scoped, firm quote rather than a range.