Welcome to the S3A blog
Why we’re starting a journal — and what we’ll write about.
This is the S3A Technologies studio journal. For years we have shipped websites, web applications, brand systems, mobile apps and growth programmes, and almost none of the reasoning behind that work has been written down anywhere outside a client document. That is what this fixes. Three kinds of post will appear here: writeups of recent projects covering what the brief was, what we tried and what actually worked; opinions about performance, typography, identity and shipping under pressure, argued rather than asserted; and reference material for clients — what to ask an agency, what to push back on, and how to read the dashboards we hand over at the end of an engagement. It is written for the person deciding whether to commission something and the person who has to live with it afterwards, which are frequently the same person.
What will you actually find here?
Project writeups. Not case studies with the difficulty edited out — the brief, the constraint, the thing we tried that did not work, and the decision that resolved it. The full case studies live under our work; these posts are the reasoning behind them.
Craft opinions. Performance, type, identity, and the compromises that come with a real deadline. So far that has meant arguing that Core Web Vitals are a growth lever rather than a checklist, and that the cheapest performance win is usually one image nobody optimised.
Client references. The questions worth asking before you sign something. Brand vs. identity vs. logo exists because three very different scopes routinely arrive with similar price tags, and knowing which one you are being sold is worth more than any negotiation tactic.
What this blog will not be
SEO-bait. Listicles padded with stock graphs. AI-rewritten LinkedIn slop. Ten Reasons Your Business Needs A Website In 2026.
That is not a stylistic preference, it is a practical one. The kind of writing that exists to occupy a keyword is obvious to read and useless to act on, and publishing it would undercut the only reason anybody would trust this page — that somebody with their name on it thought the thing was true. If we do not have something to say, we do not publish. A quiet month is a better signal than a filled one.
How do we decide what to publish?
One test: the post has to cost the writer something. A real project, a real opinion, or a decision that could genuinely have gone the other way.
In practice that means a post has to clear three bars before it goes up. It has to be specific enough to argue with — if nobody could reasonably disagree, it is a definition, not a piece of writing. It has to contain something we learned by doing rather than by reading, which is why most posts here point at a project. And it has to be useful to somebody who will never hire us, because advice that only works if you buy something is advertising with paragraph breaks.
That standard means fewer posts. It also means the ones that exist are worth the time it takes to read them, which is the only trade we are interested in making.
Who is this for?
Three audiences, roughly.
Clients, current and former. Treat this as documentation you did not have to ask for. When we hand over an analytics dashboard, a brand system or a Moodle install, the reasoning behind the configuration ends up here rather than only in a call you half-remember.
People still deciding. Treat it as evidence of how we think before you ever see a proposal. You can tell more about an agency from how it argues a technical trade-off than from its portfolio, because the portfolio only shows the projects that went well.
Other practitioners. Take what is useful. Most of what we know we learned from people who wrote things down without charging for it, and this is the part of the debt we can repay.
What should you read first?
Depends what you came for. If you are trying to make a site faster, start with the hero-image post — it is one fix, usually under an hour, and it moves the needle further than anything else at that price. If you are commissioning identity work, a brand is a system, not a logo explains what should be in the deliverable list. If you are weighing search and reputation spend, SEO gets you found, ORM keeps you trusted covers why running one without the other leaks the customers you paid to attract.
Tell us what to write about
The useful posts usually start as somebody’s question. Pricing, choosing a stack, hiring an agency versus building in-house, what an audit should actually contain, when to rebuild rather than patch — if you want any of it explained, send it our way. We would rather answer a real question than guess at a topic, and the questions clients ask in the first meeting are consistently better than anything a keyword tool would suggest.