Practical software engineering for UK businesses.
A backup policy tells you what should happen after data loss. A restore drill proves what will happen. For UK SMEs facing insurer questionnaires, client assurance requests and auditor scrutiny, the difference matters. This article sets out a practical quarterly programme for testing recovery and recording the evidence.
Statutory deadlines are not calendar reminders; they are legal obligations with evidence requirements, extensions and awkward calendar rules. This article explains how UK product teams can model deadlines as first-class domain objects, keep an audit trail and test the cases where a normal working day is anything but normal.
Most UK teams treat the December change freeze as an emergency measure announced three weeks too late. It is better understood as a product constraint you can schedule twelve months in advance. This guide sets out how to map deployments around trading peaks, the tax year, bank holidays and annual leave, and how to run a lightweight change-advisory route that still lets genuine fixes through.
Erasure requests are usually handled as legal admin, which means every request becomes an engineering incident. This article sets out how UK product teams can map personal data, build deletion and export workflows, manage backups and processors, and keep an audit trail — turning a statutory obligation into a routine, well-designed operation.
UK addresses resist tidy assumptions: one postcode can serve several towns, a flat may have no number, and a BFPO line obeys its own rules. This guide sets out how product teams should design address capture that converts, validates sensibly and stores data fit for purpose.
Most product teams design for the path where everything works. The path where it does not is where customers are won or lost. This guide sets out a repeatable audit for error, empty and edge states, and a way to rank the fixes by commercial impact.
VAT rates and invoice rules are policy, not constants. This article explains how UK SMEs can model tax as versioned configuration with effective dates, audit trails and testable boundaries, so a future change does not become a rebuild.
Go is often proposed as a cure for slow, expensive software, but for most UK SMEs it is the wrong default. This article sets out where Go genuinely earns its keep — and the questions that help a small UK team decide without outsourcing its judgement.
Buyer security questionnaires can consume days of an SME's time and stall otherwise healthy deals. This article sets out how to build a reusable trust pack — Cyber Essentials status, a data flow diagram, a subprocessor register, hosting locations and a plain-English incident response summary — and treat it as a standing product artefact rather than a last-minute scramble.
Daylight-saving transitions expose brittle assumptions in booking, payroll and reporting systems. This article sets out a practical approach for UK product teams: store UTC, pin the Europe/London time zone, test both transitions, and make time visible to users.
Design systems are usually pitched at enterprise scale, with dedicated teams and multi-year roadmaps. A small UK product team needs something far lighter: a dozen components, a set of tokens, one owner and a contribution rule simple enough to survive a busy quarter.
Most UK SMEs can name the software they pay for, but few can describe how those systems actually talk to one another. This guide sets out a practical integration audit: how to trace data flows, expose single points of failure, review the contract and data-protection side, and decide what to document, replace or retire.
A paid discovery phase should reduce uncertainty before a UK SME commits to a build. This article sets out the deliverables, red flags and questions that separate genuine de-risking from a sales exercise.
Launch is not the end of the cost curve. UK businesses need an operating model for year two that covers security, hosting, compliance and small improvements before the product quietly becomes unsupported.
Custom internal software often looks cheaper than the subscription it replaces, right up to the point where someone has to maintain it. This buyer's guide sets out the total cost of ownership, integration and staffing assumptions that UK SMEs routinely underestimate, and offers a simple scoring framework for deciding between building and buying.
Most software projects fail early and only reveal it late. A walking skeleton — a thin but genuinely end-to-end slice of the product — lets UK buyers test architecture, integrations and supplier claims before the bulk of the budget is committed. Here is how to commission one, budget for it and spot the warning signs.
UK businesses weighing AI tools face a crowded market and competing claims about cost, control and capability. This decision framework compares buying, building and blending AI-assisted software across total cost of ownership, integration effort, vendor risk, data protection and operational fit.
A software project is not complete at go-live if the client cannot operate it independently. This article examines the handover gap, the business-continuity risks it creates for UK organisations, and the contractual and practical steps leaders can take to commission a clean exit from day one.
Compliance with WCAG at level AA has become the minimum standard rather than the objective, and organisations that treat it as an audit exercise are accumulating risk. This analysis examines where automated checks fall short, why testing with disabled users matters commercially, and what UK public sector teams should be doing now.
Most UK software projects do not fail in delivery; they fail at the point of signature. This guide sets out the five warning signs that recur in agency proposals and contrasts each with the practice a well-run supplier should demonstrate, together with a checklist to work through before committing budget.