Use cases

Service pages assume you already know which discipline you need. Frequently nobody does — what exists is a problem: a spreadsheet the business depends on, a marketing budget producing nothing, a migration that lost traffic, knowledge locked in two people’s heads.

These pages start from the situation. Each one describes how it presents, why it happens, what we would do about it, where it goes wrong, and — importantly — what else you could do instead, including doing nothing.

Every use case page includes a section on alternatives, because the honest answer is sometimes that a cheaper option captures most of the value, or that the problem is not worth solving yet. We would rather publish that than pretend every situation warrants an engagement.

Situations

Working out which problem you actually have

Symptoms are frequently shared between situations. These distinctions are the ones that most often separate them.

Is the data wrong, or is the process wrong?

A dashboard nobody trusts and a process nobody follows produce identical complaints. The test is whether people have built a parallel spreadsheet — if so, the process is the problem.

Did something break, or was it never working?

A sharp fall after a launch is a migration problem. A flat line that has always been flat is a demand or conversion problem, and they need entirely different responses.

Is the bottleneck a person or a system?

Where the same two people answer everything, the problem is knowledge access. Where work sits untouched, the problem is routing and ownership.

Is the cost per lead wrong, or the definition?

Marketing and sales disagreeing about lead quality is almost always a definition problem wearing a performance costume, and no amount of optimisation resolves it.

Is manual work expensive, or just visible?

Some manual processes are cheap and merely annoying. The arithmetic matters more than the irritation, and we would rather calculate it than assume it.

Is this a new market problem or an old one, relocated?

A business entering a new country often finds the constraint it already had at home, magnified by unfamiliarity rather than caused by the market.

Why these pages include alternatives

Because a supplier who only ever recommends their own services is not giving advice. Each of these pages lists the cheaper or simpler options we would compare against, including leaving things as they are, and we mean them.

They also state how we would measure whether the work succeeded, with the baseline captured before anything changes. Where no baseline exists we will say the improvement is unmeasured rather than estimating one retrospectively.

Timelines on these pages are planning ranges drawn from comparable work. They are labelled as estimates every time they appear, and they are not client outcomes.

These six are the situations we are asked about most often. They are not an exhaustive list of what we handle, and if yours is not here the useful thing is still to describe the symptom rather than to guess at the service.

Each page separates what the situation looks like from why it happens, because the two are frequently confused. A business that diagnoses a payment problem as a brand problem will spend a year fixing the wrong thing, and the most valuable part of a first conversation is usually working out which is which.

The alternatives sections are meant seriously rather than as rhetorical modesty. In several of these situations the cheaper option genuinely captures most of the value, and in at least one of them the correct answer is sometimes to do nothing and monitor.

These pages are written from patterns rather than from client stories. The failure modes described are ones we have seen, sometimes more than once, and occasionally ones we caused ourselves early on. What they are not is anonymised case studies, because an unverifiable outcome is not evidence and we would rather not present it as one.

Each page separates symptoms from causes deliberately. The most common expensive mistake in this category of work is treating a symptom as a diagnosis — a payment problem read as a brand problem, a definition disagreement read as a performance problem — and spending a year addressing the wrong thing competently.

The alternatives sections are meant seriously. In several of these situations a cheaper option genuinely captures most of the value, and in at least one the correct answer is sometimes to change nothing and monitor. A supplier who never reaches that conclusion is not diagnosing.

Where these pages give a timeline, it is a planning range from comparable work and it is labelled as an estimate. Actual duration depends far more on how quickly decisions and access arrive from your side than on delivery capacity, and every one of these pages says so because it is consistently true.

Several of these situations compound. A failed migration and a lead quality problem frequently turn out to be the same underlying measurement failure seen from two directions, and a business can recognise itself in three of these pages at once. That is normal and it usually simplifies the plan rather than complicating it.

If none of these fit, describe the symptom anyway. These six are the situations we are asked about most often, not a boundary around what we handle, and the useful first conversation is diagnostic regardless of whether it starts from a page on this site.

One further note on how these are used. Clients frequently arrive having read one page and recognised themselves in the symptoms, then discover during a first conversation that a different page describes their actual constraint. That is a good outcome and it is why the diagnostic conversation matters more than the page that prompted it.

These pages also exist because a great deal of agency marketing assumes the buyer already knows what to purchase. Most do not, and pretending otherwise pushes people toward whichever service they have heard of most recently rather than the one that would help.