Nine website situations — and which text answers each one

This section is not a manual to read front to back. It is a switchboard: nine situations firms come to us with, one text for each.

Start with the stage your site is at. Everything else follows from that one answer.

I am planning a site and do not know where to start

The most common mistake here is not the wrong technology or the wrong supplier. It is that decisions which belong to you get taken by somebody else, because nobody asked them in time. Then the template decides the layout, the deadline the content, and whatever the supplier happens to know the technology.

Five decisions cannot be delegated: why the site exists, who reads it, how it is organised, what it runs on, what happens the day after launch. Each has a deciding criterion, and each costs money if it settles itself.

This section is for you if you have a budget and no supplier yet. If you have already chosen one, it matters more — these five answers are what a good supplier asks for at the first meeting, and a bad one never asks and builds the site on its own assumptions.

→ Website strategy — five decisions before the project starts

I do not know what a site needs in my particular trade

Universal advice about "a professional image" does not help, because in every trade something different decides whether anyone gets in touch: in construction photographs of finished work, in hospitality a menu and hours legible on a phone, in transport proof the firm is real and insured, in legal services a stated specialism instead of fourteen practice areas.

Three things are common to all of them: contact in two clicks, visible signs of life, working on a phone. The rest is difference, and the difference decides where the second slice of the budget goes.

The most expensive mistake here is spending most of the budget on the home page layout while contact is decided one page below it. We see it in the quotes people send us to compare.

→ Designing a website for a small business — what changes from one trade to the next

I have a site and do not know whether it works at all

This is the moment for an audit — provided you understand what an audit is not. It is not a report exported from a tool, because a tool lists fifty observations and never says which three matter.

An audit has three layers in an order that is not a matter of taste: technical, visibility, content. Work on the third while the first is broken is thrown away. It is also worth knowing what a free audit structurally cannot give you.

Sometimes an audit is not needed, and we say so outright. Four checks you can run yourself in a quarter of an hour settle whether to order one. That is a result too, and we would rather give it than sell a service.

→ Website audit — what we check and what it costs

The numbers in the reports do not match reality

Before concluding that the site does not work, check whether the measurement tells the truth. Four traps account for most of the discrepancy, and two pull in opposite directions — cookie consent understates traffic, "key events" mistaken for leads overstate conversions. They do not cancel out.

This is the one section worth reading before spending money on anything else. A decision taken on broken measurement is a coin toss, only dearer.

It applies to anyone who receives reports, your own or an agency's. Three of the four are invisible in the report itself, because they operate where data is collected, not where it is presented. You see them only if you know what to look for.

→ Website KPIs — four ways your numbers are lying to you

I do not know whether to rebuild or just fix

Four questions settle this faster than any quote: does the site work on a phone, can your team add a page without a developer, does a new feature need a workaround, is there traffic but no enquiries. The first "no" ends the diagnosis.

The largest cost here is a risk the offers stay quiet about: losing your URLs. A rebuild without address-for-address redirects can take away everything the site built.

The borderline case people get wrong most often: the site looks dated, but works and brings in enquiries. Looks are not a criterion — the criterion is whether you can work with it, and whether it blocks anything you are planning.

→ Website redesign or optimisation

I have been told something is expiring, or that my software is old

A different decision from the one above, and worth not confusing with it. A redesign starts from your symptoms and can be postponed. Modernisation starts from deadlines none of you set — end of support for a software version, browser requirements, legislation — and cannot. It can only be missed, and paid for more dearly later.

Three of those deadlines fall within fifteen months, and none is about how the site looks.

There is one question, and it is cheap: which software version serves your domain, and until when is it supported. If nobody at your firm or your supplier can answer within a day, that is a separate piece of information — and an answer.

→ Outdated website — when the calendar decides, not taste

I do not know where to get customers from

Four channels: unpaid search results, search advertising, social media, referrals from other sites. They differ not only in price but in what remains when you stop paying, and that is the column least often mentioned when budgets are decided.

There is also something no report shows: in a typical small-firm measurement, close to half the traffic lands in a bucket with no recognised source. We show it on our own numbers, having no way to verify anyone else's.

Before adding traffic, check two things: whether the measurement will show the effect, and what happens to the people already arriving. Doubling visits at an unchanged contact rate doubles the cost per contact, if the traffic was paid.

→ Website marketing strategy — where the customers actually come from

I am waiting for results and do not know whether this is normal

Indexing and ranking are two different clocks, and confusing them causes most of the disappointment. A site passes four gates before the contest for position begins, and the third is where it sticks most often and for longest.

We give our own measurement instead of borrowed ranges: re-crawling a URL that already exists can take half a year, and there is no button for it.

Read it before signing an SEO contract, not after three months of waiting. It also lists the things sold as acceleration that are not.

→ How long does SEO take to work — four gates and real dates

Somebody is telling me I have to be accessible

The obligation is real, but it is not universal, and that is the first thing to establish before anyone invoices for it. The European Accessibility Act covers a closed list of consumer-facing services; a firm selling to other firms usually falls outside it.

It is also worth knowing what neither an overlay nor a pass with an automated checker settles: ten criteria are left to a human regardless of what the tool scores.

If the obligation does not reach you, the question left is whether to do it voluntarily — and here the answer differs from most requirements, because some accessibility fixes improve the site for everyone.

→ WCAG compliance — who it binds and what to do

In what order, if you were doing all of it

If you were to work through this section in sequence, the order looks like this — and it follows from each step being useless without the one before:

1. Measurement, because without it every later decision is guesswork.
2. Audit, because it says whether the problem is in the site, in visibility, or in the content.
3. A decision on scope — redesign, optimisation or modernisation, depending on what the audit showed and whether an external deadline is in play.
4. Channels, last. Adding traffic to a site that cannot handle it raises the cost per contact, not the number of customers.

This order is boring, which is why it is rarely proposed. Its virtue: every step can be settled separately and abandoned without loss.