Your ads are working, sales aren't — how to find the bottleneck

You open the ad account — clicks are there, impressions are there, even the CTR looks decent. The phone doesn't ring, the cart stays empty, the enquiry box is quiet. The first instinct is to raise the budget or rewrite the ad copy, because that is the one number visible on the dashboard.

But an ad has one job: get a person to the door. It does not sell. Everything that happens after the click — the page, the form, how fast you respond — sits on your side of the fence, not the ad platform's. And the problem is rarely "everywhere at once." It is usually one specific link in a four-link chain.

The elimination method — measure, don't guess

The path from click to sale has four links: click → page → form → response. The chain breaks somewhere, but you cannot see where by eye — only by number. Measure each link on its own and compare it against what a normal range should look like. The first link where the number falls off a cliff is your answer, and you can stop checking there, because whatever comes after a break no longer matters.

Link one: the click — is the right person arriving

Click volume can look fine while the question of WHO is clicking stays open. On search ads, open the search terms report in the ad account and read what people actually typed. A common pattern: the campaign runs on broad match and picks up curious searches like "free", "job", "how to do it myself" — interested people, not buyers. Those clicks are cheap, which keeps CTR looking healthy, but they never convert because the person was never a customer to begin with.

Social ads carry the same question in a different form: is the creative reaching a broad audience or a ready-to-buy one? High CTR combined with low results is almost always broad interest getting mistaken for buying intent.

Link two: the page — is the promise kept there

Whatever the ad promises has to show up on the page within the first second. This is called message match: if the ad says "30% off", the same figure needs to be visible on the first screen — the visitor decides in a couple of seconds whether they landed where they meant to.

Load speed is its own checkpoint

Most mobile visitors open the page on mobile data, not office Wi-Fi. If it takes more than two or three seconds to load, the click is already gone — nobody waits around "just a bit longer." This is not something to guess at; it needs a real measurement.

The median landing page conversion rate across industries, per Unbounce's 2024 benchmark report, is 6.6%, ranging from 3.8% to 12.3% depending on the sector. If your own page sits well below that, the problem is not the click — it is here.

Link three: the form — how many start, how many finish

Track two separate numbers in analytics: how many people who reached the page touched the form, and how many of those who started actually submitted it. A wide gap usually has a simple cause: too many fields, fields marked required that don't need to be, the wrong mobile keyboard type (a text keyboard popping up for a phone number, for instance), or plain silence after submission — no thank-you message, no email.

Fill in the form yourself, on your phone, all the way to the end. On every field, ask whether it is genuinely necessary — each extra one loses another slice of people.

Link four: response speed — how many hours pass after the enquiry

The form went through, everything worked up to this point — and then what? Compare the timestamp of the enquiry landing in the CRM against the timestamp of the first actual contact. In a lot of businesses that gap stretches to the end of the day, sometimes to the next one, because enquiries drop into a shared inbox and nobody is sure who is meant to check it.

The cost of that gap is straightforward: the moment someone submits a form, they likely still have two or three competitor tabs open, and the decision can go there instead. The faster the reply, the longer that window stays open in your favour. Testing this is simple — submit your own enquiry and time it yourself.

An ad's job is to get someone to the door. Opening it is your job.

How to read the result

You have checked all four links on their own. The rule is simple from here:

• Problem at the click — it is an audience or search-term issue, not the ad copy itself.
• Problem at the page — a mismatch between the promise and the content, or load speed.
• Problem at the form — a design and field-count question, unrelated to the ad budget.
• Problem at response speed — an internal process issue; changing the ad will not fix it.

Raising the budget only makes sense in one case: all four links are within a normal range and you simply need more volume. In every other case, extra spend just pours more through the same narrow point.

If you want a real number for page speed, our free speed audit tool takes the page address and pulls figures from Google's own measurement source.

If you've already run this diagnostic and found where the chain breaks but don't have time to fix it, our conversion optimisation service does exactly this — measuring all four links and ranking them by how much they are costing you.