Paid adverts can bring the first enquiries within a few days. Results become steady enough to judge after two to four weeks, once the platform has learned who responds. Over three months, cost per enquiry usually improves as weak adverts are replaced. Things you do not pay for directly, such as search ranking, reviews and reputation, build over months and keep working after you stop.
A realistic timetable for paid adverts
| When | What is happening | What you should see |
|---|---|---|
| Days 1 to 3 | Brief, plan, adverts and tracking are prepared | A written plan and the adverts for approval |
| Week 1 | Adverts run side by side. The platform is learning | First clicks and enquiries. Unsteady numbers |
| Weeks 2 to 4 | Weak adverts are switched off. Budget moves to the best | Steadier cost per enquiry. A first target |
| Months 2 and 3 | New adverts from what was learned. Follow-up adverts to people who responded | Cost per enquiry improving, or a clear reason why not |
Why the first week is not the verdict
Advert platforms need results to learn from. Meta calls this the learning phase and says an advert set needs about 50 results in a week to leave it. Until then, the cost per result moves around. Judging a campaign on day three, or changing it every day, keeps it in that unsteady state.
What takes months
- Search ranking. A new page or profile takes time to be trusted by Google.
- Reviews. They accumulate one customer at a time.
- Recognition. People usually need to see a name several times before they act on it.
- Word of mouth. It follows good work, slowly.
These are slower, and they keep working when the advertising stops. A sensible plan does both.
Signs something is wrong
- Plenty of clicks and no enquiries after two weeks: the page or chat people land on is the problem.
- Plenty of enquiries and no sales: read people enquire but do not buy.
- Almost no one sees the adverts: the audience is too narrow or the daily amount too small.
- Nobody can show you the numbers: that is a reporting problem, and it should be fixed first.
How long should you commit for?
Long enough to learn something. Two weeks is the shortest useful test. A month gives a result you can plan from. Three months gives time to improve it. The budget planner suggests a period for the amount you enter.