We spent $213 on Facebook ads and got zero calls. The mistake was in the first setting.

We ran Facebook ads for our own business and got nothing out of them. No calls, no enquiries, no leads. Two months of wondering whether Adelaide simply was not interested.

Then we went and read the numbers properly, and the answer was embarrassing and completely fixable. It is the same mistake we now see on most small business ad accounts we look at, so it is worth writing down.

What we actually bought

Three campaigns, and here is what each one returned.

• $28.35 — 57,680 impressions, 81 clicks, 19 people actually reached the website, 0 leads
• $134.58 — 21,184 impressions, 678 clicks, 290 post reactions, 0 leads
• $50.90 — 4,496 impressions, 211 clicks, 144 post reactions, 0 leads

$213.83 in total. 970 clicks. 434 likes. Nobody rang.

The mistake

Every one of those campaigns had its objective set to traffic or engagement. Not one was set to leads.

That sounds like a technicality. It is the entire ball game. When you tell Facebook you want traffic, it goes and finds the people in your area most likely to click a link — which is a real skill, and a completely different group from the people most likely to ring a business. When you tell it you want engagement, it finds people who like and comment.

It did exactly what we asked. We asked for clicks and likes, and we got 970 clicks and 434 likes for $213. Nobody rang because at no point did the system go looking for somebody who would ring.

The part that hurt more

$185 of that $213 went on boosting posts. The boost button has no objective at all — it optimises for reactions. We spent most of the money buying likes at roughly 43 cents each.

If you have ever pressed Boost Post because it was there and it was easy, that is what you bought too.

And a third problem underneath

Look at the first campaign again: 81 clicks, and only 19 of those people reached the site. That is a 77% drop-off between somebody tapping the ad and the page appearing.

They did not change their minds. The page was too slow on a phone on mobile data, and they left before it loaded. We were paying for 81 visitors and receiving 19 — roughly four times the real cost per person.

What was actually working

Here is the encouraging part, and it took reading the table to see it. The two boosted posts had click-through rates of 3.20% and 4.69%. For this kind of advertising that is genuinely good — most campaigns sit around 1%.

Same city, same audience, same business. Adelaide was responding. 434 people reacted to posts about a small automation agency. The interest was real and we had simply never asked those people to do anything.

The traffic campaign, by contrast, managed 0.14% — twenty-three times worse. The difference was the creative: we had used our logo, a 182 by 169 pixel image, stretched into an ad slot.

What we would tell you to do

Set the objective to leads. Not traffic, not engagement, not reach. If the point is the phone ringing, tell the system that in the first setting, because everything downstream follows from it.

Do not press Boost Post. It is the most expensive way to buy the least useful thing.

Check your site on a phone on mobile data before you spend anything. If it takes more than a few seconds, you are paying for people who never arrive.

Do not use your logo as the creative. Twenty-three times worse is what that cost us.

None of this required a bigger budget or a cleverer agency. It required reading a table that had been sitting there the whole time.