More booking calls for a taxi company, at less than half the cost per call

A taxi company was paying for Google Ads that made the phone ring, but nobody could tell which calls turned into rides. We connected the calls to real bookings and showed Google the difference.

A rank of taxis waiting under street lights at night
Client
An Australian taxi company
Sector
Taxi and private hire
Where
Australia
more booking calls
+27%
lower cost per call
−56%
average cost per click in their market
$2.10

Where things stood

The company had been running Google search ads for some time, and on the surface they were working. The phone rang, Google reported a steady stream of calls, and the cost per call looked reasonable.

The trouble is that a phone call isn't a ride. Some callers booked a taxi. Others had dialled the wrong number, were drivers ringing the office, were other businesses, or just wanted to know roughly what a trip would cost. Google counted all of them the same way, and so did the company, because there was no way to tell them apart.

The question we started with

We didn't need to know how many calls the ads produced. Google already reported that. We wanted to know how many of those calls became a paid ride, and what each of those rides had really cost to win.

It sounds like a simple question, but nobody at the company could answer it, and that's normal. Most businesses that take bookings over the phone are in exactly the same position.

Booking calls and cost per call, before and after Google was shown which callers had paid.

Connecting calls to bookings

First we moved the calls coming from the ads onto a tracked phone line. Each call then carried the campaign and the search that led to it, so we could see what the caller had been looking for before they rang.

Then we matched those calls against the company's own booking records. Their system already knew which callers became customers. It had just never been joined up with the advertising.

What the numbers showed

Once the two were joined, it was obvious that a large share of the budget was going on calls that never turned into rides. The campaigns Google rated highest, based on clicks and call volume, weren't the ones bringing in paying passengers.

What we changed

We built a list of the callers who had actually paid and gave it back to Google as the definition of a good result. From then on, Google's automatic bidding, which decides whose searches to show the ads on and how much to pay for each click, was aiming at people who looked like real passengers rather than anyone likely to pick up the phone.

Booking calls went up by 27%, and the cost of each call fell by 56%. In a market where an average click costs around $2.10, that difference adds up quickly.

If you’re in the same spot

Ad platforms are very good at finding whatever you tell them to look for. Tell Google that any phone call counts and it will find you phone calls. Show it which callers actually paid, and it goes looking for more people like them.

More case studies

See all case studies