Facebook Ads Didn’t Work For Us… So We Rolled Our Own

Here at AddyCar we have a vision: leverage sharing economy to help commuters offset their motoring costs and empower startups to advertise outdoors just like the Big Corps can.

To achieve this vision, we need to onboard drivers in key areas of London to be able to advertise to potential consumers anytime anywhere.

For this particular campaign we focused on acquiring users in East London.

Here’s how our increase in coverage compared between Facebook and AddyCar campaigns:

New coverage from Facebook (left) and AddyCar (right) campaigns
New coverage from Facebook (left) and AddyCar (right) campaigns

In addition to raw user acquisition, we also wanted to increase AddyCar’s brand awareness among drivers. We measured this using website visits.

Website visits from FB (yellow) and AddyCar (blue) campaigns
Website visits from FB (yellow) and AddyCar (blue) campaigns

More generally, this campaign allowed us to test-drive our workflow from start to finish, and save costs compared to Facebook Ads.

Whom are we targeting?

In any marketing campaign, it’s important to understand who your ideal users are, what motivates them and how your product can empower them to achieve their desires.

As such, our ideal user owns a smartphone, drives their car during the day in East London, and is eager to earn extra income, which could be achieved by placing advertisements on their car.

The Facebook Ads way

First, we needed to establish a baseline, so we ran Facebook ads for 5 weeks targeting users in London. Facebook has a wealth of information about its users, so we further narrowed our audience to those users who own a car, whose car insurance is expiring in the next three months, who own a smartphone without AddyCar app already installed. In terms of demographics, we targeted males 18–30 years old without an advanced degree, as this demographic had the lowest cost per install based on our previous campaigns.

Our previous Facebook Ads campaigns showed that simply targeting users who are currently in London would yield a significant volume of installs from other cities. We even got an active driver in Bulgaria. So, for this campaign, we went a step further and explicitly excluded users in cities other than London.

This was a bit tedious, but helped with geo-targeting
This was a bit tedious, but helped with geo-targeting

From start to end this process took us 30 minutes to submit our ad for approval, as we’ve already gotten used to Facebook Ad Manager user interface. Two hours later our ad was automatically rejected for making “unrealistic claims” regarding drivers’ potential earnings, so we went through the appeals process to get the ad approved. Overall, FB ad submission process was relatively easy as you’d expect from online ads, and we were able to start running the ad within 24 hours.

For the sake of meaningful metrics, we defined an active user as those who installed the app and driven more than 50 miles in London. Here are the results that we got from this campaign:

The AddyCar way

When setting up our AddyCar campaign, we were also looking to test-drive our workflow, so we created an AddyCar customer account and used the search tool to find drivers in East London.

AddyCar campaign planner
AddyCar campaign planner

We chose a driver who commutes daily between Woolwich and Aldgate, as this route covers many miles, passes through some busy roads and was estimated to yield 2,300 impressions per day.

Overall it took 4 days between us creating an account and the chosen car being wrapped and hitting the road. Once the driver was selected, we designed the ad, and sent this to our wrapping supplier.

Once the campaign was underway, we were able to monitor driver’s routes, distances and gathered impressions. This was largely in line with our expectations, as it’s unlikely for a person to make sudden changes to their daily routine.

As a result of this campaign running for 11 weeks, our results were as follows:

It’s also important to note that we received just one install in the first week, but this number soon improved as commuters saw our advertisement multiple times and decided to give AddyCar a go.

Toe-to-toe: Facebook Ads vs AddyCar

As a result of these two campaigns, we saw that AddyCar provided 10.43% savings for active user acquisition and 35% savings for website traffic compared to Facebook ads.

Furthermore, with AddyCar ads it was great to see our brand spreading in real life.


By the end of our campaign, we were happy with how it turned out. We got the results that we wanted and received enough active users in our targeted location to tick East London off our list.

The driver reported feeling a lot of positive attention from pedestrians and other drivers.

“I feel like all eyes are glued to me when I’m on the road”

— Shuheb M.

Being able to see our brand on the streets of London made it feel a lot more real compared to online marketing, and it looked great! We were also able to elicit emotional response from our target audience with some of them taking photos of the ad.

As an added benefit, we also received some word of mouth signups as friends and family of our driver were asking them questions about the advertisement on their car.

Facebook Ads yielded a similar quantity of installs per ad spend, but the targeting wasn’t as good as we hoped, despite setting relatively strict filters.

As expected, outdoor advertisements weren’t ideal for app installs as the audience had to remember the brand name and subsequently look it up once they’ve reached their destination. However, it was great for raising brand awareness as well as gaining user confidence that our brand was “real” as opposed to another faceless website.

Your browser is out-of-date!

Update your browser to view this website correctly. Update my browser now