This post is a result of a campaign a colleague asked me to run. The campaign goal was to get people to attend a praise/worship session at the University of Lagos. As usual, I started thinking of campaign mechanisms and what targeting will work best. What I had concluded was that the easiest people to get to come to this event would be those closest to its venue.

As you’d expect, I tried to target Yaba, but I wasn’t sure of the results because I hadn’t done anything like this before. Which gave rise to the need for an experiment. In the coming weeks, I ran one experiment to decide if this could work or not.

If you’re all about seeing the results, you can jump to the infographics below. Otherwise, here’s the experiment that was run.

Experiment Hypothesis

The basic hypothesis behind this experiment was simple, and shockingly spot on.

  1. Participants who see the ads will be of two types, people who live in Yaba/Ikeja and people who were in Yaba/Ikeja at the time but do not live there
  2. At least 70% of the participants will indicate they were in or lived in Yaba / Ikeja

Experiment Design

Here’s the simple process the experiment followed.

  1. Lead form ads were used to collect user data
  2. These ads targeted specifically the cities of Yaba and Ikeja on 2 different days
  3. Leads were collected over a 2 day period for both cities
  4. Users were asked where they saw the ad, and where they lived

Without further ado, here’s the result from that experiment.

facebook-city-targeting-test-nigeria

Share this Image On Your Site

2 use cases for city based targeting for paid ads

 

Brick & Mortar Stores

Think about Shoprite, they have stores scattered around Lagos and other states. What if we could target the cities where these store are? Assume for a moment that different stores needed to get “rid” of specific stock peculiar to that store. Instead of running ads to everyone in Lagos saying sales, I could run a campaign for Ikeja saying 80% off electronics with directions for the store through Google Maps. Now someone who steps into Ikeja sees the ad and knows they can participate in the sales. Imagine this scaled to multiple cities, providing data on store performance across states and cities. Just pure bliss man

Remarketing + Location Targeting

Debras Grace, a female fashion store in Yaba recently launched their ecommerce website. If I were to be asked to drive in store traffic, outside of just running ads targeting Yaba, I could go one step further. Imagine I took the remarketing list of all engaged website visitors, people who added to cart but didn’t purchase. Now all I’d need to do is setup an ad for this audience specifically in Yaba with an offer of 10% off when you purchase in store. Now users who see the ad while in Yaba have HUGE context already. If conversions aren’t high on that ad, its probably jazz

These are just a few among many, I can think of three other ways right now, but we can keep those for later.

As the market continues to grow and demand more performance based marketing campaigns, things like this will become invaluable to marketers and SME owners. Next up, we’d be testing the limits for Google.