Using Location Data to Profile Your Customers and Target Your Advertising

This post aims to help you see if you can leverage your existing customer database to help better target customers by asking the following questions:

Are certain geographical locations home to higher value customers than others?  Are certain products more popular in certain geographical locations than others?  If so, what can you do with this data?

The answers to some of this may seem obvious.  You may know, for example, that your highest value customers are in London or California.  However you may also find some interesting information that you didn't previously know about.  This technique is quick and simple to do so could well be worth exploring on a quiet afternoon.

Examples of how this might generate useful information:

Do you sell clothing?  Perhaps cold weather wear is more popular in certain parts of the country than others?

Are most of your customers from a location-specific industry?

You may also find that there are groups of higher value customers living outside of the areas you expected.

This process can be useful regardless of what kind of website you have, however in my experience it is most useful for ecommerce sites.  This is usually because you will have a larger data set of customers to work with, as well as more complete address data for these customers.

Selecting The Right Tool

There are lots of tools available for plotting data on a map, however my favourite one so far is “batchgeo”.  They also have a handy example spreadsheet which is pre-prepared in the correct format for uploading to their site.  You can start by downloading this spreadsheet from their site and getting ready to copy your own data into it.

Preparing Your Data

You need to select a particular question you want to answer.  I’ll use a clothes retailer as an example, so let’s consider the question suggested above; are certain types of clothing more popular in certain parts of the country?

I would begin by selecting all the customers that bought a piece of cold weather gear, maybe a fleece, warm coat or woolly hat for example, and exporting them to one list.  The data you export should include some relevant piece of geographical information such as a postcode.

This is essentially all the information required to do a really simple evaluation.

You should then copy and paste this data into the example spreadsheet you downloaded from batchgeo.

Once you've got the data in the correct format you can copy and paste it straight into batchgeo’s tool, then hit “Map Now”.

This will now generate a map and plot the locations of all the people that bought cold weather gear from you.

If you look at this data, can you see any noticeable patterns?  Are there clear areas of the country where these products are popular?

How can you use this information?

If you run any localised marketing campaigns you could consider concentrating those campaigns in the areas you just found which show where most people are likely to purchase the products in question.

This is a great example of when to run a Pay-Per-Click campaign.  You can run really well optimised ads in these specific areas.

I've used this technique to figure out some really interesting information.  The more I play with it the more I come up with new ways to use the tool.

As with any marketing technique don’t forget to test it.  Compare the efficiency of ads generated using this information to ads generated without it.  You might just surprise yourself.

Another thing to consider is to see if you can profile what makes this particular location relevant.  You can then see if you can find other parts of the country which are similar and try a localised AdWords campaign there too.

In the example above it is likely that you might see closer groupings in colder parts of the country, however your product or service may make it more appealing in certain other locations.  For example, imagine if you sell buoyancy aids and happen to find that your data shows that they are commonly bought by people who live on large rivers.  The obvious thing to do is look for other locations where people live on the river bank but you currently have no customers, then run a localised AdWords campaign there too.

Going Further

Batchgeo lets you “group” certain parts of your data.  You can group your data in any way you like but one I have used previously is to group customers by average annual spend.  E.g. £0-50, £50-100, £100-150 etc.  Sometimes you find no obvious patterns to the plotted data, but sometimes you find that there are certain areas where customers are more likely to be in the higher spend groups.  It is worth considering increasing your maximum cost per click in these areas.

As I said before, the key is to test this yourself.  If you find something interesting in your own data, or find another great way to use location data then leave us a comment below.

Matthew Weeks


Follow Matt on Google+

Leave a comment

Make sure you enter the (*) required information where indicated. HTML code is not allowed.