How data visualisation turns marketing metrics into business intelligence

Why do marketers visualise data?

Data visualisation is a by-product of how humans interpret information. Above all, we have a limited memory capacity, which is why so few of us can remember a phone number by simply looking at it once. As a result, we humans aren’t that great at analysing raw data because we can’t store enough information to analyse more than a few data points at any one time.

This is why data visualisation exists: to group, organise and represent data sets in a way that allows us to analyse larger quantities of information, compare findings, spot patterns and extract meaningful insights from raw data.

Data visualisation doesn’t change the information contained in raw data, it simply represents it more descriptively. And, by changing the way we see and interpret data, marketers can achieve big things with all of the information we have readily available to us today.

Turning metrics into something more meaningful

To illustrate what we can achieve with data visualisation, we’re going to look at an example from one of our clients. This company has branches across the UK and they run local campaigns in Google Ads to target prospects around each business location.

To emphasise the impact data visualisation can have, let’s focus on one single metric in Google Ads: impressions.

Our client has a total of 4,728,073 impressions over the past 12 months but this tells us nothing useful. However, if we segment this data by location and visualise this data in a table, we can start to extract some meaning from the numbers.

The problem with tables is, they’re difficult to read when you’ve got more than a few columns or rows and our client has 39 business locations around the country. So we need a better way to visualise this data if we want to compare the impressions for every location.

Now, we can easily see the top-performing locations, identify the weakest performers and compare any two, three or more locations. This is already quite a jump from the total impressions figure that we started with and, by comparing this table with similar ones for CTRs, conversions and revenue for each location, we can start to build a meaningful picture of campaign performance.

However, we can still do more with this impression data by visualising it in more descriptive ways.

Turning data into business intelligence

Bar charts can help us compare impression performance for each of our client’s business locations but we want to maximise the value of this data. So, the question is: how can we visualise this in a more meaningful way?

We did this by importing their Google Ads data into Microsoft Power BI, which allows us to literally map out this impression data.

Now, we can see where our client has search visibility, the distance between these impression figures and clusters where cities combine to create a larger patch of search visibility. Likewise, we can see where our client is getting no search visibility at all and where large patches exist with no impressions.

The problem with this visualisation is that it only shows impressions and we’ve lost our branch locations in the process. So we can’t see where these impressions are in relation to our clients’ branches, which is an important insight.

In this visualisation above, impressions in the local vicinity of a branch are marked in yellow while impressions from outside a local branch are marked in red and the size of the diamond represents the volume of impressions.

So, now we can see which locations have the highest impressions, where our clients’ search visibility is greatest and – most importantly – locations where they’re generating impressions but don’t have a local store.

In other words, we’ve identified areas where there’s significant interest in our client’s business but no local branch to satisfy the demand.

Turning insights into action

In one final tweak to this data visualisation, we remove impression data from locations within five miles of a branch and this leaves us with two locations in red: Portsmouth and Manchester.

These two locations demonstrate significant interest in our client’s business but there’s no branch in the immediate vicinity. So we could exclude these locations to improve the overall performance of our client’s campaigns but this would be a short-sighted move for a company with branches across the country.

Instead, we can show our client that Manchester is the most profitable location to open a new branch and demonstrate that the interest in their business is already there and untapped – all of which we have determined from visualising one Google Ads metric to maximise its value.

Henry Carless, PPC and data scientist at Vertical Leap.