Data Discovery in Digital Marketing
The challenge from data
Data discovery is all about finding patterns and trends in data that ultimately yield useful insights.
Digital Marketing data can provide stunning insights into ways to get the best outcomes from a campaign. But it can also be challenging, because there is so much data available.
** Google Analytics, for example, has hundreds of metrics relating to website usage **
** Facebook – over 50 metrics just for tracking posts **
The same again for Instagram, Reddit and TikTok.
A Digital Marketing campaign, involving several social media and email platforms, might therefore generate data spanning hundreds of different metrics.
The challenge is to find critical insights that can explain how to deliver more value from your Digital Marketing campaign without the search for those insights becoming a time consuming and potentially expensive task.
The requirements for data discovery
For effective data discovery in Digital Marketing there are four requirements:
- Being clear about the question that you want to answer, for example, ‘has my digital marketing campaign generated more sales’. Focus on just one question at a time because the answer might determine the direction you then want to take.
- Being able to turn the question into a specification for the data you will need to help find the answer.
The question will help you to quickly hone in upon just a few metrics out of the hundreds that may be available saving you time and money and greatly increasing the chances of success.
- Gathering the data you need from different Digital Marketing sources into one place e.g. a database or a spreadsheet.
- Having the right tools to discover relationships in the data that provide an answer to the question.
What's the question?
Data discovery is needed when you are faced with questions that are important and difficult to answer. These two characteristics often go hand in hand.
For example, a question such as ‘has my Digital Marketing campaign generated more sales’ is clearly important (it’s hard to develop a Digital Marketing strategy without being clear on this point). It also hard to answer.
This is because it alludes to measures and records that will sit well outside of your Digital Marketing and Social Media platforms (e.g. most likely found in Sales, Finance or CRM systems) as well as sitting within them. No one system will have all the facts.
If you are to bring relevant data together from across your campaign and from across the business you need to be certain that you are focusing on the right data.
This is why the question matters. The clearer and more specific it is, the easier it is to identify the data that you will want to explore in order to find the answer.
Identifying the right data
For a question such as ‘has my Digital Marketing campaign generated more sales’ you are clearly going to require sales data over a period of time, plus data representing all the relevant inputs into the campaign.
This might suggest a big gathering of metrics from many different places. There will indeed be a point at which you will want to know, for example, which marketing channels are leveraging the greatest impact over sales – and which could be halted without having an adverse impact. For this you might want records of all the channels and platforms involved in your campaign.
But, for this specific question, the most relevant inputs might simply be the direct cost of the campaign, such as the cost of Ads, and the cost of peoples’ time.
If you can show that you have generated a return in the form of increased sales from the money spent on the campaign you are getting right to the heart of the question.
Assembling the right data
For this question the relevant data might therefore be straightforward to identify and collect. If you can demonstrate a clear positive influence upon sales from your spend on the campaign, the answer to the question will be a ‘yes’ – the campaign has generated more sales.
This means that you can keep going with the Digital Marketing activities in place so far and sales should continue to climb.
If there isn’t a clear positive influence then the answer has to be ‘no’.
But note the use of the word ‘influence’. There might have been an increase in sales over the life of the campaign, but it might not have been due to the Digital Marketing campaign.
The data needs to be arranged in such as way that ‘influence’ can be inferred. This means having your measure of value e.g. sales over at least the length of the campaign, and the cost of the campaign, broken down by day and placed in adjacent columns in a table or spreadsheet.
In this format, and at this level of detail, you will be able to calculate how strong the relationship is between sales and costs – and therefore how much value the campaign has created.
Using the right tools
To know how much influence your Digital Marketing spend has upon sales we need to firstly visualise your spend and sales through simple charts – but charts designed to reveal whether fluctuation in one correlates positively with fluctuation in the other. Put another way, do both sales and spend values move in the same direction at the same time. Data visualisation applications such as Microsoft’s Power BI are ideal for this purpose.
Correlation does not infer causation. In other words it might be coincidental that sales and spend of Digital Marketing fluctuate in line with each other. It might also be the case that the Digital Marketing campaign was partly responsible for the increase in sales, but other influences were also at work.
You will therefore need to pass the data through an algorithm that calculates the level of correlation between two or more sets of values – and also assigns a probability that any observed correlation is purely an outcome from chance.
It will also calculate how influential the campaign has been, in other words the extent to which other influences are in play.
This means knowing how much you have to spend on Digital Marketing to achieve an uptick in sales – with those other influences at work, and also if those influences were no longer present.
This is powerful stuff because it then allows you to set a budget for your Digital Marketing campaign confident about the outcomes it will deliver under a range of different scenarios.
Explaining correlation
The analysis so far will not tell you which Digital Marketing activities are having the greatest impact over sales, only that there is a correlation between the overall spend, and sales.
To understand why there is correlation you will now need to take the step that you may have so far avoided – the gathering of multiple metrics from different Digital Marketing platforms.
This is to discover whether specific outcomes from Digital Marketing activities correlate with sales, and whether they can explain why the spend on the whole campaign correlates with sales.
In a recent example involving a UK University, we demonstrated that spend on Digital Marketing campaigns correlated strongly with Applications to study on Undergraduate programmes.
To understand why, we selected some specific metrics from different platforms e.g. levels of engagement with social media posts and Ads, and website page views arising from Google searches.
We used the same algorithm as described earlier to discover how much each metric correlated with Applications to study. We also examined clusters of different metrics.
Indeed Meta and Google Ads together were shown to have the greatest level of influence over Applications.
Less is more
The finding was of great value. It confirmed the value of the advertising spend on those platforms but also highlighted the poor performance of other advertising platforms.
This meant that spend could be focused just upon the most effective platforms leading ultimately to better results being delivered on the back of a reduced Digital Marketing budget.
This is where we can help
Digital Marketers’ super-power is creating compelling content that turns heads, sways opinions, creates clarity, motivates and moves.
At FiguringOutData.com our superpower involves working alongside you to bring your data alive through analytics powered dashboards. We can develop them for you, or train and support you to develop them.
—-
Paul Clarke is the Director of FiguringOutData.com, a Data Analytics Service Provider supporting teams make the best use of data to guide Digital Marketing and Social Media management.
For information about our services and for guidance in the use of data to support Digital Marketing please get in touch using the contact information above.