Data 101: Becoming data literate
by Graeme Griffiths, Associate Director of Research, Institute of Practitioners in Advertising (IPA)

How can you make sure the data you are working with is reliable and trustworthy? Graeme discusses data literacy in marketing. 

20th May 2022 Read time: 5 minutes Watch time: 1 hour

Data should sit at the front and centre of everything you do as a growth-minded marketer. But how can you make sure you are using the right data and that the data is reliable and trustworthy? Misinterpretation and misrepresentation of data can have disastrous effects. There is a growing need for data literacy in marketing. 

That is the view of Graeme Griffiths, the Associate Director of Research at the Institute of Practitioners in Advertising (IPA). He joined the What’s Possible Community to cover the data basics.

Data is the magic sauce

Data informs so many of the decisions that we make every day in both our work and our personal lives. Across the media and advertising spectrum, data is what we would call the magic sauce in transactions between buyers and sellers,” Graeme said.

Graeme is passionate about data literacy in marketing. Marketers use data to justify the choices made about media investment and to appraise the effectiveness of advertising spend. However, some advertising spending decisions made by organisations might have used proprietary data that could be unreliable and misleading. This is what he wants to remedy through teaching data literacy.

Question everything

Graeme says those interpreting data should be sceptical (but not cynical) about the data they are grappling with. As a rule of thumb, he says, you should question any data you encounter. When scrutinising any dataset, Graeme suggests using the 3 M’s rubric:

  • Measurement: What is the data representative of?
  • Methodology: How was the data collected?
  • Motivation: Does the provider have an ulterior motive?

He says marketers should embark on data collection with clear objectives – and be realistic about what data they are after.

Could a dataset claim to be representative of pet owners with more than three children who visit a supermarket on a Tuesday? That’s not realistic,” Graeme says.

When marketers receive data from data providers, they should do a round of checks. This is to make sure whoever has collected the data has been vetted and that they are reputable. Graeme thinks that having an internal team member within an agency with a good relationship with a data provider who can vouch for them internally is a good idea.

Bias, consent, and presentation

Sampling bias is another thing marketers should look out for. Marketers could draw faulty conclusions from inaccurate and unrepresentative sample groups. He also warns against falsely claiming data is nationally representative when the data may exclude certain vital regions.

Graeme says you can randomise data collection to avoid sampling bias to ensure individuals with similar characteristics do not skew responses.

Graeme says marketers should consider how the data they are interrogating has been collected. Was the data given with consent? Survey respondents are warier than ever of giving their data up without being informed. Marketers need to ensure they have sufficient rights to use the data and know their GDPR position. The penalty for GDPR non-compliance is 2% of a company’s worldwide annual revenue.

Graeme encourages marketers to be as transparent and descriptive as possible when presenting data. Marketers who present misleading data can influence poor decision making. Graeme suggests having many eyes on the data for further scrutiny, which can help reinforce the data’s reliability. 

The future of data

The kinds of data marketers have access to are changing. This is due to privacy concerns from consumers. Graeme states that these disruptions are nothing new – he has seen his fair share of changes in the data world in his long-spanning career. 

I do remember a time before the proliferation of digital data. The data available to marketers and researchers has done a bell curve. The internet exploded, and the availability of census-level data was enormous – and is enormous,” Graeme says.

What we’re seeing now is a tapering off of that access to data. I’ll put my neck out and say I’m not concerned by that personally. I welcome it. The concentration of data is going into the walled gardens of the big tech companies. It makes marketers look for alternatives. There’s been a reliance on this data that never provided context, just raw numbers.”

Graeme sees another wave of monetisation of personal data on the horizon, which may provide the alternative sources of data that marketers are looking for. He also notes that with more walled-off platforms (like Netflix) opening themselves up to ads in the future, they will have no choice but to evidence their user data and make it available for marketing professionals.

Therefore, data literacy in marketing is essential as platforms change and collecting data becomes more complex.