Why data quality is the forgotten challenge of performance marketing

George Paton-Williams,
Head of Marketing

The growth of digital causes as many challenges as it does opportunities. For example, the explosion of options and technologies for marketing, advertising and product creates a fragmented world where connecting everything up is not quite as straightforward as it should be.

Integrations between technologies and ad platforms require work to enable the seamless flow of data back and forth. This flow is essential for the successful measurement, analysis and future targeting of digital advertising. But poorly connected technologies can create discrepancies and issues in data that must be solved to avoid issues later down the road. And get it wrong and we lose vital efficiencies, see performance worsen, and decision-making errors creep in.

So why is ensuring data quality such a challenge?

  1. Automated, regular extraction of data across all sources

Manually pulling all your data from across all your channels and technologies every day is hugely time consuming, getting in the way of the more valuable role of using that data (for analytics, for example).

  1. Inconsistencies in data across channels

The format of data from one ad platform is going to be different to the format of another ad platform, and different again to the format of your CRM. If your data isn’t transformed into a consistent format across all channels then that data cannot be easily analysed like-for-like.

  1. Naming conventions

How you set up and name your campaigns, ad groups, creatives (and so on) needs to follow one standardised approach (where possible) to maximise consistency of data and easier manipulation for insights.

  1. User consent to data use

User consent to 3rd party cookies and IDFA from Apple is very low. Similarly, approximately one-third of people use an ad-blocker. Not only does this limit the scale of usable data, but also presents a challenge of how to ensure you only ingest the data that you have permission to use.

  1. Data inaccuracies

60% of data errors are due to incomplete or missing data. Sometimes tracking code and pixels are not present on all pages of a site, or tagging is simply incorrect or outdated.

  1. Data sampling

Some free analytics tools (and even paid ones) will only analyse a sample of your data, rather than your complete, granular data. Clearly this risks you not understanding the full picture of user behaviour.

  1. Differing attribution windows

Setting different approaches to attribution across different activity will make it hard to assess the true value of each channel’s advertising. Maintain the same attribution window and attribution type (i.e last-click vs last-impression vs DDA).

How do you maximise your data quality to improve performance marketing?

a) Tagging and pixel set-up
Do you have your tracking set up correctly through your app or website? Test that events are being recorded correctly at the right time where possible, and regularly verify they are working.

b) Naming conventions
It will make your life much easier if you plan ahead to understand how you want to manipulate your data before you actually get around to doing it. Setting the same naming convention for all activity within a single channel is important. Doing so across channels where possible is helpful, but not always possible, and can be solved in the next step.

c) Extraction and transformation
Typically one tool here can do both parts of getting the data out of all your channels automatically and aligning the data sets behind one standard format. The challenge is still in integrating your accounts to the tool and setting up rules for which data needs to be transformed and how.

d) Data Monitoring
Real-time alerts when there is a discrepancy or issue in your data is important so that problems can be diagnosed and fixed before they lead to mistakes (such as budget misallocation, for example).

e) First party data strategy – maximising depth and breadth of consensual data collection
In a privacy-first world, increasing the amount of data people allow you to use is crucial. Strategise how you aim to maximise customer opt-in and how you plan to make the most of that data. Make the most of technologies such as ‘consent mode’ from Google to ensure compliance.

f) Loading data into dashboards, analytics, and BI tools
The final step is getting that now clean, accurate, and consistent data into the right places so you can use it. That typically includes measurement and analytics tools, or into ad platforms for targeting ads. Set up dashboards so you can view the most important and frequently used metrics.

It is important to design and follow proper technical and strategic processes for managing advertising data. But much of it is technical and not in your typical marketer’s skillset. Get in touch with us if you need some help with your data quality processes.

Get UA insights straight to your inbox


Linen Hall
162-168 Regent Street


+44 203 096 5108

How can we help?