How to Solve Three Big Data Challenges for Better Audience Targeting

Here in the Consult team we have been thinking about what makes a great Data Strategy and why it’s so important to have one. It’s not always easy to know where to start when developing your own Data strategy, so in our latest blog we’ll share some insights on how we have solved three data challenges this year.

Here’s a look at the challenges and corresponding solutions:

Building better use cases with a Data Management Platform (DMP)

Challenge: You already have a DMP, but you are not happy with existing use cases and you don’t feel that you are getting enough out of the technology that’s been implemented.

Solution: Some use cases are often overlooked because of a perceived difficulty in implementing them, sometimes the use cases stretch the capabilities of the technology or require more groundwork than normal to get up and running. We’ve found that whilst it’s important to ensure that these requirements are not overlooked in the technology on-boarding process, it’s equally important to have a plan to keep on top of each use case as they are rolled out alongside your marketing activity.

One example of an under-utilised use case is what we refer to as single vs combined traits:

  • A single trait is a simple segment based on an event e.g. someone who viewed a video called ‘Sport’ or someone who visited a given page on your website.
  • Whereas a combined trait is a segment that’s made up of multiple events and perhaps even excludes several others e.g. visitors who engaged with a recent marketing campaign but have not visited your website.
  • For most companies, these traits (also referred to as audiences) can be quite long and to be useful the traits needs to be kept up-to-date to reflect the variety of campaigns, content and customer touchpoints at any given time.

We have also found that activation is often held back purely for resource reasons. When scoping any work with a DMP it’s always important to look at the resource that will be required, not just during the rollout period but post-rollout as well. We recommend going back to basics to solve this; reviewing how the technology will fit in with your existing processes and reviewing where the resource gaps are as things currently stand.

Improving campaign ad frequency

Challenge: You’re not sure if frequency per user (especially in display) is being capped effectively. On the one hand, high frequency can mean wasted budget and on the other hand, it can simply serve to scare off any prospects long before they have visited your site, as they will be bombarded by ads.

Solution: Working with the Salesforce DMP as well as a number of Demand Side Platforms (DSP) we have started to see success in capping frequency on a number of different levels for our client’s businesses, such as:

  • Between different technologies e.g. DSP A frequency capped against DSP B
  • Between different brands or products e.g. Capping exposure of product variant A against product variant B
  • Between different portfolios e.g. capping exposure between different verticals within the same business, such as insurance, loans, saving and credit cards

And whilst frequency capping capabilities are by no means perfect perhaps there is still some steps that you can take to improve your current process?

Identifying the right audiences

Challenge: How do you identify and prioritise audiences given the sheer amount of data out there? For instance, at one end of the spectrum there are one-off reports and insights that might be useful in guiding your data strategy, whilst at the other end there can be real-time data from your website or DMP; and all this is before you have even touched on the gamut of 2nd and 3rd party data in between.

Solution: Whilst it’s tempting to be extremely detailed when it comes to data strategy we recommend keeping things as simple as possible (at least to start with). It’s important to recognise that each advertiser has different requirements when it comes to building audiences.

When working with one of our clients this year it was clear that age targeting would be a critical ingredient, so we  came up with a solution to improve our targeting capability and validate the data, this included devising:

  • A package of third party technology for pre and post validation to improve optimisation
  • Creating pre-validated data segments to achieve the 95% target
  • Embedding in-house capability into marketing tech e.g. within DSP or DMP
  • Creating bespoke data partnerships – alternative data providers where we can compare demographic information

In summary, it is important to understand the end goal and what you are trying to achieve with your data strategy. Often, having a roadmap will help to plot key milestones and ensure implementation is on track. The above are just some of many examples , that should help to give you a starting point to build your strategy.

comments powered by Disqus