Data-driven decisioning is the zeitgeist of contemporary marketing practice, but barriers to entry, such as lack of quality data and skillsets, etc, are commonplace. Lloyd Greenfield, client partner at The Programmatic Advisory, talks through his consultancy’s three-step approach to getting started.
The explosion of data-led insights, plus martech to manage its output is leading the way to potential increase in ROI on media spend, and the minimization of wastage.
However, this is not to say the management of data is without its challenges. For example, as more-and-more marketers attempt to plan and implement at scale (see chart) the challenges of structuring across channels becomes evident.
A competent approach to data-driven marketing, ie making decisions based on insights derived from consumer data analysis, has some key pillars: the presence of data-driven decision making in an organization; confidence in this activity.
There most commonly cited barriers to taking such an approach are listed (see bullet points below). But this can be remedied by realizing the four key pillars I’ve outlined below, an effective strategy to mobilizing an effective and scalable approach to marketing.
- Lack of actionable or useful data
- Lack of tech or skills available to manage data
How to create a strategy for data
Putting a data strategy in place allows you to prioritize the data available, and how much you use it, as well as helps marketers identify some of the barriers listed above. The below flow chat should help marketers structure their approach to formulating such a strategy. Although the two important points for marketers to remember include: what decisions do you want data to analysis to inform; this doesn’t have to come with a huge price tag.
Understanding your first-party data sources
Many marketers will struggle to understand the quality of data available to them, with acting on the below three points aiding activation.
- Recognize and understand all available data sources
- Identify consumer behaviors across the purchase funnel (both online and offline)
- Understand where your strong first-party data is, then prioritize its use
Getting more out of your external data sources
Advertisers that lack quality first-party data will often look to external sources for such insights.
Such data sources can be used as an additional layer of insight to identify consumer insights. This includes advertisers and publishers packaging their own valuable data (which they are not using themselves) and then making them available to marketers; with these established sources of ‘quality data’ commonly referred to as secondparty data.
When dealing with third-party data sources it is important to consider how this data is collected/modeled/segmented to ensure a balance between scale and accuracy.
First party data is often not used to its full potential. This is often the case as it is too difficult to track, or it has not added enough value to earlier ad campaigns.
To prevent this, marketers should ensure their marketing stack is equipped to track, and activate its strongest data sources. However, it is important to remember that not all new technologies on the market are needed. Although once you are activating data from multiple sources, it may be time to bring additional technology on board. Particular applications that may be of interest include helping teams to collect, centralize and then analyze this data.
Technology won’t replace the need for talent
No matter how good your tech stack is, someone will always be needed to refine this data, and make sure it is collected and analyzed in a compliant way – an in important consideration with the EU General Data Protection Regulations set to come into force.
This requires different types of expertise and new technology is constantly emerging, requiring an accompanying skillset to make the best use of it. This includes: analyzing; measuring; modeling and tracking data. It’s important for marketers to work closely with their agencies and tech providers to understand the collective gaps in their expertise.
Setting this approach in motion
Every marketer will require a different approach, as not all have the same wealth of data, but not all advertisers need complex technologies.
An audit of data, technology and talent against your brand’s specific use cases will begin to help you understand the investment required. From here, it’s best to plot them in terms of the investment needed, and them compare this with the projected long term value they will provide.
While predicting business value can be difficult, but a test-and-learn approach in the use cases with the lowest barrier to activation is advisable way to begin.
TPA recommends that advertisers follow the above process to making the transition to becoming a data-led organization, and better drive value for their business.