A good data strategy starts with surprising the customer!

Data. It’s a magic word. These days, just mentioning the word data sparks immediate interest: Data Management Platform, data & insights, data driven, data warehouse, data science. All marketers and companies have data, big data!


However, having a lot of data doesn’t necessarily mean a distinctive competitive advantage. Rather than being a discipline in itself, data is a resource that can contribute to a successful strategy. But what is data then? Data shows you trends, gives you insights into previous customer behaviour and can be predictive. However, there are also ultimately people behind the data who aren’t always predictable.

That’s why it’s important to take a creative approach, both with and without your data. Allow room in your total strategy to surprise customers with subjects that they hadn’t thought about themselves. As well as a data scientist, include a creative person and a marketer on your team. Data can be the driver of new creative ideas, and conversely, a creative idea can be better supported by collecting and analysing data.


Data is therefore fundamental!

Even so, many organisations have obtaining a lot of data and a data platform as their goal, which eventually leads to them gathering a lot of meaningless data and losing sight of their customer. To prevent this, organisations should ask themselves what data and having a good data strategy really means for them. In doing so, it’s important to keep the following three steps in mind:


Step 1 – Determine to which goals data should contribute

What are the most important goals that you contribute to as marketing or CRM team? Do you want to prevent churn on current subscriptions, cross- and upsell on the basis of purchasing behaviour, or would you like to use data to increase your service level? Data can often also help with an internal goal, such as showing a purchasing or visual merchandising department how customers purchase clothing. You can use this to adjust your purchasing policy or lay out a store differently. Only after the larger context is clear, can you look at data in a more focused way.


Step 2 – Look at which data you already have available

As previously mentioned, many companies already have data. This is often the so-called ‘what’ data: facts such as age, gender and purchasing behaviour, and the ‘how’ data: through which channels can a customer best be approached. This is good to know, but how many companies are able to establish a date of birth? How many companies follow this up with a birthday email?


Exactly. You have the data, but you have to use it to distinguish yourself and thereby surprise the customer. The match between the goals and available data determines what your data strategy should eventually look like. Don’t only pay attention to which data you need and therefore have to gather, but also look at the quality of the available data. Filled in data fields are not necessarily useful. ​


Step 3 – Test and learn

Start out with a broad strategy: join forces with your marketing strategy and set clear goals. Then go smaller again. It’s simple to test the validity of hypotheses within direct channels. Within your strategy, consider which assumptions about the effect of data you would like to test, and work out these cases. This will help you to keep learning and optimising in a ‘lean & mean’ way. In addition to staying in line with your customers’ changing behaviour, this also means that you can create concrete internal support for plans, and also allocate funds more effectively.


In short, start out with your end in mind, but especially with the customer in mind. Know which data you are gathering and why, and to which goals you are contributing. Establish how you are going to prove this, but don’t go overboard. Ultimately, there should be space in every strategy for surprises.


This article is also posted on the Dutch platform Baaz.nl


Would you like to know more?
Contact me, and I'll gladly tell you more!

Albert Blom
Albert Blom

Data Scientist


"You can have data without information, but you cannot have information without data" - Daniel Keys Moran