Why brick and mortar retailers should pay more attention to data quality

Katerina Bukhvostova

Data quality might not be the trendiest of topics discussed amongst retail peers, but we believe it to be the base for all the future communications with your customers. And you need to get it right from the very beginning.

The inspiration for writing this blog post is the ever increasing number of retailers beginning to rely on in-store data insights to help them with their daily operations. We find it absolutely fascinating, to say the least. At the same time, the amount of new tech vendors filling this space, trying to find their place under the retail sun hardly makes it simple to choose the right tool for the job. Take the example of one of the latest infographics on the subject, which, by the way, seems to already be outdated. What makes this space even more interesting is the constantly developing technology, inevitably forcing the change regardless whether it is welcome or not. 

Looking at the companies listed, you are probably asking yourself whether any of these solution providers are the right ones for you. We are not in a position to tell you what suits your needs, as no retailer is the same. However, we can pinpoint certain things to look for when choosing the right vendor. If you are relatively new to the world of in-store data analytics, do not despair. We will try to make it simple enough and explain why data quality is crucial to your success.

What data should you start with?

With data gathering you can integrate many different data streams, but to get started we believe the three most important data sets are:

  1. Point of Sale
  2. In-store behaviour (Wi-Fi analytics)
  3. People counting

The majority of retailers already use at least one of these data sets for analysing store performance, but very often they are used in isolation. By combining these three key data feeds, the value of unified data set becomes a more powerful tool than the addition of their value as separate data sets.

Once you have these three combined data feeds as unified analytics, you can start adding other data sets like online behaviour, weather, staffing schedule etc.

The top factors that determine the quality of data

Making sure you are getting high quality data works on many different levels. Some of the most widely recognized properties of data quality include:

  • Relevance
  • Accuracy
  • Timeliness
  • Comparability
  • Correlation
  • Completeness

To give you an example, some of the basic questions retailers should be asking: Do I collect the data that meets my current needs? Is it flexible enough to meet my future needs? Can I act on it in real time? Can I answer very specific questions with the data I’m getting? How can the data be compared across all the stores in my chain?

Any data scientist will tell you how important it is to have the correct set of data in order to draw conclusions and be able to serve your customers well. Just one small error will not pose any big problems and might go unnoticed if you are a small shop owner. However, if we are talking about a thousand stores, the game changes drastically. The numbers across all stores turn into a large amount at the end of the year - and suddenly you find yourself in a heap of Big Data!

Considering the above, it is important to note that it is the combination of all the properties that makes the data quality high enough to act upon, which is not often the case for in-store tech companies. 

How to sustain the high data quality level

At Walkbase, data quality is about trusting the data we give to our customers. It’s a very vague and hard to prove statement that is thrown around all too easily. To us it means that we can deliver 1 or 1000 sites to our customers and the real-time data we produce can be used to reliably compare them regardless of their size or layout differences.

This stems from our never-ending quest to ensure data correlation between our data sets. Without data correlation our data output quality would not be very high, and it would be impossible to connect different data sets. The difficult part about making data sets correlate is that we have to remove noise, outliers, which would otherwise skew the actual results. The removal of this noise, i.e. data filtering, is something we have invested a tremendous amount of time into, and something we are very proud of (on the same note, make sure to check our blog post on the technologies we use for getting the largest possible sample size). 

We are often asked how our data is different from, let’s say, large Wi-Fi vendors or footfall companies. The answer to that would be – we make sure to collect the right data for you. Not the extensive (more is not always better), unfiltered data, which is often the case for pure Wi-Fi vendors, but not limited solely to footfall, for example. 

Without having your data filtered and cleaned, you are unable to make the right decisions or trigger marketing messages at the right time. This ultimately leads to retailers always lagging behind and not being able to serve their customers right: having a store associate instantly help a currently dwelling customer or sending the discount including text message to the returning customers at the moment they enter the store would remain things that can only be done in theory. If this doesn’t happen, we all know well where modern consumers go when their wishes are not being met – to your competitors.  

As the leading retail IoT analytics and marketing automation provider, we are able to integrate very large amount of data sources and put them together just right. This means that the quality of data we are producing will always remain high, and is part of the very core of our product. Even if you are at the very start of the omni-channel journey and may only be integrating footfall data with POS, you can count on Walkbase providing you with a high quality data every step you take towards the connected store of the future. 

Are you wondering how advanced is your company on the data quality radar? Take part in our 3-minute test to find out whether you are a Data Unicorn!

Test: How well are you using retail data


Walkbase provides a retail analytics solution for improving the impact of marketing on physical stores and personalising in-store shopping experience.