4 Steps to Becoming a Data-driven Retailer

2019-11-04 10_43_37-Free Road Image on Unsplash

Today’s customer expects retailers to know and follow them through online and physical interactions with their brand, providing personalized experiences as they go. Which means that retailers need to leverage all of the data customers provide them.

No problem! More and more customers are willing to trade their data if it means they’ll get personalized experiences in return. But do you know how to make the most of the data you're getting from your customers? According to HBR Closing the Customer Experience Gap report, only 23% of retailers say they can act on all or some of the customer data they collect. There is clearly a gap between what customers want and how retailers handle the data to actually deliver that experience.

Here is a 4-step approach to becoming a data-driven retailer:

1. Identify all your data collection points

Don’t limit yourself to only the data sources you actively use now, like your CRM and POS; you may have more sources of data than you think. For example, what if a customer comes into your store and asks a question about a product she wants to buy for her partner’s birthday? The fact that this customer is married and her partner’s birthday is great data and could help your company market and sell to her more efficiently in the future. But are you capturing that data? And further to the store interactions, what about a customer’s email behavior, or any live chat engagements they’ve had?

If you’re looking for a good way to identify all of the possible data collection points you have, try going through the customer journey and figure out all of the touch points he or she has with your brand along the way. All of those touch points present moments where you can and should be capturing data on your customer.

2. Capture the data and consolidate

First, at each possible data collection point, ensure you’re gathering as much data as possible. To do that, you need to create value for your customers - give them a reason = reason to give you their data. Examples of this include giving customers discounts, or exclusive member benefits, for trading in their data. And, as a bonus in your physical stores, this will give the store associate more of a reason for asking for that data in the first place, when they know they’re giving the customer a chance to get more value, and not annoying them with questions.

Next, consolidate all of the data you have. Which is no small feat. As we need to break down organization silos and look across all channels for customer data, we also must break down the data silos, and combine it all into one source of truth. This eliminates duplication, inaccuracies and inconsistencies in the data, and will ensure your data is reliable and trustworthy, and there are no issues around whose data is right.

3. Analyze and predict

Now is the time to find patters and create propensity models. Which is a fancy way of saying you take your customer characteristics and match them to what you expect them to do next, based on what other similar customers have done before. This is a better gauge of what the next action will be of a customer, which can inform both your marketing efforts and your inventory forecasting. Here are some examples of what I mean:

  • Marketing: If a customer identified as having a kid between 1 and 2 years has shopped online only between 12:00PM and 2:00PM on the weekends, then use that information to serve this customer with promotions in and around that time for their next marketing communication (FYI, if you’re actually seeing a customer with this shopping pattern, it’s likely their toddler’s mid-day nap is their reason for their online-shopping session, something that I personally can confirm is a thing).
  • Inventory forecasting: Part of the data collection process can and should involve where each customer is shopping, in terms of both brick and mortar stores and the online store. Then you can look at the customer base profile for each store. If you identify that a given brick and mortar store over indexes in a customer with similar characteristics, this can inform an adjustment to your product selection available at that specific store. For instance, store A has more customers with young kids, and store B has ore young adult customers. Store A may need to carry more baby-and kid-friendly products. This concept is called localization of inventory, and it’s an emerging trend in retail.

4. Automate

The beauty of creating a propensity model is that you can automate the next action, and be sure that that it is informed by real data.There are a ton of possibilities of how to do this in your marketing through marketing automation. Delivering personalized recommendations based on the last action taken, with products selected based on the customer’s last actions, are some examples.

At your store level, look at your customer profile base for the specific store, and the propensity for that base to purchase specific products. Then set up min/max levels for your predicted bestselling products, in your in-store technology. Ensure your in-store technology has the ability to set these min/max levels for individual products, at the store level.

Want to see it all in action? Join our webinar

In retail, we are flooded with real and potential customer data points.  The challenge now more than ever is to figure out not how to capture it, but what to do with it, to ensure it has an impact to your business. If you want to learn more about turning your data into real action, join our retail experts for our Turning Data into Action webinar.

Watch Webinar