7 Actionable Analytics for In-Store Customer Journey

Published by Ronny Max on

What is Actionable Analytics for the In-Store Customer Journey?

The In-Store Customer Journey is quantified by (Anonymous) Tracking, and it includes the Path to Purchase (where), TouchPoints (what), and Buyers Behaviors (why).

In 2018, tracking people in physical stores is (finally!) feasible

Customer Experience is mapped by Location & Time (x,y positioning) of people in motion, which generate the insights for Behavior Analytics.

To increase Profit and Conversions, we design Optimization Projects (A/B Testing in Physical Stores) that end with ACTION!

In-Store Customer Tracking

Here’s In-Store Customer Analytics you can start doing TODAY:

  1. Identify Purchase Points
  2. Map Path Trajectory
  3. Quantify Local Demand
  4. Prevent Friction Points
  5. Measure Engage Time
  6. Monitor Service Time
  7. Optimize Product Positioning

Because People Tracking Technologies are becoming cost-effective – with Good Enough Accuracy – we get data-driven insights from quantifying the behaviors of customers in physical stores.

By tracking (anonymous) customers, we translate the In-Store Buyer’s Journey into retail metrics and actionable insights.

Before we dive…

Some basic concepts:

Accept the Offline Customer Experience is NOT Online Shopping Behaviors

Despite the media hype, there are distinct differences between online and offline shopping behaviors.

Here are the Top 3:

  1. Linear vs. Parallel Path: Online behaviors are dictated by how people move from webpage to webpage (Impressions and Click). Reality is a LOT messier. STOP thinking about Linear Customer’s Journey Maps.
  2. Sole vs. Social Visit: Online, people buy for themselves or for someone else. Regardless of the influences on the decision to buy, online visits are designed for sole individuals. Physical Stores are Social.
  3. Conversion vs. Profit Goals: Conversion Rate Optimization (CRO) is a well-defined discipline for online commerce. Digital products are hard to produce but easy to replicate – that’s why CRO exist. But physical locations have has unique costs and constraints. Conversion is super important. But Profit is the secret to physical stores.

Our research found that one consequence of the new world of marketing complexity is that more consumers hold off their final purchase decision until they’re in a store – McKinsey & Company

Define Your Data Analytics for Personalized Marketing or Anonymous Behaviors

Another source of confusion is the difference between Personalized and Anonymous Customer Data.

Location Marketing deploys location-based data that is tied to a specific individual. Due to privacy laws, this requires the customer’s opt-in.

The catch, of course, is that the opt-in is often done to third parties such as Apple, Facebook, and Google. Even Snapchat.

Most retailers are not so lucky…

Location Analytics depends on Anonymous Data.

We don’t need to know the identity of a person, just to track anonymous objects. The behaviors – captured by people tracking technologies – gives us actionable insights.

With People Counting we can quantify demand for the store, compare stores, and proximity traffic. And in 2018, the advances in tracking technologies allows us to move inside the store.

To optimize store performance you don’t need personalized data.

Moreover, sometimes working with the customer’s purchase history is distracting at best, and destructive at worse.

Because our objective is to learn what happens before the payment.

While retailers digitally interact with customers during the in-store visit

In Behavior Analytics for Physical Stores we focus on:

  • Customer Engagement
  • Retailer’s Calls to Action
  • Preventing Friction Points

To optimize the physical store, these actionable insights come ONLY from Anonymous Analytics of Customer Behaviors

Even at the highest levels, retailers don’t really understand either the rewards or the risks associated with anonymous vs. non-anonymous customer geo-location data. – RSR Research

Analyze Customer’s Experience as UX Analytics

Customer-centric retailers design stores based on the principals of User Experience.

In In-Store KPI frameworks, we first need to clarify our definitions.

“Customer” and “Shopper” are generic terms in retail. In Behavior Analytics for physical stores, we need to better way to classify people.

Here the top 3 In-Store Customer Classifications:

  • Visitor: a shopper in the store, from Entry to Exit points
  • Consumer: end user of the product (in/out of Shopping Groups)
  • Buyer: the person who pays (sole shopper or in Buying Groups)
Shopping Group [Behavior Analytics Academy]
Shopping Group [Group Tracking Case Study – Behavior Analytics Academy]

How do you know how to classify a person?


It’s a judgment call

Based on data from tracking people

And Advanced Analytics helps

We can create a context for the customer experience and behaviors

For example:

  • Point of Sale (POS): the customer’s purchase history points to departments and products that may be of interest to a specific person.
  • Order Management: the tracking of products, categories, and brands points to popular purchases, specialized products, and odd items.
  • Assortment & Layouts: gives an indication of what is available for sale in the local store, and where is it located inside the store.

Each time a consumer is exposed to improved shopping experience, their expectations are reset to a higher level of experience. – Brendan Witcher, Forrester

Next, we corrolate the customers and product data into the “actual” motion of people inside the store.

Tracking People with Behavior Analytics

How do we track a “behavior”?

To start we need a technical definition.

“Behavior” is quantified by Location & Time Position of Objects

Assuming we can identify and exclude the onsite employees, we track the (anonymous) customers by their activities in departments, aisles, and in front of products.

Regardless of accuracy, we can categorize people tracking by:

  • Location (Virtual Zone): The virtual location of the object is defined as a relative position within an image, layout, or geo-location area
  • Time-Based Tracking: How a people tracking solution defines Time impacts the way we design the KPI Framework
  • Moving vs. Static Objects: Objects are tracked within a virtual area, in motion across virtual zones, or as static positions

While the business benefits vary, the following frameworks can be deployed regardless of the technology, solution, and data quality.

Now we’re ready for Optimization Projects

Let’s dive in –


The first optimization project is actually the “one before the last step” in the Purchase Funnel.

When retailers think about the last step in the Path to Purchase, they talk about the Checkout Process.

While the Checkout is the point where money changes hands, the decision to buy occurs inside the store. The location where people make the decision to buy – inside the store – is the Purchase Point.

Purchase Point refers to the “Location Position” the customer decided to buy the item

Offline Purchase Points are distinctly different from Online Purchase Decisions. Even senior retailers can be confused about the differences between the checkout process and purchase points.

Purchase Points [In-Store Tracking | Behavior Analytics Academy]

Here are some examples:

  • In supermarkets, Purchase Points occur when a customer puts an item in the shopping cart
  • In apparel stores, often the Purchase Point occurs when a customer tries the shirt either in front of a mirror or inside the fitting room.
  • In digital fitting rooms, Purchase Point and Checkout are merged into a single step because the customer can do both with Smart Mirrors.

Bottom Line: Purchase Points pinpoint the WHERE and WHEN customers decide to buy a specific item. By mapping the path before and after a Purchase Point, we gain more clarity to the WHY people buy.


The In-Store Customer Journey cannot be described in a direct linear map, but we can do Path Analytics.

In-Store Path Analysis is about trajectory:

  • Entry: How people entered a specific virtual location
  • Exit: How people left a specific virtual zone
  • Frequency: How many times did a person stay/pass a particular area

Path Analysis provides the “Local Traffic Patterns” or “Local Demand”.

People Tracking Path Analysis

Here are 3 cases for Path Analysis:

  • Store Layout: Do people move around the store as you planned? The IKEA store, for example, take has a predefined shopping track. Do customers follow the predefined path? I usually don’t. Do you?
  • Department: Which retail sector is challenged in 2018? Department stores. Well… even small stores have departments – just think about Women or Men (Gender) or Kids Section (Shopping Groups), and so on.
  • Islands: In-store design we used to talk about Aisles. Today the topic is Islands (cases or displays in-store).

If you watched over 1,000 hours of video like me, you know that people always surprise you. Always!

Bottom Line: Path Analysis paints a picture of how people move inside the store, identify normal and outlier behaviors, and add information on WHY people searched for a specific product.

Once we mapped possible in-store customer flows (what “we think” customers do), we turn to specific TouchPoints


Occupancy is one of the most versatile and least appreciated metrics. It is used in workforce analytics, path analysis, and friction points. A practical analytics is to use Occupancy to define “Local Demand”

Occupancy counts the number of customers staying in a (virtual) zone, per period of time.

Local Demand” includes this analytics:

  • How many customers visit a virtual zone, per period of time?
  • Is the number of “Actual Local Visitors” above, below or normal?
  • What is the “Optimal Local Demand” (in context to conversion) for that particular virtual zone?

Occupancy is an ideal metric for quick A/B testing:

  • On special events (i.e. July 4th weekend) does demand changes for a particular product (i.e. meat and corn)?
  • What is the impact of assigning an expert associate (i.e. in painting) to work as a cashier (to speed the checkout during heavy traffic)?
  • What is the relationship between traffic and engagement? For example, if we put a coffee station in the high traffic aisle, will people stop?

Bottom Line: Occupancy is a versatile metric to test for “Local Demand”. It’s used in quick A/B Testing of core assumptions and designing layouts.


Abandons, bottlenecks, and checkouts create obstacles to the shopping process and the customer’s experience.

The most understood point of friction in the store is the Checkout.

With Queue Management, Kroger Supermarkets reduced their average waiting time from 4 minutes to 30 seconds and increased sales by 1-2%.

And, of course, there is Amazon Go

Always migrate your audience to the path of least resistance. Kintan Brahmbhatt @ Amazon

Abandon Behaviors consist of:

  • Leaving Activity (i.e. people leaving the line to the checkout)
  • Not Starting Activity (i.e. not using counter service in lunchtime)
  • Not Entering the Store (i.e. not entering bank on the 1st of the Month)
  • Not Returning to Store (i.e. people really remember “bad service”)

Bottom Line: Preventing friction is currently a much-discussed topic in retail innovation. But the solution is not technology. Many times retailers can avoid frictions just by paying attention to policies, procedures, and process.

Now that we identified the Path to Purchase and we quantified the TouchPoints, we track customers behaviors.


There are 2 dimensions in the analytics of Customer Engagement:

  • Tracking shopping behaviors within a virtual zone
  • Personal tracking of a customer across the store

The definition of Time-Based Metrics may vary, but the time a customer spends in a particular location indicates engagement.

Here are some case studies:

  • Engagement Threshold: In an apparel store, the engagement threshold for black pants [on a flat display] was over 20 seconds. For premium-priced shirts, the threshold was 40 seconds.
  • Optimal Engage Time: In a beauty store, the optimal time in front of a product display is between 15 seconds to 90 seconds. In the makeup counter, an optimal time is above 60 seconds.
  • Engage Time & Path Analysis: In a supermarket, customer behaviors changed per path. In other words, Engage Time depended on previous engagement behaviors.

Bottom Line: Time-Based Metrics are ideal for quantifying the Buyer’s Journey & Customer Engagement. The Optimal Engagement Time depends on the product positioning and customer behaviors.


What’s the value of a “great” salesperson?

If you ever walked into a store with the intent to spend $20 and left with sales of over $1000… you know what I’m talking about…

Telecom companies, for example, have an Optimal Service Time to sell their phone and internet plans. And they design their concept stores accordingly.

Service is a critical component in the success of physical stores. Just ask Best Buy.

Where to “measure” Service Time?

  • Structured Customer Service, such as counter service in a deli
  • Roaming Sales Associates, such as selling jewelry or electronics
  • Ad-Hoc Encounters, which are typically found in no-touch dept stores.

Bottom Line: A variety of technologies provide insights into Service Time. Together with Service Intensity and Service Productivity KPIs, Service Time plays an important role in analyzing the power of sales associates.


What can retailers do today & quickly?

Quick Wins come from Product Positioning!

Online CRO (Conversion Rate Optimization) dedicates much to the optimization of Calls to Action (CTA).

In physical stores, there are no red buttons that customers can click.

Instead of linear concepts of step-by-step frameworks, we work with real-time data on planograms and pricing.

Some studies on Product Positioning:

  • Clearance Racks: Many Merchandising Mavens think that stores follow their guidelines. Tracking customer behaviors around clearance racks are eye-opening to retail executives who care about profits.
  • Staff Expertise: In certain products – such as jewelry, electronics, and even construction – the availability of an expert associate makes a huge difference in sales
  • End Cap Positioning: Brands tend to pay extra for positioning their products in high-traffic end caps. Here’s a hard truth – high volume of traffic does not necessarily mean customer engagement.

Bottom Line: Optimization Studies on Product, Promotion, Price, and Place can generate a spike in purchases. This is data-driven analytics for physical stores.

It’s not enough to know what people needs and desires are. We also need to know how that interacts with the store’s environment – Ted Burks (YouTube)

The Advantages of Anonymous In-Store Tracking (& Dynamic Customer Journey)

Regardless of the tracking technology and solution provider, anonymous in-store tracking brings physical stores to the forefront of customer analytics.

In Behavior Analytics, we merge and adapt Behavior Science, Online CRO, and Location Analytics to optimize the retail store for conversions and profit.


Click here to subscribe

Ronny Max

Ronny Max is a recognized author, expert, and speaker. She is the Principal of Silicon Waves, which provides C-Level Consulting on In-Store Optimization, Location Analytics, and People Tracking Technologies. In 2015, Ronny served as the retail expert in Stanford University Vision Project. In 2017, she founded Behavior Analytics Academy.