Top 5 BI Business Benefits for Retail

Tackling Retail Business Challenges with BI

Retail businesses face a lot of challenges that threaten efficiency and profitability but did you know that many of them can be significantly improved or even eliminated by an effective Business Intelligence and Data Analytics solution?

We’ve decided to focus on just 5 of them:

Benefit 1: Improved Demand Forecasting
Benefit 2: Better Visibility of Sales Performance
Benefit 3: More Efficient Stock Management
Benefit 4: Increased Understanding of Customer Behaviour
Benefit 5: Better Visibility of Promotion and Pricing Impact on Profitability
How Do We Do It?
Summary

Retail Business Data Challenges

The challenge that troubles you the most will ultimately depend on the size and character of your business, and where you’re starting from. When there are no specific Business Intelligence skills in-house, BI projects may exist, but they can be disjointed or implemented in isolation from one another.

Many retail businesses already have good reporting but might be struggling to get the reports produced in time for key meetings where they’re needed for decisions. This is often the case when a lot of different purpose-built applications are being used across various business sections, resulting in isolated and disjointed sets of data or data silos. This makes it very hard to get a total view of the whole business.

Another issue is the frequent stand-offs between IT and Business with business needing quick results to drive better decision making and IT needing to ensure that all tech stack solutions follow their data strategy guidelines. IT will be thinking about issues like whether platforms are supportable, whether they adhere to security and data governance policies, and whether they risk driving operational systems into overload with excessive calls to retrieve data.

When departments work independently of IT data people often use Excel and other tools to bring the data together. Users locate the data they need, exporting it from various live systems to CSV files and cobbling it together through a mix of supporting lookup and mapping tables. But data is out of date the minute it is exported and the grain and format of the data is very much dictated by the particular tool and even the person defining the export. This can mean lack of consistency.

There is often minimal or no cross department communication to define common requirements and definitions, resulting in much confusion. Core KPIs (Key Performance Metrics) built independently by different business units using complex Excel formulae might be different from one department to the next.

For example, how many different definitions do you have across your business for;

Sales revenue
Like for like store sales
Gross margin
Customer conversion rate
Inventory turnover
Sell through rate
Customer retention rate
Net promoter score
Customer acquisition cost
Out of stock rate

With so many different performance metrics and data dictionaries describing your common business terms, people risk coming to meetings with completely different sets of figures. That leads to confusion, misunderstandings, mistrust and poor decision making.

So let’s take a look at 5 key areas where organised data systems and a defined Business Intelligence plan can help you.

Benefit 1: Demand Forecasting – Required data Inputs

Demand forecasting helps to ensure product is in the right place at the right time, so that when a customer wants a product they can buy it. All retail businesses want to ensure that sales equals demand. If sales are less than demand for whatever reason, a customer can’t buy what they want and a sale is lost. Perhaps it was in the warehouse, but not in the store, or maybe it was in a different store. Perhaps it was in stock but was damaged. Perhaps it is in stock, but we don’t have enough of them.

For accurate demand forecasting we need to have visibility over many things, including data from some or all the following:

Historical sales data – units sold, revenue, daily/weekly grain
Product attributes – category, brand, size, colour, style, price, lifecycle (core, new or discontinued)
Pricing and promotions – list price, sale price, promotion type, duration, coupons, bundles, offers, loyalty offers
Inventory and availability data – on-hand inventory, in-stock/out-of-stock flags, backorders, substitutions
Channel and store data – store location, layout and size, channel type, local trading trends
Customer loyalty – repeat v new customers, basket content, loyalty segmentation
Calendar and seasons – holidays, events, payday, week and day of year impact, retail events such as back to school or university, Christmas
External data – weather, local events, economic indicators

Some of the data which influences demand forecasting are external such as weather forecast information. But much of it comes from different business units such as:

Merchandising
Operations – Store and Supply Chain
In-store Operations
Supply Chain, Logistics and Warehousing
Sales and Customer Service
Marketing and Advertising
Finance and Accounting

To successfully improve the accuracy of our demand forecasting, we need data from all these business areas.

Demand is impacted by:

Marketing, advertising and social media campaigns which result in increased purchases
Product being in stock in stores or online for customers to buy when they want it
Seasonal events such as Easter and Christmas driving higher sales in chocolate, flowers, toys and other gifts
Holiday seasons driving higher sale of warm weather or cold weather clothes, camping gear, outdoor gear
Pay day dictating when consumers have money available to buy the things they want
Weather events such as heavy rain and heatwaves driving purchases of things like umbrellas, wellingtons, waterproofs, sun cream, fans, air conditioning units
Local events such as roadworks and festivals driving sales down due to access issues, heavy traffic, parking issues. Or driving sales up due to increased footfall
Store location and layout affecting the consumers experience and impacting whether they buy or don’t buy – can they find what they need? Was the shop warm? Did the shop smell nice?
Customer experience in a Bricks and Mortar store, or in an online store. Good service can drive an increase in sales. Poor service may result in no sale.

To maximise profitability, we need to bring all the relevant data sources together so that we can link our historic sales data to seasonal patterns, marketing campaigns, social media trends, store condition and user experience.

A well-designed and well-implemented BI solution can bring together all the relevant demand forecasting data to drive more accurate demand forecasting in a timely manner.

Benefit 2: Better Visibility of Sales Performance

Sales Performance encompasses:

What is selling?
Where is it selling?
Why is it selling?
How profitable is it?

Some of the core KPIs that drive visibility of sales performance in the retail world include:

Top line sales – overall performance and sales growth
Total Sales Revenue – by region, channel, store, product, product category, by year, month, week, day
Like for like sales growth – exclude outliers such as closed or new stores, new or discontinued products to get true growth or decline
Sales per square foot – Store and location productivity
Sales per employee - Staffing productivity and levels
Conversion rate – transactions v footfall, online orders v online sessions
Average Transaction Value (ATV) – Average item value
Average Order Value (AOV) – Average order value
Units per Transaction (UPT) – pricing v volume relationships
Sell Through Rate – Units sold v units received
Top/Bottom Product and Category performance – sales and growth contribution to identify popular stock and dead stock
Channel and Omnichannel performance – store v online v marketplace, channel shift, buy-online-pickup-in-store (BOPIS)
Markdown rate – discounted sales v total sales
Promotional Sales Uplift – sales v sales baseline
Gross Margin – by product, channel, store
Contribution margin – sales minus variable costs including discounts, fulfilment and returns

Once again, a lot of data is needed to calculate these KPIs.

Sales data is generated in a number of operational systems:

POS
E-commerce
Marketplace

It is also fragmented across different regions and countries, who have different currencies and different tax rates.

In many cases one business may have acquired another, and with that may have inherited numerous variants of POS and e-commerce platforms and systems.

This can make it very challenging to bring all sales related data together.

Not only will we need to bring all the data together with all the challenges that relating data from different systems brings, but we will also have to handle:

Choosing a base currency to work in
Handling foreign currency exchange rates – average day rate, end of day rate, consolidated rate
Handling tax – deducting tax at the right tax rate where systems expose gross figures rather than nett figures

A well-designed and well-implemented BI solution can bring all sales data into a single view enabling sales to be tracked by channel, store, product and time. This in turn allows us to create dashboards that answer the key questions every business wants to know, such as;

Are we growing?
Where is growth coming from?
Are we profitable?
What does our sales trend over time look like?
What does our channel and store performance look like?
Where do we need to focus our attention?

Benefit 3: More Efficient Stock Management

Stock Management or Inventory Management is about ensuring stock is in the right place at the right time, and that we are not overstocked. Challenges we face with stock and inventory management include:

Overstocking – Inventory turnover too slow, aging stock that becomes obsolete, unsaleable or must be marked down
Understocking – Product not available for demand, missed sales
Inaccurate demand forecasts – results in overstocking or understocking
Replenishment levels not adequate for demand – fixed ordering quantities regardless of season, campaign or trends
Stock in the wrong place – disconnected warehouse, store and e-commerce inventory
Poor inventory allocation – equal quantities to all stores rather than following store trends, newer stock moved before older stock leading to shelf-life challenges
Shrinkage, Loss and Inaccuracy – reduced stock items due theft, damaged goods, misplaced goods and administrative errors
Cash flow impact – cash tied up in stock that isn’t selling at the expected rate

Overstocking results in cash being tied up in stock and potential cashflow challenges, where understocking means we cannot meet demand. You can see there is an overlap here with demand forecasting, as we need to know what the likely demand is to have a chance of getting stock management right.

Not only is understocking going to hit our immediate sales performance, it can lead to customer dissatisfaction and lack of trust, leading them to do their shopping elsewhere, not just now but for the longer term.

Theft and damaged stock not only leads to effective understocking but is also a big threat to our profitability.

Stock sitting in the warehouse when it should be in a store, or stock quantities that do not meet supply needs will lead to frustration for a store manager who cannot fulfil customer demand due to stock shortages. Where individual performance is measured by store performance this is a frustrating issue.

Stock replenishment which does not take into account regional, store and channel trends or ignores varying demand can lead to dissatisfaction amongst both staff and customers where product is unavailable when they need it.

It is extremely hard for a retail business to get all their stars aligned – that is; their marketing messaging, their store layouts, their staff training, all in such a way that wins the confidence and business of consumers. It can take a very long time to achieve. Imagine then that you discover the stock is not available when and where it is needed - all that time and money has been wasted. A sale is not real profit until it is sold, delivered and paid for.

Common KPIs for driving inventory visibility and decision making include:

Inventory Turnover – cost of goods sold v average inventory
Days Inventory on Hand (DOH) - average inventory cost v average daily cost of goods sold
Weeks of supply (WOS) – Inventory on Hand v Weekly Demand
Sell-Through Rate – units sold v units received
Inventory productivity (GMROI) – gross margin v average inventory cost
In-stock Rate – Time product in stock v total time in period
Stockout Rate – Out of stock events v total product/store combinations
Fill Rate – unites delivered v units ordered
Inventory Record Accuracy (IRA) – (System quantity minus physical quantity) v physical count
Shrinkage Rate – (Book inventory minus physical inventory) v book inventory
Aging Inventory % - number of products older than x days (define x) v Total inventory
Markdown Rate – Markdown sales v Total Sales

So, to build a useful Inventory dashboard we would need to pull together demand forecasting data, sales data, and inventory data. We need to take multiple data sets and integrate them, relating them by location, product and time.

A well-designed and well-implemented BI solution can do all this, bringing inventory and demand data together to provide insight and allowing us to answer these questions:

How much stock do we have, is it adequate?
Where is our stock?
Where and what is the demand?
Is our stock in the right place?
Are we turning product over fast enough?
Do we have any aging stock that needs to be shifted?
Are we replenishing stock effectively?
What is the value of our stock?
Are we marking down too frequently due to overstocking?
Are we seeing restocking due to theft and damage?
Are we losing sales and profit due to poor inventory efficiency?

Benefit 4: Increased Understanding of Customer Behaviour

As a retailer it is important to understand your customer.

Who are they?
How do they shop?
Where do they shop?
When do they shop?
What products do they buy?
What products do they buy together?
Do they interact with campaigns?
Do they take up offers?
Are they repeat customers or one off?

Understanding who our customers are and how they behave is key to successful retail businesses. If we know our customer’s shopping habits, we can fine tune our marketing, forecast demand and stocking plans accordingly.

We can anticipate their needs based on historic behaviour and market intelligence gathered from customer surveys, mystery shopping and market research campaigns. This in turn can help us to personalise our sales and marketing approach to our customers.

Customer Intelligence can be gleaned from a variety of sources:

Point of Sale
Loyalty Schemes
CRM Systems
E-commerce Platforms
Mobile Apps
Digital Marketing Platforms
Customer Support and Services Systems
Instore Technology – footfall counts, dwell time, heatmaps, queue times
Returns
Payment Systems
External Data – market research, location, weather, events, demographics

Common KPIs for measuring customer behaviour include:

Customer count – active customers
Customer Growth Quality - New v Returning Customer Rate
Purchase Frequency – total transactions v total customers
Average Transaction Value (ATV) – total sales v transactions
Units per Transaction (UPT) – Units sold v transactions
Store Conversion Rate – Transactions v Footfall
Online Conversion Rate – Orders v sessions
Customer Lifetime Value (CLV /LTV) – average order value x purchase frequency x customer lifespan x margin
Revenue per Customer – total revenue minus (customer cost of goods sale plus discounts)
Customer Retention Rate – repeat customers v customers in prior period
Churn Rate – 1 minus customer retention rate
Repeat Purchase Rate – customers with 2 or more purchases v total customers
Loyalty Program Participation Rate – loyalty customers v total customers
Channel Mix per Customer – single channel v omnichannel
Net Promoter Score (NPS) - % promotors minus % detractors.
Return Rate – returned units v units sold

Other customer behaviour we might want to pay attention to is how they speak about us to others either online or in person. Detractors and Promoters are defined by segmenting customers into groups of Promotors (9-10), Passives (7-8) and Detractors (0-6) based on the question “How likely are you to recommend our company/store product to a friend or colleague?”

So with all these metrics in the mix we need to grab data from a wide number of different sources from across departments and then relate them to a customer, product and point in time.

A well-designed and well-implemented BI solution can do this, bringing all this customer behaviour and demographic related data together to provide insight into:

Where we can sell more to existing customers
Optimising store layouts with associated products
Reducing abandoned baskets
Suggesting other products when online shoppers checkout
Improving personalised offers
Growing repeat customers
Reducing customer churn
Optimising pricing and promotions
Improving product offerings
Ensuring stock is in the right place
Linking NPS, returns and complaints to customer behaviour and experience
Measuring campaign success and ROI on marketing and operations

Benefit 5: Better Visibility of Promotion and Pricing Impact on Profitability

Promotions and campaigns are frequently launched on gut instinct, with no data intelligence behind them. But if you can back up decisions on what to promote, who to promote to, how and where, then you can say goodbye to random acts and start to home in on data driven campaigns.

In order to implement cost effective promotions we need to understand the Return On Investment they deliver:

Can we measure the effectiveness of a promotion?
How responsive is a customer to price changes?
What is the impact of discount on profitability?
What is the impact on demand?
What is the impact on sales growth?
What is the impact a new product launch has on cannibalisation – does a new product reduce the sales of older products?
Are stores delivering promotions correctly?
Are promotions increasing volume sales but eroding profit margins?
Are the right customers being targeted for promotions?
Are promotions personalised adequately for audiences?
Do demand forecasting and inventory levels account for promotions?

Poorly planned promotions could very easily lead to eroded profits, empty shelves and dissatisfied customers.

A successful campaign is much more than getting a record-breaking number of likes, clicks or products in baskets. Increased demand must be co-ordinated with supply chain so there is enough stock available. Without it customers unable to access the offer will be unhappy, leading to bad social media and negative feedback on ecommerce platforms and ultimately damaged reputation.

By collating data from all related business areas, such as demand forecasting, inventory, sales and marketing it will be easier to have visibility over whether all the requirements are in place before a promotion or campaign is scheduled.

A successful promotion also needs customer intelligence regarding demographics, and data on historic campaign performance to help us avoid patterns that have failed before.

Core data sources that provide valid insight to drive successful promotions and campaigns include:

Historical sales performance
Historical Promotion Performance
Product Pricing and Margin
Inventory and Stock Position
Customer Behaviour and Segmentation
Loyalty Programs
CRM Systems
Digital and Omnichannel Behaviour
Store and Location Performance
Calendar and Seasonal Performance
External Data Sources – weather, local events, market trends
Marketing Channel Performance – impressions, clicks, email opens, cost per acquisition

Here is another clear example where multiple data sets need to be converged and multiple business units need to collaborate to drive accurate data driven insights and business success.

The most effective retail promotion campaigns will bring together expertise on customer demand, inventory risk and profit margin protection.

A well-designed and well-implemented BI solution can bring together all the intelligence data needed to run a highly successful promotion. It can also provide insight into:

Who to target
What to promote
When to promote it
How much to discount or bundle
Where to promote

How Do We Do It?

How Do We get a True 360 View for a Retail Business?

For a retail business to get a true 360 view of their customers, products and the performance of their core business functions, collaboration is key.

Bringing Data Together

Being able to bring together all the relevant data that relates activities such as sales, returns, campaigns, stores and stock back to each other, is essential when running a successful retail operation. It enables us to calculate a true profit margin including hidden costs as well as the more obvious ones, such as product purchase cost. With rich data we can bring in a number of factors such as supply chain costs, value of returned products and bad debts for a more complete view.

Trying to do it manually is incredibly time consuming and it’s often simply not possible to pull all the data together in time. This means information which could make a difference is often missed and then profitability takes a knock.

Bringing Departments Together

Enterprise BI projects should be a collaboration across the whole business. This enables a common, unified data dictionary to be pulled together ensuring that all business units use the same terminology and the same definitions for core KPIs.

It is also an opportunity to identify data sources common to multiple business functions so that data collection and integration can be done once and the results shared across the business.

Collaboration means that a single source of truth can be created, a single business semantic model that can be used by all departments, enabling data from different systems and sources to be related and presented in a single dashboard and contributing to critical KPIs that can drive better and more timely business decisions.

Integrating Business Data

To achieve our goals we need to do the groundwork that allows us to relate many data sources together e.g. sales data with marketing data and do it consistently. Where different systems have different methods for identifying customers, products or stores you could have problems – for example different IDs, codes, name spellings or multiple records for a single customer.

We need to be able to map customers, products and dates across the whole of the underlying data estate so that we can show one customer with all that relevant information attached to them – from sales to sent campaigns etc

We achieve this by creating:

a single customer entity which contains all the relevant attributes from every system we are linking the customer to
a single product entity which contains all the relevant attributes from every system we are linking the product to
a single date entity that contains all our date extractions required for reporting such as year, quarter, month, week, season. We would also add attributes for multiple calendars such as financial year, financial year month, financial year week

You can see why we need collaboration! We have different data owners, different system administrators, different definitions of business objects and different key attributes that specific business units require for their own reporting. But we also have different business insight and experience to bring meaning to business data and KPIs.

Conforming data from across the business will often be achieved by creating shared repositories such as lake houses, data warehouses, and shared semantic models/data sets that can host the single data dictionary, single source of truth, and be used for all dashboard and reporting needs. Click through to learn more about these topics.

Summary

A successful BI solution will:

Turn reactive decision making into proactive decision making
Turn data rich, insight poor businesses into insight, data driven decision making businesses
Turn fragmented silos of business data into a central single source of truth
Drive improved customer and staff satisfaction
Drive increased profitability

It is by no means a simple journey to get to an enterprise-wide BI solution, but if you start with that goal in mind, you can break your journey into smaller chunks, designing and building your solution in readiness for expansion into all business areas.

We suggest you identify one business area to start with - perhaps Sales Performance as that is key to all the areas we have been talking about here, then build it for sharing and expanding by integrating further business areas into one solution rather than separate disjointed BI projects.

If you would like to talk to us about how we have helped our Retail clients do this please get in touch. A successful business intelligence and data analytics solution is within your grasp!

Blog Post