My neighbourhood supermarket is open 100 hours per week. Traffic peaks at about 5 pm and the average store visit lasts just 10 minutes. That’s 10minutes for your product to get noticed, found, considered and purchased. The aisles are narrow and don’t invite leisurely browsing and, for a small store, the assortment is pretty good, so there’s plenty of competition.
As a professional FMCG company, you have planned for this. You have worked out your “perfect-store” plan for how to price and display your products in the very best way to attract the attention of the time-poor shopper.
As a sales leader, your job is to make sure your plan gets implemented in-store for the whole 100 hours of every trading week. So naturally you have a set of retail KPI metrics and knowing what you would like to track is not too much of a problem. The measures fall into 3 categories, right place, right price, right promotion.
|RIGHT PLACE||RIGHT PRICE||RIGHT PROMO|
|Is my product available and positioned correctly?
– Out of stocks
– Position on Shelf
|Is my product priced correctly?
– Price gap versus competitor
Are my promotions deployed correctly?
– Distribution of secondary displays and promotional materials
– Promotional price
– Distribution of competitor promotions
Traditional audits: accurate and credible but not so actionable
The harder part is getting the data you need at the level of accuracy, timeliness and specificity you need to make it actionable…without breaking the bank. Traditionally, sales teams have turned to retail audit companies (think Nielsen, IRI) for this. This retail data scores well for accuracy and its independence means it has credibility within your business and with retailers. The in store data collection is done by professional fieldworkers in a representative sample and can be easily combined with EPOS sales data enabling you to evaluate what works. Quite likely your Insights team already subscribes to some of what you need. It scores less well on timeliness and typically you can’t get store specific information. So you may find out you have some problems to fix but you don’t know where they are. So it’s good at giving you the big picture retail insight on how effective you are but falls short as a tool to help you identify how to improve and where to take action.
with Image Recognition & AI
|Crowd collection with Image Recognition and AI|
|Key characteristics||– Accurate and credible
– Expensive (but your Insights team may already subscribe to some)
– Not store specific
– Typically not quick enough
|– Store specific
– Not independent – credibility can be questioned
– Collection uses up selling time
|– Store specific
– Cost effective
– Can help route planning and visit priorities as well as in-store actions
|Credibility (accuracy & independence)||High||Medium||High|
|Actionability (timeliness & specificity)||Low||High||High|
Tech helps Sales Forces see more
Technology is bringing new possibilities. Your teams can now use smart-phones/tablets and apps to collect data. Image recognition technology and AI are combining to enable quick and easy data capture in store. Over the last few years the reliability of the technology has improved and a number of FMCG sales forces are using it to collect in store data: the rep takes a picture of the shelf, software uses AI to compare the picture to the “perfect store” model and identifies what needs to be fixed. There are advantages and disadvantages to this model:
- Data collection tools can save time, and even plan your routes for you
- With the right tool, data can be augmented with crowdsourcing tasks
- Data credibility can be questioned – your team are “marking their own work”
- Data collection uses up time that could be spent fixing the issues
Using the crowd to drive speed and efficiency
How can CPGs gain full visibility of their stores? Enter the crowd. An innovative wave of companies is putting image recognition technology into the hands of millions of shoppers. Done well, crowd-sourcing can be a real win-win: citizens can make a little extra money for data collection “missions” and the corporation gets efficient and flexible data collection with a wide coverage of stores. The process is driven by an app that delivers the mission instructions and communicates with a governing system that ensures quality and manages payment. A typical mission would be to visit a store and take a picture of the Hair Care fixture. The image is transmitted back to base and the AI does the rest. The processed data is delivered to the sales reps and management via an app and a dashboard and is made available immediately making it very actionable. This approach to gathering field insight has the following key advantages:
- It can be used to assess which stores are a priority to visit as well as what needs attention when the rep arrives
- Sales teams are able to focus on what they are best at; selling and merchandising
- It is more cost effective than using an audit company
- The data is independent
Proper management of the “crowd” is critical to getting the high quality retail KPIs to support decisions and actions. Building and maintaining a crowd resource is not easy. Keeping them engaged but not over-using them is key, knowing how to give simple, clear instructions and operate quality controls is also vital. So there are some key questions you should ask of vendors:
- How do you recruit and reward your data collectors?
- How do you manage their work in terms of volume and quality? What automated quality checks do you have?
- How do you ensure that you comply with relevant employment laws where you operate?
- Do you comply with ethical standards such as paying a living wage equivalent?
What’s next? Technology continues to advance and we see retailers starting to invest in smart shelves and also see costs for smart packaging coming down…..but that’s a topic for a future blog. Watch this space!
If you would like to learn more about data sources and technologies for retail execution, you can read the full outlook in our white paper “Perfect Sales Execution: The ultimate guide to data sources and technologies” here.