By Andrew Dawson
In today’s digital world, where ecommerce platforms are becoming the de facto standard for many shopping experiences, the retail value chain has access to more data than ever before.
The explosion of business intelligence and visualisation tools, and the creation of data lakes to pool information and optimisation structures to mine this data and represent it in a format that makes sense, have led to a plethora of tools and infrastructure and a veritable river of data, reports, and dashboards.
The challenge is that often, the people who need to gain insight from the data are not being given the right information in the right format or in an easily accessible manner to help them make the right decisions.
The key to success is using these platforms to deliver intelligence to key individuals who can effectively influence decisions, in a way that helps them to resolve any issues the data is highlighting.
Right data, right time, right place
The retail space is by its very nature fast paced and requires decisions to be made quickly, for better or worse.
However, key account managers, sales managers and customer managers need very different information to improve their decision-making ability.
In addition, they are often too busy to log in to a dashboard and pull a report, never mind trying to decipher what various dashboards and information is trying to tell them.
While predictive analytics is powerful, it is often not used to its fullest extent because the right people are not receiving the right intelligence at the right time to effectively influence decisions.
This leaves us with a conundrum in that investment into mining and visualisation tools means people are identifying issues, but they are unable to resolve the issues the data is identifying.
It is essential that we use these platforms to correct the issues and exploit the opportunities being presented, which means we need to drive actionable insight from data.
What is the data actually saying?
Granularity is key to improving insight in the retail space.
For example, Key Value Items (KVIs) may differ, and even if they are the same, they will have different rates of sale depending on the regions and communities they are being sold into.
If this data is averaged out and procurement is done at a group level, all stores will receive similar stock levels of products. However, all stores may not sell these products at the same rate, leading to potential over or understock problems.
On the other hand, if you have granular data based on individual stores, combined with line of sight into the distribution centre and what stock is sitting on the shelf and in the ack of the store, it is far easier to accurately track the real rate of sale per store, identify trends and patterns and far more accurately plan the next replenishment cycle.
Adding in a further layer of intelligence, if you can auto-generate a notification to the right individual at the right time to highlight a stock shortage and even give the number of days until stock runs out, this is an actionable insight that can drive corrective behaviour.
Measure it, manage it, make it better
Actionable analytics is where the real value lies and is the next step in leveraging the data that is flowing through the retail value chain.
This relies not on understanding key roles in the industry, and on the ability to mine data effectively, but on having the tools and capabilities in place to identify deviations, interpret them and convert them into a message that is delivered to the right person in real time, with the necessary detail, to enable them to action the insight and prevent revenue loss.
Once you have the capability to deliver the right insight to the right person at the right time, you need to measure on the actions taken, track the data to check whether insights have been actioned and then measure what the impact has been.
Once this is in place, it will spawn a whole new channel in the applied data space and create competitive advantage, improve customer service, and drive enhanced profit within the retail space.