What is image recognition?
At a high level, image recognition can be defined as the following:
“Image recognition is a term for computer technologies that can recognize certain people, animals, objects or other targeted subjects through the use of algorithms and machine learning concepts. The term “image recognition” is connected to “computer vision,” which is an overarching label for the process of training computers to “see” like humans, and “image processing,” which is a catch-all term for computers doing intensive work on image data.”
In VERY basic terms, image recognition is the act of an algorithm scanning a data set and looking for some sort of pattern and/or correlation. In this instance, the data that’s provided is an image, and the pattern it looks for is dependent on the purpose.
How does this apply to retail?
In the world of retail, image recognition is used by CPGs as a productivity tool to enable faster audits, high accuracy and boosted efficiency for their reps.
Consumer brands have always struggled with finding the right method to perfect their retail execution. Due to changing variables like size and space available within a store, it’s generally accepted that there will always be some level of disconnect between their planogram and the physical display itself.
That said, when CPGs have been known to lose up to 25% of sales due to poor execution, it’s still high on the priority list when it comes to getting it right.
Thankfully, with the use of this image recognition technology, the task of tracking their visible in-store KPIs is more accessible, efficient and actionable than ever.
How does it work?
To complete a store audit using image recognition, all the user needs to do is stand in front of their desired category and take a few pictures. The best parameters for these will be guided by visual AI cues and doesn’t need any prior training – making it a foolproof task.
Once the photos have been taken, the user can continue with their other duties in the store, such as speaking to the store manager, doing inventory checks, auditing other categories, and so on.
While they do this, the image recognition technology, which uses an algorithm know as a ‘neural network’, will work in the background to recognise the relevant KPIs.
What kind of KPIs can it track?
As long as the algorithm has been trained to do so, an image recognition tool can collect any visible KPI within a retail setting.
With over 1000 possible in-store KPIs, it’s not possible to list them all. But if a CPG was aiming for ‘perfect store execution’, image recognition could help them in tracking:
- On-Shelf Availability
- SKU Presence
- Stock Levels by Threshold
- On-Shelf Visibility
- Number of Facings
- Share of Shelf
- Planogram Compliance
- Assortment Compliance
- Promotion Compliance
- Number of Displays
Who can use it?
Thanks to the visual cues given when taking the pictures, image recognition takes the responsibility out of the hands of the user. Instead of extensive training on collection methods and how to track particular KPIs, all that’s needed to use an image recognition tool is the simple competence to follow instructions and a smartphone.
This gives CPGs the opportunity to reassess their KPI monitoring strategy – taking into account how far they want to reach, the most cost-effective field method and what type responsibilities they want their employees to have.
There are three main schools of thoughts as to who should carry out image recognition audits:
Internal sales representatives can use image technology during their store visits. The duty of data collection will sit alongside their other tasks, but because of the effectiveness of the tool used, will take much less time than other methods, such as manual auditing.
Members of the crowd, freelancers, independent auditors – no matter what they’re called, this group of people will carry out store audits on behalf of the CPG. Their only duty will be to collect in-store data, and the frequency at which they complete store visits is up to them.
Because no training is required to use an image recognition tool, the accuracy of these audits remains high.
External field agencies
Field agencies fill the gap between sales reps and the crowd; while they’re separate from the CPG, they’re still specialists.
Usually, field agencies have free range and CPGs have to trust their actions, with little control over their activities out in the field. Image recognition changes this and gives manufacturers a view into what the field agencies are doing, which creates a more reliable system overall.
Is image recognition right for me?
Part of the beauty of image recognition in retail is that it looks different from company to company.
A flexible solution at heart, it has the capability to fit the needs of both big and small CPGs. Whether they want to focus on two or three main KPIs or monitor a wider spectrum, the benefits of image recognition apply is every scenario.
So, is image recognition right for you?
If you’re a CPG that’s looking to monitor visible in-store KPIs in an efficient, accurate and actionable way, this technology should be on your radar. Because on top of that, it could help to boost your revenue numbers by 5% within the first year alone.
If you’d like to learn more about how your team can use image recognition to complete more efficient stores audits, read the full outlook in our white paper “SCANNING THE SHELVES – How Image Recognition Helps Consumer Brands To Boost Their Retail Execution” here.