How to Choose an Image Recognition Vendor: 5 Criteria to Keep in Mind

In modern retail, data is king. Vast swathes of customer and product information can now be harnessed to improve the in-store experience and make operations more efficient.

Brick-and-mortar retailers may have been latecomers to the game, with many already having fallen victim to the so-called “retail apocalypse” after failing to innovate quickly enough. But finally, retailers and brands are waking up to the reality that without embracing a data-driven future, there’s no future to be had at all.

Image recognition technology may be the most significant of these new advances. Digitizing the laborious process of shelf-checking saves valuable time and effort that can be better spent on building relationships with store teams and negotiating better placements.

It is definitely worth investing in, but with the market becoming increasingly crowded, it can be difficult to know which image recognition vendor is the right one for you.

Here are 5 of the main criteria we’d recommend to keep in mind when making your decision:

Short on time? Watch this handy video blog instead:

#1 - Speed

For busy field reps, auditors, and brand ambassadors, time is of the essence. With so many sites to visit, low-value manual tasks like counting each individual product on a shelf are a real drain on their already limited time and resources.

A good image recognition solution will work in real-time, so that the user doesn’t have to wait around for results and can focus more of their attention on value-added tasks such as negotiating more shelf space, fixing planograms, and covering more stores in less time.

Ideally, store visits should take no longer than 30-40 minutes, a goal which is much more achievable with image recognition.

#2 - Efficiency

The problem with some new technology is that its very “newness” ends up being more of a hindrance than a help.

Remember when self-checkout machines were first introduced? Nowadays we’re all used to them, but initially it was a mess!

Shoppers just couldn’t get to grips with the technology, and store teams spent more time helping people with “unidentified items in the bagging area” than they would have done just scanning products the old-fashioned way.

To avoid this kind of friction, image recognition technology should be easy to integrate into current routines (price tracking and management, for example) rather than requiring a complete overhaul of store operations.

#3 - Deployment & Scaling

Image recognition is the ideal technology for huge companies that stock hundreds or even thousands of products from countless different brands, as it provides a simple, headache-free way to track those products across your entire network.

But this kind of technology is not always easy to deploy at scale. Make sure that whichever vendor you choose is able to handle, and provide examples of, large-scale projects - across countries and perhaps even continents - without any issues.

#4 - Corporate Social Responsibility

As far as we’re concerned, real image recognition technology is artificial intelligence embedded with machine learning.

However, there are some companies who instead use low-skilled workers in developing countries, who look at each image manually in order to gather data. When making your decision, you might want to think about the ethical implications that come along with that.

#5 - Pricing

Retailers and brands are already having to achieve more with less these days, so budget will undoubtedly be a major factor in the decision-making process. With financial restrictions in mind, it’s important to consider the pricing model of your provider.

A lot of companies offering image recognition software charge per image, meaning clients may have to be careful about how many pictures they take. Not ideal for a system that’s meant to make life easier for the client.

If you have a lot of shelves to cover in a lot of stores, a vendor that charges per user - with unlimited pictures - could be a far more cost-effective option. It’s certainly something to bear in mind!

---

Companies like Sanofi are using YOOBIC’s image recognition technology to get results in less than 60 seconds, with over 96% accuracy. Find out more by requesting a free, personalized demo below.