News & Blog

Manu Krishna
Manu Krishna Director, Content Marketing at Trax

10 concerns about image recognition in retail and how to address them – Part 2

In our previous post, we explored the attributes essential for the best image recognition solution or product. Here we’ll look at factors that will play a key role in ensuring that you enjoy a hassle-free purchase, implementation and ownership experience.

Image recognition in retail today is at the intersection of digital business and analytics at the store shelf. This disruptive technology provides manufacturers with insight into conditions on the retail shelf in the form of a digital image, and guides field personnel and managers in near real time on how to drive greater value.

But with a proliferation of vendors and shrinking tech budgets, a well-researched buying decision is worthwhile rather than simply chasing the new shiny toy in town.

Buyer’s guide: Factors that drive impact

6. Price/Performance value: Does the product offer the best value for the price, compared to similar offerings in the market?

Deep fine-grained recognition is not easy to achieve. It often involves high computational costs, requires large sets of training data and is dependent on Graphical Processing Units (GPU) for training speed. But the benefits far outweigh the costs – just fixing out of stocks, which is the most popular use-case of IR saves millions of dollars annually. Seek evidence that the vendor has a demonstrated ability to achieve high accuracy, scale and short time to insight. Trax, for example, falls into the ‘best’ category according to Gartner’s benchmarks:

  • 95% + accuracy
  • Insights within five minutes in-store
  • Scalability above 50,000 stores
  • Reporting, query and configurable dashboards

7. Customer purchase experience: Do you feel you are buying the most optimal solution that addresses both your unique needs and your unique constraints?

Although they have detailed shipping information and point-of-sale data, brands have almost no visibility of the store shelves. The only technology alternative to image recognition is manually entering observed data into a mobile device and capturing a picture that will substantiate the observations that have been entered. However, the substitutability is very low due to the inability to deliver recommendations and KPIs on the spot. To further justify your investment, consider a well-designed pilot that compares results between similar stores using this technology and others that aren't. This will help in building a business case and calculating actual ROI.

8. Customer ownership experience: Would you be proud to own the vendor’s product and have a positive experience throughout the life of the product?

A product or service enabled by image recognition is generally available at a monthly managed-service fee, as a licensed solution or SaaS with annual contracts. But prioritize functionality and benefits above delivery method. Constantly monitor the business benefits by asking these key critical questions:

  • Sales force productivity: Will this lead to an increase in stores/aisles visited?
  • Incremental sales: Will this improve in-stock facings, promotion and planogram compliance and hence lead to incremental sales?
  • Better shelf insights: Will shelf condition data be acted upon more often?

Trax clients have achieved 3-5% sales uplifts per category, and increased store coverage manifold with IR products.

9. Customer services experience: Is the customer service accessible, fast, stress-free, and of high quality?

A customer account representative should be on-hand to handle any questions and solve any problems, from the initial onboarding, through rollout and for the entire duration of the customer relationship. The solution should be easily configurable, and you should be able to create reports, add users, add fields, or perform other such tasks with minimal dependency on the vendor.

10. Brand equity: Do existing customers view the vendor positively and exhibit high brand loyalty?

Seek evidence on past implementations that helped solve challenges similar to yours. Consider the vendor’s long-term customers, and history of responsiveness in terms of ability to change direction as customer needs evolve, and market dynamics change. Coca-Cola Amatil was one of Trax’s earliest clients, and the project’s success resulted in a long and successful relationship. Today, dozens of other divisions worldwide are using Trax to improve retail execution and gain new market intelligence for their brands.

Whichever image recognition system you choose, make sure it fulfills your specific retail needs. Get it right, and you can make the changes required to improve positioning, enhance market share and uplift sales. Get it wrong, and you may as well go back to pen and paper.

Download the Frost & Sullivan report here.

AUTHOR
Manu Krishna
Manu Krishna Director, Content Marketing at Trax Manu is a technology marketing professional currently leading Content Marketing at Trax. When he's not hacking inbound and building content that lasts, he is usually in a park reading a book.

Related posts

ARTICLE
by Corliss Y. Collier

“The missing link”: Leveraging in-store execution data to inform category strategy

The shelf is a constantly changing entity. Several events trigger major and minor shelf actions. Product or category innovation, shopper leakage, market and regulatory changes, retailer merger or acquisition – all are reasons for retailers and CPG companies to review categories periodically. Category reviews present an opportunity for CPG companies to disrupt the market; both brand and category performance depend on what products make the shelf and how they are segmented.

Read more