Us And Them

Besides being the title of a great Pink Floyd song, "Us And Them" is a very a propos title for our latest blog posting.

Our position as a leading provider of on-site video analytics gives us a unique view into the entire video analytics ecosystem. Lately, we've started to understand at a much deeper level where we fit, and how our approach to video analytics compares to existing technologies in terms of providing business intelligence, whether for retailers, quick service restaurants or other types of physical locations.

The emerging picture we see is that there are a number of security analytics softwares that are being marketed as suitable for business intelligence (BI) analytics. The key features that attract retailers to these software suites is that they are relatively inexpensive and integrate easily into digital video surveillance systems. Yet, fundamentally, these security analytics are ill-suited for retail BI at best, and at worst deliver incorrect, flawed data that leads to damaging business decisions.

Consider a simple task that many security analytics tools perform for BI purposes: people counting. From a video analytics perspective, this is simply noting a large-scale change in recorded pixels within a trafficked area. For the purposes of security analytics, the software is typically looking for what would be considered an "exception" - that is, a person where one should not be. This works great in a security system where a single person is that exception. 

Yet from a BI perspective for retailers, the capabilities of these tools weaken significantly. Consider, for instance, a scenario where a supermarket wants to use security-based analytics to record shopper counts in aisles to create basic store heat maps and conversion rates for categories. The typical supermarket aisle is crowded. You've got a large number of shoppers, shopping carts and employees. Depending on the view of the camera, a line of three shoppers can be recorded as a single shopper. Carts can be recorded as people. And shoppers that appear and disappear from view can be counted twice or more. With security analytics systems, these errors are being captured continually, adding layer upon layer of false data. If the supermarket retailer's merchants, marketers and operations personnel use these data to support decisions that impact their business, the results can be disastrous.

Our solutions have been specifically designed as BI video analytics tools. The methodology behind our analytics excels in challenging retail and QSR environments where heavy crowds, mixtures of customers and employees and objects such as shopping carts abound. Thus, the insights we deliver in the same supermarket situation noted above would have a vastly reduced error rate. As an added benefit, our solutions layer in additional insight such as consumer demographics, consumer-to-employee ratios and interactions, and consumer engagement with merchandise, for instance. 

To summarize, our ground up approach to develop BI focused analytics versus repurposing security analytics results in significant error reduction and much richer insight. Between "Us and Them," we're clearly the right choice for retailers and QSRs looking for scalable video analytics for business intelligence. 

Finally, for retailers and QSRs who are curious about what we can do, why not request a free trial? All you need to do is send us a video file from your in-store surveillance system, and we'll show you what we can do.