• Suss your stock issue shortages with image recognition

    Grow sales. Drive efficiencies. Deliver better insights. Availability is a fundamental factor when it comes to sales success and ensuring stock levels are maintained (and in view!) is obviously essential. Using the latest image recognition technology, you can quickly identify on-shelf availability and address problematic stock situations without breaking the bank.

Better data

richer, more accurate data

Easy analysis

of your own products against competitors

Drive efficiencies

in your field team, reduce the time taken to collect and record data

Instant action

point to required action instantly. Improve your profits
Image recognition image 01
Image recognition image 02

Wield the power of insights

Using the latest image recognition technology, you can quickly identify on-shelf availability and address problematic stock situations. Consider difficult conversations around compliance a thing of the past. With this undeniable evidence at your fingertips, you can very easily query a lack of ‘facings’ with retail partners.

How image recognition works

Image recognition technology now makes it possible to monitor shelving with a quick snap of a photo. It works for any category, shape, or size of product. Crucially, it’s cost effective, easy to set up and eliminates the need for complex planograms.

Less faff, more sales.

The images gathered and read by the system mean brands can access an immediate view of their overall share of space at any one time.
Benchmark and compare against competitors, confront a lack of retailer adherence, and spot and react to any stocking trends on a hyper-local, regional, or national scale.

Power to the people

Of course, we mustn’t forget the importance of marrying the tech eye with the human eye. Reports are far more powerful when combined with trend analysis produced by the professionals routinely visiting stores on your brand’s behalf. We know from experience that having the best people representing your brand means you get the best results – better recorded data, more effectively.

Image recognition image 03
Image recognition image 03

Our merchandisers are within 10 miles of 97% of stores in the UK, ready to support with 365-day promotion for your brand.

Shelf-space. Positioning. Visibility. Product selection. A lot of thought and expertise goes into maximising your sales. We get it. And whether its POS installation, check and fix, auditing or range changes, we combine big brains with big data to make sure the right improvements happen. 

60,000 photos per week for client reporting

40,000 merchandising hours completed per week

Paint a clearer picture with image recognition technology

Over the last few years, image recognition technology has advanced by leaps and bounds. We've all watched in wonder as a team needs to pull off a long-winded heist to get into a vault - but today, its perks can be found anywhere. Beyond security, image recognition software is in use in the healthcare industry, assisting with microsurgical treatments and kinematics when an extra pair of eyes is required.

The automotive business, as well as, predictably, the gaming industry, are employing this sophisticated technology to push the boundaries of innovation, with applications such as optimising production processes and utilising facial recognition to boost player immersion.

But how does it work?

Image recognition, a subset of artificial intelligence, is used in practically every area of our life, with its machine-learning skills identifying product properties and features in digital photos or video.

Dee Set's industry-leading image recognition system sees the world in a unique way. Its digital image processing recognises that a picture is a picture, but digs deeper into the components that make it up. Pixels, with their plethora of numerical data, are the best approximation of what we humans see with our eyes, according to an image identification technology. 

It's vital to highlight that without some deep learning, computers will fail to distinguish between photos. This is because computers 'see' in numbers rather than concepts or other knowledge. 

Like most retail machine identification solutions, our technology converts numerical data into a tapestry of patterns and regularities, using these recognisable elements as a "representation" of what it's looking at. Isn't it fantastic?

Image recognition, on the other hand, should not be confused with object detection, which is another automatic human talent. Computers are amazing - they wouldn't have gained such widespread popularity if they weren't - but they require some very clever technology to mimic our eyes and brains in comprehending the world around us. 

Image recognition technology is a good approach for computers to process visual input in this way. This is due to the fact that, as the name implies, it can process aspects of an external image and classify it into several product categories.

How does image recognition work, then?

So, how can one of our smart solutions figure out what it's looking at? When it comes to learning about actionable insights and product placement - even artificial intelligence - we all have to start somewhere, just like our own team of retail industry experts. It all starts with the establishment of a neural network that processes each pixel in a picture individually. 

The team will feed the network pre-labelled images relevant to its objective, such as renderings of a shop shelf, to train it how to recognise comparable imagery without prompting, in order for it to 'learn' what it is seeing and pick up on the visual display.

The more data we give the image recognition software, the better the neural networks, which are made up of multilayer perceptrons (MLPs), convolutional layers, and pooling layers, can develop paths to assist it to grasp what it's seeing. For example, in this retail application, artificial neural networks see the share of a shelf - the total amount of shelf space your brand occupies - as a grid. These layers act as filters, scanning pixels and passing on pertinent information to the next one, assisting the system in narrowing down and digesting what it is viewing.

We at Dee Set have not only witnessed these recent developments but have also led the way in the field of image recognition. We not only find the most up-to-date data, but we also crunch it in a way that makes sense, serving it up in reports that make marketing magic happen. 

Why do retail businesses need to be using this technology?

With so much potential, it's easy to see why we'd want to get our hands on this technology and put it to use in the retail industry. Competitive intelligence tools are already being utilised by field reps all over the world to help improve customer experience and expedite the packing, processing, and promotion of consumer goods in a variety of industries. Visual content given to these algorithms can identify even finer parameters like price detection and general quality, allowing businesses and customers to improve overall shelf health.

Other aspects, such as enhanced biometrics, have been added to image recognition technologies, allowing businesses and marketers to receive a full picture of how their items are used and engaged with on the shop floor. This is demonstrated in a variety of ways:

  • Product audits: We understand that the shelf is probably not the most exciting spot to look for those critical figures. The idea of physically auditing physical stores makes the entire Dee Set team shudder, which is why we're delighted that image recognition not only speeds up but also smooths out the process. 

Because our time is valuable, picture recognition technology eliminates false audit data and sloppy retail executions that don't match the suggested layout - understandable, given that we're all only human. Artificial intelligence, on the other hand, thrives where our brains struggle with data alignment and standardisation. This data, which computers can analyse down to the smallest detail, may be used to determine not just how much shelf space each brand has, but also pricing trends and retail sales. An intelligent audit is unlike any other.

Retailers should know how to look after the general upkeep of a shelf according to the terms and conditions of a contract, but a variety of variables, from short staffing to shipping concerns, can make that compliance difficult to achieve. This is where technology can assist retailers and brands alike bridge that crucial gap, allowing them to stay on top of their planogram game.

Image recognition can employ a single photo to evaluate whether the items on the shelf match the reference planogram, resulting in a layered algorithm that sends out an alarm if there are any differences. We can't all be in the same location at the same time, but this technology allows firms to meet client demand wherever their products are sold.

  • Empty shelf detection: Shopper dissatisfaction is a genuine problem. Stockouts (also known as shelf shortages) are one of the most common reasons for customer unhappiness, and merchants are struggling to meet rising demand while dealing with supply challenges.

When shelves are bare and stores are understocked, consumer and brand concerns develop, which is fair given that this is visual evidence of money missing from the industry's normal revenue streams. It's a looming problem made worse by current geopolitical and global influences, but technology has been able to not just recover but revolutionise how businesses can keep their shelves stocked for the sake of their customers - and their bottom lines. When a critical SKU is missing from its designated location, image recognition software can pick up the slack, alerting workers to the need to replace stock or place a new order.

As our systems get more intelligent, these machines will be able to learn more complex algorithms to anticipate a shortage, analysing trends and customer bases to guarantee that business continues to thrive. Who doesn't enjoy that?

woman taking a picture of cosmetics in a store

What’s it worth to your brand?

The adaptability of image recognition is one of its key advantages. Digital picture processing transforms empty shelves into a product library ready to be assigned with a single snap from one of our Brand Champions, allowing marketers to contextualise gaps into potential income. With data obtained directly from the source, these real-time photographs give our clients insight into what's happening right now in the stores where your items are being used.

Our sophisticated picture recognition technology that collects crucial numbers where you need them most, powers our solutions. Stock availability and comparative data from selected brands may be supplied directly to your device with this handy application, as well as the ability to guarantee store compliance and planograms are followed. 

This clever software can detect and apply pricing patterns from a variety of sources, saving you time, money, and effort. Our software can recognise product sizes, forms, and categories when assessing things as part of its report, so there's no need to take multiple photos. This allows our brands to personalise and alter the parts they want to focus on.

We make the shelf edge more efficient. But there's a crew behind every outstanding piece of equipment that not only innovates but also enhances its use. 

The Dee Set team visits stores regularly during our Data-Led Calls to capture in-house photographs and gain a sense of how your product offering is regarded in the retail environment. We work as an extension of your brand, cultivating important relationships and representing your company while also figuring out how to get the most up-to-date data for our reports.

Dee Set's team of data scientists makes over 6,000 calls per week in stores across the country, making our approaches among the most effective in field marketing. We employ picture recognition to swiftly show and explain how your items might be underperforming in particular sectors, armed with a passion for information and a taste for the crunchiest of numbers. 

This useful information is compiled into simple, easy-to-understand reports that can be updated on a regular basis, providing insight that helps your team decide which course to take next.

Our retail technology takes your business further

We believe in working smarter rather than harder. We recognise that consumer brands want to understand customer behaviour, which is why most of our data-driven, analytical solutions include picture recognition. We can deliver reliable data straight from the source for whatever our businesses' needs are, from unified strategy planning to provide actionable insights from a wide range of stores and potential shelves.

The use of image recognition eliminates the need for time-consuming documentation and human error. This technical wizardry may scan a visual display or reference planogram from an input image and equate them to a number of datasets, indicating which component of shelf availability, planogram compliance, or product positioning has to be focused on.

When you work with Dee Set for retail store analytics, we combine experienced expertise with cutting-edge technology. Image recognition is always harnessed by a group of award-winning field marketers with a wealth of experience, using our own designed method for gathering visual data.

We understand that gathering large volumes of data and running it through a reporting system isn't enough. Because not everyone talks geek - and because we can only take so much number crunching - we try to make our data speak for itself about what businesses can do to enhance their standing. 

Dee Set develops data-driven solutions that are human-powered. This gives us the best of both worlds, with field agents that can apply our industry-leading technological solution to a variety of brand issues, converting them into crucial retail data.

This dual approach sets us apart from the competition, combining technical competence with a team that enjoys delving into the latest trends and industry analyses. 

Our reports are basic and transparent, allowing our clients to construct plans around essential facts. Your objectives are also our objectives, thus we work hard to provide solutions that not only solve your difficulties but also promise growth month after month. Although it isn't rocket science, we like to think that our service is truly extraordinary.

Clients we work with

  • Morrisons logo
    Accessorize logo
    Lavazza Logo
    Waitrose logo
  • Claire's logo
    Kimberley Clark Logo
    RB logo
    Innocent logo
  • Sainsbury's logo
    Kelloggs logo
    GU logo
    Castrol logo
  • B&Q Logo
    Homebase logo
    Morrisons logo
    Accessorize logo
Dee Set Logistics Ltd/Dee Set Confectionery Ltd, trading as Dee Set, registered in England, Scotland and Wales. Registered No: SC208421/04297287.Vat No: 896110414.