Your shelf data is not reliable… Unless you have Shelf Recognition

Modern sales are an area of commerce that increasingly relies on data. In a world where almost everything can be measured and calculated, the area left to subjective evaluation is shrinking. While an element of emotion, risk or madness may even be advisable in the creation of business visions, it should be kept to a minimum in the execution and implementation of business concepts.

How important is data for sales?

Working with clients at eLeader, we seek to reduce uncertainty and waste of resources through process automation, diligent reporting and accurate analysis. Only a business creed formulated in this way can guarantee that decisions determining the fate of quite a few organizations will be accurate and ensure success.

One area our clients pay special attention to is the in-store display. This is the place where the sales strategy ingratiates itself with the consumer and jostles for their attention. It is at the product shelf, after all, where the verification of sales managers’ assumptions takes place and where the future of specific goods is decided.

It should therefore come as no surprise that among the duties of representatives visiting stores are the tasks of checking whether the shelf complies with the standards, what the competition is doing, whether something can be done to immediately rectify the deviations from the standard, and finally, whether the visual strategy needs to be revamped.

In order to fulfill the above objective, it is necessary to provide an adequate range of data. This cannot be done without information gathered from many points, sourced frequently and arranged into business-relevant indicators. The quality of the data itself is also of importance here – it must reflect the actual situation. And this is where the hard part begins, as there is a deadly chasm between theory and practice, i.e. strategy and execution.


The store is not as we see it

Field representatives serve as your eyes, ears and hands in the market. You delegate them to provide data and information about the reality at points of sale and to take action to change this reality in line with the company’s vision. An example of such reporting is a merchandise display audit meant to gather information on the appearance of the product shelf on which your own and your competitors’ products are stacked, and compare the shelf to the standard. Employees are rewarded with a salary for providing this information. Bonuses are awarded for achieving high compliance or exceeding the display target. Knowing the situation on the shelf, you plan further action – a fulfilled standard reassures you that execution is going according to plan, a good share of the shelf tells you that the market position is good, and the correct location of products shows that both sales representatives and store employees have grasped the standards and are displaying goods in accordance with them.

And now, find out what it is really like.

In 1992, two American psychologists Arien Marck and Irvin Rock described in professional literature the phenomenon of the so-called inattentional blindness. In our everyday lives, we do not see many things that are in front of our eyes (like half of the people who failed to perceive a huge gorilla in an experiment using the video embedded below). Inattentional blindness is our natural filter that protects us from too much information that we are unable to process. Obviously, our perception cannot tolerate any “white spaces”, therefore it replaces various objects with other elements, following goal-directed selection. According to this psychologically proven principle, our mind selects for perception of reality mainly those elements that support the assumptions it has made about it (which may be called a form of wishful thinking).

The choices we make based on sensory impressions are subject to the goals we set for ourselves. So if our goal is a perfect shelf, we see each product shelf as better than it actually is. The mind also automatically fills in gaps in the image we see (the principle of closure and continuity) and groups similar objects into a single set (the principle of grouping). In practice, this results in seeing objects that are not there in reality. Psychology describes these phenomena as principles of perceptual organization. Another regularity known to psychologists is the so-called change blindness. Known already in the 19th century, this phenomenon affects about 75% of the population. It refers to schematic perception of the world, i.e. seeing in a way that follows the originally “encoded” image and failing to notice even significant changes in the environment. When examining a store display, whose primary feature is change, this peculiarity of our psyche certainly does not help.

As thinking human beings, we are subject to laws that make us fallible (illusionists use these mechanisms in their tricks), and this fallibility is linked to the ability to concentrate, i.e. maintain attention, which decreases with fatigue and repetition of daily activities. Sales representatives in the store are no exception to the rules described above, so in spite of their sincerest intentions and good motivation, each day they have no choice but to capitulate to the laws governing biology.

The pitfalls of bonuses

Apart from our biological conditioning, there is another key factor contributing to the distortion of field-sourced data. Unfortunately, the culprits here are the very “victims” of this phenomenon, i.e. sales managers. Subjecting bonus payments to the achievement of a proper display standard is a decision showing knowledge and appreciation of the rules governing consumer behavior – we buy what we see, most often at eye level and abundantly represented on display, when we feel that we have a choice.

Shelf information is therefore key to answering whether those in charge of setting up merchandise are ensuring these conditions in stores. Unfortunately, by simultaneously imposing on these persons the obligation to manually report indicators that are relevant to their bonuses, supervisors are forcing them to be judges in their own case and impairing the objectivity of the audits. If you remember the psychological rules governing human perception, you will understand that, unintentionally, this leads to an undesirable idealization of the image of the store shelf, i.e. a wishful perception of it as looking somewhat better than it actually does (because this determines the objective – a bonus for meeting the standard).


So it does not take any deliberate manipulation of the shelf audit results for there to be a feedback loop along the line: demanding a standard – meeting a standard – raising the bar – meeting a standard – etc… As a result, in a short time, an idealized but completely inconsistent with reality picture of the position in the market is created, on the basis of which strategic business decisions are made. At eLeader, we know from experience that such a situation can persist in companies for years, with participants (structures, management and board) mutually reinforcing each other’s misconceptions.

The only logical remedy for this situation is to objectivize the product shelf audit, that is, relieve the sales representative of the role of a judge in their own case. The methods available so far included inspection visits (audits of sales representatives performed by independent structures or external companies) and visual analyses of photos documenting the effects of shelf work, followed by verification of these results against the reports in the SFA system. The problem, however, is the cost of conducting such activities on a regular basis, which – for advanced standards – could amount to half the cost of maintaining the field force.


For one of our clients, we performed an undercover audit of sales representatives’ visits. We used a manual count of the products in the photos taken during that visit. Then, we compared the results with what the sales representatives had reported. In individual points of sale of the traditional channel, we recorded discrepancies in the range of 37,5 to 115% for present products, which shows how heavily distorted data flows into analytical and bonus systems.

The illusions of managing sales

As you probably already understand, decisions that rely on data obtained from the field are in fact not always the best. Practice shows that, especially in companies that have decided to implement an automated display audit tool, irregularities in reporting translate mainly into:

Distorted perception of your own market position – if you gathered at one table the CEOs of the five largest FMCG producers, e.g. in the sparkling beverage category, and they started boasting about their shelf shares based on company reports, it would appear that together they have a 170% shelf share.

Unfair and ineffective bonus system – a system rewarding achievements that do not exist is detached from reality. Subsequently, it makes room for abuse and frustrates the representatives who are trying to achieve good results in an honest way.

Illusory management – there is a difference between managing sales when you have 20% and when you have 50% of the product shelf. Distorted data means that you are operating in a parallel reality, while real decisions make themselves or someone else makes them for you.

What to do?

Knowing the challenges of companies around the world, in 2013 we started developing a solution to authenticate the in-store display data, significantly increasing its scope, as well as the time of its acquisition and calculation of standards. We have relieved human senses and minds by means of artificial intelligence in those elements of merchandising where objective evaluation of facts is crucial and it should not be subject to the intention or condition of the auditor. The result of this development, eLeader Shelf Recognition AI, successfully supports the work of the sales force of companies such as Maspex, NUTRICIA, DANONE, Lactalis, and Lagardère. During the tests of the solution, we had the opportunity to monitor the previously reported shelf parameters. Once Shelf Recognition started measurements, there was a dramatic decline in the readings in comparison to the declarative manual reporting. On many occasions, this “moment of truth” was a test during which our clients could not accept the fact that for years they had been stuck in ignorance about the reality of the store shelf and were not aware what they had actually been paying bonuses for. Even proving the quality of automated audits hardly helped here, as it was simply too much for them.

Nevertheless, many companies decided to adopt the tool. Where AI-supported verification of reporting was implemented, it proved beneficial to the organization, because it marked the beginning of working with good data and drawing conclusions from a real, rather than idealized, situation. However, in order for all parties in the project to benefit from the implementation, you should take care of a soft landing on the new territory.

A great reset

Implementing eLeader Shelf Recognition AI is a way to obtain reliable data from the market and streamline the work of field teams. Making a decision to entrust artificial intelligence with display audits should, first of all, be accompanied by preparing all those involved in the changes for a bucket of cold water. No one is going to feel good about the plunge in metrics, but it is important to remember that this condition is nobody’s fault but a result of an imperfect execution model commonly used in the market. Instead, one should be glad to get closer to reliable data.

Second, the new reporting model is an opportunity to remodel indicators. Shelf standards, which have been overestimated for a long time, should be made more realistic to serve as a starting point for setting new targets (e.g., in the perfect store model).

Third, special care should be taken to inform salespeople about the planned remodeling and making the bonus program more realistic. It is crucial that the team is aware of and prepared for the planned changes. It is particularly important to ensure that diligent and honest employees will not lose out on the proper execution of standards.

Fourth, a transition period is recommended to allow time for employees to become accustomed to the new shelf audit methods, break free from the currently functioning, flawed processes, and to allow them to set sales targets in accordance with the SMART principle. If you follow the guidelines above, you have a much better chance of success in the field of managing and achieving sales targets.

Find out for yourself!

We know that it is hard to accept that there may be irregularities hiding in our own data acquisition processes. Therefore, we encourage you to independently verify the scale of this phenomenon. Both verification of photographic documentation and incognito visits by auditors immediately after the verified visit should provide an answer to the question of the extent to which the sourced information reflects what is actually happening in the market.

We also encourage you to contact our specialists, who will be happy to support the process of verifying the quality of reporting. They will help you make a strategic decision and evaluate whether the internal and external conditions are met for you to “employ” eLeader Shelf Recognition AI for merchandising. In many cases, it is possible to test the solution on a selected product category.

eLeader Shelf Recognition AI

eLeader has 20+ years of global experience in providing software solutions for salespeople. The company’s portfolio includes SFA, BI, CRM, IR and low code business solutions.


eLeader Shelf Recognition AI is a solution for automated online shelf audit. Its quality and effectiveness have been entrusted to a separate AI R&D department, whose task is to develop the product, and to specialized customer and implementation support departments responsible for analyzing the organization’s needs and capabilities when implementing automatic display audits and accompanying processes.

Our specialists will be happy to support you in the process of digital transformation by introducing you to both the technical and business aspects of implementing automated store display auditing solutions.

Would you like to learn more? Download our free e-book and discover all the benefits of the solution:

Recommended articles:

Subscribe to our newsletter: