Revenue Growth +32% for a Perfumery and Cosmetics Manufacturer

The place of inspiration for the development and testing of the methodology was a large Russian manufacturer of perfumery and cosmetic products, which has been on the market for over 70 years.

At the time the algorithms were applied, the enterprise’s annual turnover was more than $200 million, and quantitative distribution in many areas reached 100%. The products of this enterprise were sold by almost all outlets in the country, however, the number of assortment positions within the outlets was far from ideal.

My name is Irina Lukina, I’m a co-owner and the Head of R&D at AGATA. During the implementation of this case, I joined the Client's enterprise as a Brand Director of the toothpaste business. At that time the bulk of the products in this area was illiquid. The employees of the department could not single out locomotive and basic products from the entire product range and the number of assortment positions was determined randomly. And while competitors like Colgate and P&G sought to meet the demand for new tastes and useful properties of pastes, the group's assortment expanded without re-evaluating old positions.

At the same time, the new products could not offer consumers anything new, their difference from the old ones was insignificant. The marketing department could not reasonably answer the questions:

The enterprise had market data from ACNielsen, which could have been applied to analyze the distribution, knowledge level, advertising costs of each competitor and identify connections between them. However, this data was viewed only as a rating:

“Oh! We are in 5th place in terms of revenue, and 3 in terms of sales. Great?”

Globally, the enterprise lacked a single decision-making center to back up the answers to these and many other questions with market data:

How many SKUs should there be?
To what extent do you need to produce this or that product?

What gramme would be optimal?

Do I need to form a higher price category?

Why are competitors spending exactly that amount on advertising?

First, the task was to form an ideal assortment in terms of the number of SKUs, sales per 1 SKU, the level of knowledge, quantitative and qualitative distribution. We started our solution by analyzing internal and competitive data, as a result of which we divided our range of toothpastes into three price categories:

• low (3 SKU),
• middle (5 SKU)
• upper middle segment (4 SKU).

Then, for each category, we calculated the average value of SKU and sales per 1 SKU - these were the benchmarks for the entire market. When comparing the indicators of market standards with our assortment, we found out:

Further, to create new SKUs, we analyzed market demand. At the time, buyers preferred natural and healthy ingredients like Colgate's propolis. We needed to get into the trend, but not repeat the products already on the market. So there was a sea buckthorn toothpaste, sales of which grew so much that competitors began to copy the taste.

The situation is similar with the upper price segment of toothpastes. At that time, chemical bleaching was relevant, and we offered only mechanical bleaching, so we finalized the line with corresponding high-price products.

When developing new formulations, we were faced with the task of creating a high-quality topical product, while not lowering the marginal profitability indicators below the planned one. We began to approach the selection of components in detail, comparing not only their properties, but also the conditions of suppliers. Since this was already substantive work, the suppliers were only glad:

Thanks to the transformation of the approach to the components selection, we not only ensured the planned margins of new products, but also optimized all other products. The total cost of the brand decreased by 8%.

It was the benchmark algorithms that set the very rigid framework on which we have been working.

The enterprise's sales project also included a "historically formed" volume of toothpaste sales - 380 million rubles a year. Planned indicators for the year, taking into account this volume, were taken out of thin air: "Let's do, for example, 420 million." Relying on this goal seemed to me an irrational decision, so I calculated the plan using aforementioned benchmarks and the bottom-up method.

As a result of the calculation using algorithms, the planned sales volume was 560 million rubles, instead of 420 million. And this was perceived by the management as a shock! I was asked how this is possible, and I showed a mathematical model. They looked at the calculations and agreed. At the end of the year, we really received the desired 560 million rubles!

It took me 3 years to develop a working mathematical model:

• 1st year, I got acquainted and understood the assortment and market specifics
• 2nd, I began to look for interdependencies between indicators and derive methods
• 3rd year I was busy with the implementation of the plan

And yes, in the end, sales amounted to 560 million rubles, as planned! Now the AGATA algorithm contains the very mathematical models that allow you to avoid 2 years to search for dependencies.

All that needs to be done is to enter the external and internal data of your market, and then only plan and monitor indicators in the platform.