How to do an abs analysis correctly. How joint ABC and XYZ analysis will help you in business. How ABC Analysis Can Be Used in Combination with Other Strategy Tools

Today, the essence of assortment optimization is increasingly reduced to identifying and developing the most significant groups of products in terms of profitability. It is not necessary to use complex marketing theories. As a basis, you can take, for example, data on profit in the context of product groups and connect the ABC analysis method. Let's take a closer look at how to analyze the assortment of goods using only basic financial indicators.

In this article, you will learn:

The assortment of many companies was formed spontaneously. New suppliers and brands, new nomenclature were added. A large assortment served a large turnover. But at the same time, all this required large financial resources for the purchase, production and storage. Today, this is the case for companies, many of which lack working capital, an unaffordable luxury.

The main signs of a poorly selected assortment line:

• an increase in warehouse balances both in kind and in value terms, and at the same time the lack of goods demanded by customers;
• an increase in the number of positions, accompanied by a decrease in profits;
• lack of resources for procurement.

Moreover, expert methods, when the assortment is formed only on the basis of the opinion of the heads of sales departments, do not work in the case of a large number of product names. A tough mathematical approach is needed. In fact, the essence of assortment optimization work is often closer and more understandable to the CFO than, for example, to the head of the marketing department, and boils down to highlighting profitable product groups and, accordingly, developing them, the remaining ones - carefully minimizing.

Personal experience
Mikhail Podlazov,
Before optimizing the assortment, it is necessary to determine what the company plans to achieve. Typically, enterprises pursue three key goals: increasing revenue and profitability of sales, optimizing stocks of finished products, expanding sales markets and conquering new ones.
At Nidan Juices, four key metrics are assessed and analyzed before adjusting the product portfolio:

• arithmetic mean sales volume;
• product line size;
• storage facilities, transport network and production;
• profitability.

An example of ABC analysis of product groups

The initial data can be information downloaded, for example, from 1C to Exel and containing information about profit in the context of product groups. The more honestly the profit for them is calculated, that is, all costs for each group are taken into account as much as possible (purchase prices, delivery, packing, storage in a warehouse, etc.), the more accurate the optimization will be.

It makes sense to start with an ABC analysis using the example of a profit indicator. Moreover, as practice shows, the use of some more complex techniques simply does not justify itself. Data for analysis (groups of goods and the profit received by the company on them), as a rule, are taken for the maximum possible period of time (a year or more) in order to smooth out seasonality different groups goods and errors in the planning of purchases or production, as a result of which the product is temporarily out of stock. But due to the fact that the situation in many markets has changed dramatically since the beginning of the crisis, it is better to limit ourselves to analyzing the results from the beginning of 2009.

How will help: Maximize profits by managing inventory and not keeping money in stock.

How will it help: identify and eliminate surplus stocks, implement a system for monitoring and optimizing stocks.

So, to conduct ABC analysis, you will need to determine the profit generated by each specific group of goods, its share in the total amount of profit and rank the groups by this indicator, as well as calculate the share in the cumulative profit (see Table 1).

Table 1. An example of ABC analysis for product groups

Serial number in assortment	Product rating in terms of "profit share"	Group name	Profit for the first half of 2009, rub.	Share in profit,%	Accumulated profit share,%	Profit category
10	1	Beer	3 324 754	20,7	20,7	BUT
14	2	Perfumery and hygiene	2 157 010	13,4	34,1	BUT
1	3	Alcohol	2 040 270	12,7	46,8	BUT
12	4	Snacks	1 504 268	9,4	56,1	IN
3	5	Soft drinks	1 482 471	9,2	65,3	IN
5	6	Confectionery	1 469 275	9,1	74,5	IN
8	7	Meat products	1 205 017	7,5	82,0	IN
11	8	Cigarettes	1 093 273	6,8	88,7	WITH
4	9	Frozen food	724 245	4,5	93,3	WITH
2	10	Grocery	332 012	2,1	95,3	WITH
13	11	Juices	270 044	1,7	97,0	WITH
9	12	Household products	201 096	1,3	98,2	WITH
7	13	Milk products	191 609	1,2	99,4	WITH
6	15	Coffee Tea	80 046	0,5	99,9	WITH
15	16	Bread	10 832	0,1	100,0	WITH
TOTAL			16 086 221	100

It should be said right away that ABC analysis using the example of a company's assortment assumes the allocation of groups in slightly different proportions than the classical Pareto rule suggests. When conducting ABC analysis, it will be quite reasonable to use the following criteria:

• category A - groups with the largest contribution to profit (share in profit), which together give up to 50 percent of the company's total profit;
• category B - product groups, which in total give about 35 percent of the profit, and their accumulated share in the profit is from 50 to 85 percent;
• category C - groups with the lowest share in profit, bringing in the remaining 15 percent (cumulative share from 85 to 100 percent).

By the way, at the stage of analyzing the share in the profit of each product group, it would not be superfluous to compare the data obtained with similar indicators, for example, for the same period last year. A decrease in the share in profits may well indicate a post-crisis reduction in demand for certain goods, and it is advisable to take this trend into account in further work assortment optimization.

Personal experience
Mikhail Podlazov, Deputy General Director for Economics and Finance "Nidan Soki"
For us, the profitability of a commodity item is one of the most significant indicators that allow us to judge the need to produce a particular item. Precedents, when the company refused some commodity items due to their low profitability, took place in the practice of "Nidan Soki". Thus, two headings in the product line, which included 8 headings, demonstrated low profitability for six months. The reason is great losses during equipment reconfiguration and high cost of packaging. The finance department calculated that in order to reach the minimum required level of profit, it is necessary to increase the sales volume of each item by 80 percent.
In conditions competitive market it was impossible to do this, and the company had to abandon the production of these two commodity items. Of course, it also took into account such indicators as sales volume, product line size, storage, transport and production capabilities. After the decision to reduce commodity items is made, available raw materials and materials are developed that are used only for the production of unprofitable SKUs. Then the equipment is reconfigured, the production schedules are changed. The final stage is the adjustment of price lists and the withdrawal of product leftovers from store shelves.

After conducting the aggregated ABC analysis (example above), it is necessary to expand and deepen the strong groups in terms of profitability and reduce the financially disadvantaged categories.

Optimizing the most profitable product group

Product groups allocated to category A bring the company half of all profits, and their optimization should have the greatest effect. But in order to work with this group, you will have to analyze its composition.

Let's make a reservation right away that a product group can be homogeneous in composition, such as, for example, "beer" or "juices". In them, the products will differ only in brand, taste or package size. Or the group contains subgroups. For example, the group "perfumery and hygiene" contains the subgroups "soap", "wipes", "shampoos", "deodorants", etc. In order to organize all this, the concept of width and depth of assortment is most often used. The width is determined by non-fungible subgroups of goods. For example, for the “footwear” product group, the assortment width will be subgroups: “winter”, “summer”, “beach”. The depth is appropriately determined by the interchangeable goods. They are usually located within a subgroup. For example, goods in different packaging volumes or similar in their characteristics, produced under different brands.

So, group A optimization assumes that the width and depth of the group should be maximized. In practice, most often in one subgroup in depth it is justified to keep no more than 6–7 positions, but in relation to group A there may be exceptions. This is the key idea of ​​optimizing the assortment of goods that bring the greatest profit, and the algorithm for reducing the names that are unprofitable for the company will be as follows.

Step 1. Checking the number of titles in subgroups. Within the group of category A, an ABC analysis is carried out, an example of which was described above. The bottom line is to determine the subgroups of category C and check the number of names for them.

According to the universal rule, the number of items in each subgroup assigned to this category should not exceed the product of the number of headings in group A multiplied by the share in profit. For example, if the whole group A, in particular "perfumery and hygiene", consists of 300 items and includes such subgroups as "shampoo", "nail polish remover", "soap", "toothpaste", etc. the subgroup “nail polish removers” was categorized as C (profit share 0.9 percent). Accordingly, there should be no more than three types of nail polish removers (0.9: 100 X 300 = 2.7). If this is not the case, their number will have to be reduced. To determine the choice, an ABC analysis is carried out for specific headings - the next step.

By the way, if the number of items in the studied group of category A is less than 50 pieces, then the first step can be neglected and immediately proceed to the analysis by headings.

Step 2. Analysis of names. The logic of actions at this stage is the same - ABC analysis by share in profit with the only difference that the object is the analysis of specific product names (see Table 2). Category C falls under the abbreviation. Although there are some exceptions that need to be considered, namely:

• if the position of the company is not critical, you should not cross out the products that were launched recently from the assortment. It is clear that the profit on them is lower, if only because they are sold for less time than all the others;
• accessories and related products can fall into category C, which stimulate sales of category A.

table 2... Example for trade names of the subgroup "shampoo"

№	Subgroup	Profit for the first half year 2009, rub.	Share in profit,%	Share in profit as a cumulative total,%	Profit category
1	Shampoo "Niveya" dmuzh 250 ml	58 636	16,86	16,86	A
2	Shampoo "Nivea" dry 250 ml	49 985	14,38	31,24	A
3	Shampoo "Nivea" for hair	41 090	11,82	43,06	A
4	Shampoo "Nivea" with extrapod 250 ml	27 551	7,92	50,98	A
5	Shampoo "Nivea" jir vol 250 ml	26 211	7,54	58,52	A
6	Shampoo "Niveya" dmuzh 250 ml	19 582	5,63	64,15	B
7	Shampoo "Nivea" for perch. with extber 250 ml	18 451	5,31	69,46	B
8	Shampoo "Nivea" volume 250 ml	17 351	4,99	74,45	B
9	Shampoo "Belito" kefir 500 ml	17 107	4,92	79,37	B
10	Shampoo "Belito" beer yeast	15 165	4,36	83,73	B
11	Shampoo "Belita exclusive" eggs 585 g	13 459	3,87	87,60	B
12	Shampoo "Daf" osmilvol 200ml	8646	2,49	90,08	C
13	Shampoo "Belito exclusive" henna 585 g	7729	2,22	92,31	C
14	Shampoo "Timothy" Cherry / cotton 400 ml	7217	2,08	94,38	C
15	Shampoo "Palmolive" for hair color	5250	1,51	95,89	C
16	Shampoo "Timotey" henna 400 ml	4811	1,38	97,28	C
17	Shampoo "Palmolive" for light hair	3937	1,13	98,41	C
18	Shampoo "Niveya" for hair	2849	0,82	99,23	C
19	Shampoo "Daph" docr 200 ml	1312	0,38	99,61	C
20	Shampoo "Antoshka" club 320 ml	1239	0,36	99,96	C
21	Head & Sholders shampoo basic care 200 ml	132	0,04	100,00	C
TOTAL		347 712	100

It will also be justified to reduce the range of items that fall into category C, getting rid of "unstable" products. We are talking about those items whose sales fluctuate greatly from month to month. It is quite risky to bet on it, since in the event of an unfavorable development of events, this threatens with overstocking of warehouses, an increase in volume illiquid assets in short, significant losses for the company.

XYZ analysis is used to assess the stability of sales. For each product of the analyzed group, the coefficient of variation is calculated (shows the degree of deviation of the data from the mean) according to the following formula *:

where x i is the volume of sales for the product for the i-th period;

x is the average value of the sales volume for the analyzed product;

n is the number of periods.

As the initial data, data on sales of goods of the group for several periods are used. The sales volume can be calculated in rubles or in physical terms. Although the latter are preferable. The fact is that if a company has recently revised its pricing policy, then the results will turn out to be incorrect.

Two important remarks. First, the number of periods must be at least three. Secondly, for products with a pronounced seasonality, the period should be longer than the seasonal cycle. Another option is to use periods within seasonal ups (or downs, respectively).

• X - variation does not exceed 10 percent. Stable sales, therefore, the main concentration of efforts and resources. Such a product does not promise large losses for the company, even if it is purchased (produced) in a larger than required volume:
• Y - variations in the range of 11-25 percent. A less stable category than X, however, a fairly reliable product;
• Z - Spread is greater than 25 percent. It is better to remove such a product from the assortment or work with its deliveries (production) under the order.

* The formula for calculating the coefficient of variation can be otherwise represented as the ratio of the standard deviation to the mean value of the indicator. In Excel, the standard deviation is easy to calculate using the "STDEV" formula.

Table 3. Analysis of the stability of sales

Name of product	Sales volume, rub.			Standard deviation	Average value, rub.	Variation,%	Group
	IV quarter. 2008	I quarter. 2009	II quarter. 2009
3	116 285	114 926	116 195	760	115 802	1	X
5	47 818	50 697	48 299	1542	48 938	3	X
1	305 922	276 658	335 817	29 580	306 132	10	X
6	34 500	27 865	32 289	3379	31 551	11	Y
8	37 929	36 685	30 750	3837	35 121	11	Y
2	255 420	245 089	327 870	45 108	276 126	16	Y
4	79 036	48 999	102 571	26 851	76 869	35	Z
7	12 346	33 786	32 502	12 025	26 212	46	Z

What to do with products that bring in 35 percent of the profit

So, we have sorted out the groups classified as category A. A slightly different approach is applied to the group of category B - in table. 1 is "snacks", "soft drinks", " confectionery". The sequence of actions can be as follows.

1. ABC -analysis of names. It is quite justified to carry out an analysis immediately by heading for the entire group, without dividing it into subgroups. We will definitely leave the positions of category A. From the remaining products of categories B and C, we will have to complete the assortment.

2. Highlight related products. If you focus only on profitable positions and there will be no related products (services) in the assortment, then instead of an increase in sales, revenue will decrease.

There are two ways to identify them. First, expertly, you can make such a sample of the marketing department employees. Second, related products can be found through cross-analysis. It is done by analyzing checks in retail or by analyzing invoices in wholesale companies. The bottom line is to collect data on which kits are most often purchased. And as a result, leave in the assortment those items from category B, which are most often purchased with category A.

3. Maintain stable products. Based on the XYZ analysis discussed above, the X and Y categories are returned to the assortment. Regardless of sales volume and profit margins. If some goods from month to month consistently (plus / minus 10-25%) bring the company, albeit small, but profit, most likely it will be inappropriate to refuse them.

4. Increase the depth of category A groups. For example, in the “juices” group, cherry, orange and apple juices of the same brand are in category A. This is a reason to add other flavors of the same brand from category B.

5. Determine the commodity items that are significant for the buyer. There are several categories of goods that should be in stock:

• "Traffic generators" - a category that provides a flow of buyers. These are products with a high purchase frequency. However, they do not necessarily bring significant profits. But, coming for them, clients simultaneously acquire other names;
• "Cash generators" - a category that provides a large volume of sales, that is, has the maximum turnover in the range of the group. In conditions of liquidity shortage, it is important not to exclude it in the pursuit of profit;
• "Defenders" - a category of goods for which the buyer makes a conclusion about the general level of prices in the organization. For example: "milk is cheap, butter is cheap, so you can look at the price tags with less preference." As a rule, it is built on a price basis from goods - traffic generators.

The marketing department will also insist on the sale of branded goods. Whether this is justified in category B depends on the strategy of the business and its resources.

Personal experience
Dmitry Ivanov, main CFO Wimm-Bill-Dann
Wimm-Bill-Dann has over a thousand product names. Of course, the idea of ​​reducing and optimizing the assortment would seem obvious. This will automatically lead to a decrease in the required warehouse space, eliminate the need to work with a huge number of items of purchased raw materials, equipment changeovers, illiquid stocks, etc. But we must not forget that the products support each other. You cannot significantly reduce the number of titles and not lose in revenue. When a customer comes to a store, he should see, for example, ten varieties of juices. If there are fewer of them, you take up less shelf space, you are less visible, you are bought less and, therefore, your market share is less.
But I would not categorically declare that it is impossible to reduce the assortment. This can be done, but with extreme caution. If you are reducing the number of juice names from ten to nine, that is perfectly acceptable. This is exactly what we periodically do, trying to save our "penny". But reducing the assortment to six types of juices is a big mistake. Such optimization will have an extremely negative impact on the company's revenue. In other words, you can cut back relatively painlessly. product line by 10, but not 20 percent.
In Russia, more than 25 percent of the market is for orange juice, about 25 percent for apples, the same for tomatoes, and peach and other flavors take up the rest. Nevertheless, we cannot limit ourselves to these names. Our consumers want variety.

6. Save new items. Regardless of the situation in the company, its financial position will be more stable if new products periodically appear in the assortment. The fact that some product is profitable now does not mean that tomorrow customers will lose interest in it. In order not to miss the moment of "degeneration" of "profit generators", it is necessary to regularly analyze the results of sales for the main goods (at least once every six months or a year), to monitor changes in the share in profits.

All other products from category B can be thrown away and disposed of.

What products to exclude from the assortment

Groups of category C, identified during the preliminary analysis of the assortment, it is better to completely exclude and not spend on them financial resources companies. Of course, making exceptions for items that have recently appeared in the assortment, are important related products for category A, traffic generators.

Expert opinion
Alexey Fedoseev, general director group of companies "Intalev"
A number of our clients deliberately reduce the number of brands and suppliers they work with. And all this is done in order to optimize costs as much as possible. Many people finally began to consider the prime cost of commodity items, which practically no one had done before. It turned out that a number of brands are very expensive. One client of ours reduced the number of suppliers of non-fungible goods from 150 to 10. In the next quarter, he lost 20 percent in revenues, and profits fell 3 percent. This was done last year, even before the crisis. But due to the fact that the company began to work more efficiently with the remaining customers, profit in the next quarter increased by 40 percent.
The company began to receive large discounts for purchase volumes and large rebates. Of course, the measure was risky. Throughout the quarter, the company's managers were, to put it mildly, in a state of stress - whether it will work or not, they constantly monitored the market reaction. But in the end, the reduction in the assortment paid off, although one or two commodity items were returned at the end of the quarter.

ABC analysis technique

The idea of ​​the ABC analysis method is based on the Pareto principle: "a relatively small number of reasons are responsible for the majority of possible results", currently better known as the "rule - 20 to 80".

This method of analysis has been greatly developed due to its versatility and efficiency. The result of ABC analysis is the grouping of objects according to the degree of influence on overall result.

An example of MSExcel spreadsheet (packed in Zip 19Kb format) for ABC analysis. The initial data are the results of the activity of the regional retail network for the 1st quarter of 2002.

First step: Define Analysis Objects

Second step:

Average inventory, rubles; Sales volume, rubles; Income, rubles; Number of sales units, pcs; Number of orders, pcs. etc.

Third step: Sorting the objects of analysis in descending order of the parameter value.

Fourth step: Definition of groups A, B and C.

To determine the belonging of the selected object to the group, you must:

3. Assign group values ​​to the selected objects.

Group A- objects, the sum of shares with a cumulative total of which is the first 50% of the total amount of parameters.

Group B- objects following group A, the sum of shares with a cumulative total of which ranges from 50% to 80% of the total amount of parameters.

Group C- the remaining objects, the sum of shares with a cumulative total of which ranges from 80% to 100% of the total amount of parameters.

I highly recommend getting creative with the definition of objects and analysis parameters. Feel free to experiment. Having grouped the product by one parameter, compare the result with other parameters. Group C can bring you 20% of the income, make up 50% of the inventory and occupy 80% of the warehouse area.

Example:

ABC analysis of goods by sales volume shows which goods provide 80% of the Company's turnover. Analyze the same goods, but by the number of units (or the number of orders for them) and as a result you will receive 20% of the goods purchased by 80% of customers, and this is already the attractiveness of the goods for the client and the Company's turnover. The same result can be used when planning the placement of goods in a warehouse or in trading floor store. The analysis of goods by income will show where you make money, a similar analysis by costs will allow you to understand where you are spending it.

If you are selling tiles or clothing, and it is difficult for you to collect data on nomenclature items, make an analysis by collection, and then within the collection ..

Important: Remember, an ill-conceived reduction in group C products (20% of the company's income) will lead to the fact that after a while the remaining goods will be distributed according to the same law, but the overall result of your activities for the company may decrease by 50%.

MethodologyXYZanalysis

The main idea of ​​XYZ analysis is to group the objects of analysis according to the homogeneity of the analyzed parameters (according to the coefficient of variation).

Formula for calculating the coefficient of variation:

, where

x i- i- th period,

NS

n - number of periods.

The square root value is nothing more than the standard deviation of the variation series. The larger the standard deviation, the further from the arithmetic mean the analyzed values ​​are. The standard deviation is the absolute measure of the dispersion of the variations in a series. If the standard deviation is 20, then with the arithmetic mean of 100 and 100,000 it will have completely different meanings. Therefore, when comparing the series of variations, the coefficient of variation is used. The coefficients of variation of 20% and 0.2% make it possible to understand that in the second case the values ​​of the analyzed parameters differ significantly less from the arithmetic mean.

Sample MSExcel spreadsheet (packed in Zip 26Kb format) for XYZ analysis. The initial data are the results of the activity of the regional retail network for the 1st quarter of 2002.

First step: Define Analysis Objects

Customer, Supplier, Product group / subgroup, Nomenclature unit, etc.

Second step: Determine the parameter by which the object will be analyzed

Average inventory, rubles; Sales volume, rubles; Income, rubles; Number of sales units, pcs; Number of orders, pcs., Etc.

Third step: Determine the period and the number of periods for which the analysis will be carried out.

Week, Decade, Month, Quarter / Season, Semester, Year

This method of analysis makes sense if the number of analyzed periods is more than three, the greater the number of periods, the more indicative the results will be. Moreover, the period itself should be no less than the planning horizon adopted in your company.

For example: Analysis of milk and bread sales in retail store can be posted by the amount of sales per week. Deliveries are made every day, sales too. But if we compare the sales of milk and Absolut vodka (which is ordered once a month and sold 1 bottle every 2 weeks), the result will be less indicative. With such a period, 99% of the assortment of the store will fall into the category "Z", 1% in the category"Y", And what conclusion can be drawn? You work in extreme conditions in an unpredictable market? In this case, it would be optimal to analyze monthly sales.

A more interesting situation arises when analyzing sales and inventory in companies that trade household appliances, building materials, spare parts for cars, etc. Financial plan in a company it is often compiled for a month, but the really necessary planning horizon should be for six months. Analyzing data with a period less than a quarter just doesn't make sense. All products fall into the category "Z».

Fourth step: Determine the coefficient of variation for each object of analysis.

Variation coefficient formula:

where,

x i- parameter value for the evaluated object for i- th period,

NS- the average value of the parameter for the evaluated object of analysis,

n- number of periods.

1. Do not try to write the entire formula in one cell, break the formula into multiple cells.

2. Squaring -^2, root extraction -^0,5

Example formula of the radical expression = (( C3-G3) ^ 2 + (D3-G3) ^ 2 + (E3-G3) ^ 2) / 3,

Then extracting the root and dividing by the mean -= H3 ^ 0.5 / G3

3. Pay special attention to the objects of analysis that have periods with zero values. Either exclude them from the analysis, or change the formula for calculating the coefficient of variation.

INMSExcelThere are a couple of standard formulas that make life much easier:= QUADROTKL (range up to 30 values) Is the numerator of the radical expression

The whole formula will take the form:= (SQUARE (C3: E3) / AVERAGE (C3: E3)) ^ 0.5 / AVERAGE (C3: E3)

and= VARP (range up to 30 values) - this is all a radical expression.

Now the formula will become quite compact:

= VARP (C3: E3) ^ 0.5 / AVERAGE (C3: E3)

The easiest option:

STDEVPA (C3: E3) / AVERAGE (C3: E3)

pay special attention for the presence of zeros in the cells. If a cell is filled with zero, then this cell is counted as significant. If the cell is empty, then it is not taken into account in the calculation. If zero is the objective value of this parameter, it should be left. If the product appeared in the analyzed period, then the cell can be made empty and then only the necessary periods will be included in the calculation. In other words, you have the opportunity, without rewriting the entire formula, to change the value -n(number of periods)

Very comfortably , for reference, add a cell with a formula -= COUNT (D3: F3) , and to get background information over how many periods the value of this coefficient of variation was calculated.

Hope, this will make it easier for you to carry outXYZ- analysis.

Fifth step: Sort the objects of analysis in ascending order of the coefficient of variation.

Sixth step: Definition of groups X, Y and Z.

GroupX- objects for which the coefficient of variation does not exceed 10%.

GroupY- objects, the coefficient of variation for which is 10% - 25%.

GroupZ- objects, the coefficient of variation for which exceeds 25%.

Combining the results of ABC andXYZanalysis

First step: Conduct an ABC analysis

Before starting ABC analysis, create an index field, i.e. a cell containing a numbering that does not change during sorts. At the end of the analysis, "embed" the values. Copy the cells containing the formulas and use: menu "Edit", "Paste Special ...", "Paste,Values ​​".

Second step: Conduct XYZ Analysis

Before the beginningXYZanalysis, create an index field the same as in ABC analysis (or do both analyzes in one file), i.e. a cell containing a numbering that does not change during sorts. Once the analysis is complete, you can "embed" the values, but this is not required.

Third step: Combine the results

One of the most common assortment analysis methods is ABC analysis, which is based on the Pareto principle. The latter, in turn, states that 80% of the causes are responsible for 20% of the consequences. This rule, which is also called 80/20, means that in any process a small percentage of causes (20%) are vital, and the remaining causes (80%) do not have a serious impact on the final result.

This rule can be applied to various socio-economic phenomena and business processes. For example: 80% of the work is done in 20% of the time; 80% of the company's profits are provided by 20% of customers; 80% of the warehouse area is occupied by 20% of inventory; 80% of sales are provided by 20% of sellers; 80% of problems are caused by 20% of defects, etc.

In each specific case, this proportion may deviate from the exact parameters 80/20 and be 90/10 or 65/35. However, the essence of this does not change and lies in the fact that the largest percentage of the result is provided by a lower percentage of costs.

As part of the assortment analysis, this rule says that 20% of goods provide 80% of the store's turnover.

ABC analysis method will allow you to determine those 20% of goods that are priority for the store. In fact, this method assumes ranking the trade assortment according to various parameters. Traditionally, the entire assortment is divided into three groups of goods, depending on their contribution to the turnover and profit of the store:

1) goods of group A - the most important goods that provide the first 50% of the results;

2) goods of group B - goods of medium importance, providing another 30% of the results;

3) goods of group C - the least significant goods providing the remaining 20% ​​of the results.

Traditionally, ABC analysis is carried out in three stages.

Stage 1. Determination of the object of analysis and parameters of its assessment.

The object of analysis can be:

product groups;

commodity items;

suppliers.

Evaluation criteria can be:

sales volume (in kind and in monetary terms);

turnover;

inventory, etc.

Stage 2. Compilation of the list of objects of analysis in descending order of the value of the selected parameter.

Stage 3. Determination of groups A, B and C, for which it is necessary to calculate the proportion of the parameter from the total sum of the parameters with a cumulative total and distribute the objects of analysis into groups in accordance with the obtained values.

Consider ABC analysis by product group on specific example a grocery store operating in the “shop near the house” format.

Stage 1. Within the framework of this analysis, the object of research is product categories, including:

milk products;

bakery products;

confectionery;

alcoholic beverages;

fruits and vegetables;

meat, sausage;

juices, water.

To conduct an analysis, it is necessary to systematize information on the turnover of each product group for certain period(month, quarter, year). After that, it is necessary to calculate the share of each product group in the total volume of the store's turnover (table).

Stage 2. Sort product groups in descending order of their share in the turnover.

Turnover of commodity groups for the 1st quarter of 2010

table

ABC analysis of the assortment of the store

Thus, the conducted ABC analysis shows that the main share of the store's turnover is provided by such product groups as dairy products, bakery products and meat, sausages, which belong to group A. Product groups - juices, water and fruits, vegetables belonging to the group С, need development and require additional sales promotion actions, for example, in the form of price reductions or expanding the range.

However, ABC analysis by product group gives only a superficial idea of ​​the structure of a store's assortment. For a more detailed and in-depth analysis, it is advisable to conduct a similar analysis by product categories within product categories, as it is more informative and allows you to manage the assortment more efficiently.

You can use XYZ analysis to explore the assortment in more depth. It allows you to create a more complete picture of the shopping process in the store.

XYZ analysis also involves dividing the assortment of the store into groups X, Y and Z, while the criterion for this analysis can be the profitability of goods or the stability of their sales.

Moreover, most often the need for XYZ analysis arises when the store is faced with the task of analyzing the assortment simultaneously by several parameters and presenting its result in the form of a matrix.

Consider the combination of ABC and XYZ analysis using the example of a grocery store.

The following product groups are the object of research:

milk products;

bakery products;

meat and sausages;

confectionery;

alcoholic beverages;

fruits vegetables;

juices, water.

The combined analysis includes the following steps.

Stage 1. Conducting ABC analysis. As a criterion for dividing goods into groups within the framework of this analysis, the share of the product group in the total turnover of the store is used. The analysis results are presented in table.

ABC analysis

Stage 2. Conducting XYZ analysis. At this stage, the criterion for dividing goods into groups is the share of the product group in the gross profit of the store. The results of the XYZ analysis are presented in table.

table

XYZ analysis

Stage 3. Combined analysis (ABC- and XYZ-). It is necessary to combine the results of ABC and XYZ analysis, as a result of which the entire assortment of the store is divided into 9 segments based on two criteria - the share of the product group in the store's turnover and the share of the product group in the profit of the store. The results of the combined analysis are presented in table.

table

ABC and XYZ analysis

The analysis made it possible to identify the most profitable and least profitable groups of goods for the store.

So, the most profitable commodity groups for the store are AX, BX, AY - the groups that provide the greatest contribution to the turnover and the formation of the store's profit.

Groups requiring special interventions to improve their effectiveness include BY and CY. These groups have sufficient potential, but in order for them to move to the group of profitable goods, it is necessary to optimize the assortment and pricing policy in relation to these groups.

Finally, the BZ and CZ segments are among the least profitable and poorly traded goods. These product groups require special attention from the store management.

A successful business in many cases depends on the correctness of working with numbers. This can occur both at the level of the simplest calculations in the course of comparing "debit" and "credit", and in the aspect of complex, multilevel analytical calculations. These experts include ABC and XYZ analysis. What are these methods? What are their practical significance? How to use them correctly?

General information

What is ABC Analysis? This is understood as a method by which one can classify a particular resource, depending on the degree of its importance. The basic principle used in this type of analysis is the Pareto rule. In the generally accepted interpretation, it sounds like this: 20% of actions bring 80% of the total result.

In relation to ABC analysis as such, this principle can be interpreted as follows: reliable control of 20% of a certain system (as an option - sales or enterprise management) determines its efficiency by 80%.

ABC analysis implies the classification of certain operations or sections of a resource by dividing them into several categories (depending on the degree of value) - A, B and C. Type A includes the most valuable of them (the ones that bring 80% of the result, and their, respectively, 20%). Type B actions are "mediocre", 30% of them, and they provide 15% of the result. The activities of type C, in turn, are the least valuable. Despite the fact that there are 50% of them, they give only 5% of the result.

Analysis methodology

The practical use of such a tool as ABC analysis, by and large boils down to making a "rating" of the usefulness of certain actions. The criterion here, as a rule, is statistical information or expert assessments to identify the "most valuable" transactions.

As a rule, in the course of the ABC analysis, you can build graphs, the X axis of which will be the number of actions, and the Y axis - the performance indicators. Thus, you can calculate which activities will be the most effective. This kind of graph is sometimes referred to as Pareto curves. As soon as the researcher ranks the effectiveness of all actions, a statistical analysis is carried out, the calculation of the most useful activities for all charts, and, as a result, the formation of their final "rating".

Analysis sequence

In what order should the ABC analysis be performed? Experts recommend sticking to the following algorithm:

1. We pose the main question. The effectiveness of actions in relation to what result we are interested in in this case?

2. We select activities that are most relevant to the task at hand.

3. We draw up graphs for each of the actions in comparison with the performance indicators of each.

4. Choose 20% of the most effective, 30% - mediocre, 50% - least significant.

The specific methodology for each of the four items can be selected based on the purpose of the analysis. In some cases, an entrepreneur, for example, wants to show the investor that such and such a product sells better, and it is necessary to invest more actively in it. Another option is to analyze the feasibility of allocating resources allocated to certain purchases. Also, the purpose of the ABC analysis may be to identify the effectiveness of advertising aimed at "promotion" of certain types of goods.

The practical benefits of analysis

How can the analysis in question be useful in practice? There are many options here. Take the sales area. Let's say we need to identify which of the commodity items bring the most revenue. A competently conducted ABC analysis of sales will allow us to find not just a scattered list of well-selling products, but 20% of them, which provide 80% of the profit. The situation is similar with the service sector. ABC analysis of customers can help find those 20% of service consumers, on whose activities 80% of revenue depends. It's the same with industry. ABC analysis of stocks of raw materials or semi-finished products will identify 20% of their varieties, which are used in 80% of the volume of production, and therefore are the most valuable. That is, those who need to be given priority in the procurement and distribution of capacitive resources in the warehouse.

We can see how versatile ABC analysis is. There are more than one example of its use. The areas that are compatible with the application of this technique are very different.

XYZ analysis

There is one more method that complements the research on the ABC methodology - XYZ analysis. What is it? It is believed that this type of study makes it possible to classify the reserves available in the company depending on the intensity of their consumption, as well as forecasting the dynamics of the emergence of needs for them in relation to a specific time cycle. What does it mean?

Resources are classified in three categories - X, Y and Z. Those that belong to type X have stable consumption dynamics, minimal adjustments over time, and, as a result, their consumption is quite easy to predict. As a rule, the difference between the minimum and maximum consumption indicators recorded within time periods does not exceed 10%, or even tends to zero.

Resources of type Y, in turn, have a noticeably less stable consumption dynamics, however, they are still quite well predictable. The difference between the minimum and maximum indicators is 10-25%.

Resources categorized as Z are characterized by highly volatile consumption dynamics. There are no pronounced trends, it is difficult to predict something. The values ​​of the minimum and maximum consumption indicators for a time period may differ by 25% or more.

An interesting fact is that the same resource can belong to different categories in different periods of measurements. This can be predetermined, for example, by the season, yield, or the specifics of demand. For example, tangerines are traditionally sold well in stores in winter. But the specific dynamics of their implementation during the winter will most likely be different. Between, say, early December and the 20th of the month, tangerines are likely to be classified as a type Y product - with relatively stable but variable demand. However, due to the fact that this fruit is very popular in New Year, then from the 20th of December to mid-January it will most likely be sold at a constantly high rate, which will make it a resource of type X. this product becomes close in criteria to the category Z.

Combination of two tests

ABC-, XYZ-analysis can be combined. Moreover, in many cases, the study will be incomplete if each method is used separately. How to carry out sequential ABC-XYZ analysis? We will now consider an example of an algorithm suitable for this purpose.

Let's say we are faced with a task: to analyze the range of grocery products in order to determine which units bring the most revenue and which of them are characterized by the most stable demand. In the first part of the study, the ABC analysis of the assortment will come in handy, in the second - already XYZ. How to proceed? What results can we have in both cases?

First, we identify the best-selling product for, say, the past month. We take data from a CRM system or other kind of accounting source, reflecting the number of units sold by day. We reveal that 80% of all revenue came from sausage, chips and carbonated drinks. These are goods of group A. Next, we look at how many receipts for each of the commodity items are punched on each day of the month. It may turn out that soda was sold in the amount of 100-102 units per day. Sausage - on one day - 50, on another - 153, on the third - 10, on the fourth - 181 units. In turn, the results for chips may show that this product was sold like this: on the first day 80 units, on the second - 125, on the third - 91, on the fourth - 114. It turns out that among the goods of group A, soda is the most stable, and it can be attributed to category X (and therefore it is safe to purchase from suppliers under favorable terms for sale). Chips are a product with an average stability of demand, it will belong to group Y. Sausage is a product of group Z, the dynamics of sales of which often changes.

Similar procedures can be carried out for goods of type B and C. Experts recommend, based on the results of a comprehensive study of the assortment, when the ABC analysis method is combined with the XYZ method, to single out the leading goods (which will belong to the AX type), as well as outsider positions ( classified as CZ). In addition to them, you will get 7 more products (in total - 9 possible combinations, 3 squared, and when measured in different periods, when the dynamics of sales of the same products can change, the total number of options can reach 27, 3 to the 3rd power) ... All of them can be ranked and a "rating" can be made that reflects the combination of profitability and sales stability. For the convenience of calculations, we can try to carry out XYZ-, as well as the previous ABC-analysis in Excel. The example we've looked at is simple enough that we can use simplified tools like a spreadsheet.

Practical usefulness of classification into groups X, Y, Z

We noted above that, having identified the most profitable and most stable product, we can adjust the policy of relations with suppliers. However, this is not the only advantage of XYZ analysis. How else can the results of such a study help us? Let's consider the specifics of their practical use in comparison with each of the three groups of goods.

So, the products of type X are characterized by the most stable demand. The most important criterion for the usefulness of having such information is inventory planning. We can establish interaction with suppliers in such a way that our warehouses are used as efficiently as possible. We will know exactly how long the products of group X will be there from the moment they are loaded until they hit the counter. Consequently, we will be able to plan the delivery of less dynamic, from the point of view of demand, positions Y and Z so that they always have where to place them.

Priority in procurement

Group Y goods are characterized by relatively stable consumption dynamics. Main function of such products - to support the main demand generated for goods of group X. In some cases, correlations are possible, reflecting the dependence of the dynamics of demand in class X on the availability of products of type Y on the shelves. Analysts probably believe that the psychological aspect plays a role here. A customer who sees empty shelves - let's take the case when goods of group Y are not presented by a retailer - hesitates to make purchases in such a store even for those items that are usually characterized by stable demand. In turn, if there are enough products of type Y, then the demand for goods X is "warmed up". The main task for the store owner in this case is to ensure optimal utilization of warehouse capacities, to find the ideal combination between the cost of purchasing auxiliary Y-positions and the real economic effect their presence on the shelves.

In turn, the products of group Z are difficult to optimize in terms of warehouse management. Their direct influence on the dynamics of sales of "flagship" goods of type X also may not be. And therefore, experts recommend that they give them a minimum place in the total volume of purchases. Or, alternatively, replace them with new products, goods that have not yet been tested on the market. In this case, at least there will be a possibility that the fresh brands that have appeared on the counter will grow from the Z category into more significant ones from the point of view of sales stability.

Play in your "league"

Let's make a reservation right away: when interpreting the result of the analysis, it should be understood that, say, goods of group Z belonging to category A (and this is unusual integrated analysis) will be more valuable than products of type X for category B. Moreover, their direct comparison is not entirely correct - it's like, relatively speaking, considering the capabilities of football teams from different leagues. Therefore, when analyzing the prospects for goods in categories A, B and C, it is wrong to linearly compare their distribution in groups X, Y and Z. Consistency in the interpretation of product results in relation to their "leagues" is important.

So, let's summarize:

Category X goods are "flagships" of sales, their purchase from suppliers must be stable, supply channels are established and, if possible, diversified (in the event of "sanctions" and other kinds of phenomena beyond the control of the business);

Class Y products must also be present on the counter, performing a supporting function in relation to X goods and stimulating general demand;

Goods of type Z can, if not be excluded from circulation, then try to replace experimental designs which could potentially acquire the status of products of categories X and Y.

All these conclusions take place provided that we are talking about the analysis of goods within the same group - A, B or C. As we said above, it does not make much sense to identify "averaged" indicators here.

Nuances of interpretation

Of course, this kind of recommendation is valid if only the results of the combined ABC-XYZ analysis can be interpreted unambiguously. The research methodology should be accompanied by multidimensional criteria that will make it possible to draw undeniable, from the point of view of statistics, conclusions regarding the sales prospects of a particular product. When we considered the question of how an ABC analysis can be carried out (an example with a sausage), we divided the products into the corresponding categories very conditionally. It's the same with the XYZ part. In practice, the analysis methodology is much more complicated. Moreover, researchers rarely carry out, as in our example, ABC analysis in Excel using calculations, in fact, manually. As a rule, much more complex analytical programs are used in order to minimize the probability of errors, since we are talking about real business, where mistakes are undesirable, unlike theoretical scenarios.

Work with non-target customers is. Your managers may be tech-savvy, have perfect scripts and work experience, but there will be no result if they are knocking on doors not to your customers.

ABC analysis: test to check the target audience

How to identify problems?

To see if you have such a problem, check out our checklist:

• Managers hold many meetings, but there is no result;
• Build a business from the experience gained in another type of business;
• There is no buyer qualification process;
• There is no definition of the target portrait of the client,
• No ABC analysis.

Even if you agree with one of these statements, build an ABC analysis.

Why do you needABC analysis?

ABC analysis is a study of the customer base in two directions: in terms of volume and frequency of purchases. ABC analysis allows you to understand:

1. Who pays you more and more often
2. What is the portrait of your customer
3. Are target customers entering your funnel?
4. Where to direct the main efforts of managers
5. How to increase shipments to new customers

ABC analysis: algorithm for its implementation

The study will show on the ABC axis the distribution of buyers by revenue, on the XYZ axis - we will see them broken down by the frequency of transactions. You will see immediately,.

As a result of the ABC study, 20% of counterparties c should be included in group A the largest volumes purchases, in B - 60% with average purchases, in C - 20% with small. The same should be done by XYZ, placing those who most often contact you in the X group, in Y - irregularly, in Z - make single purchases.

ABC analysis: building a portrait of the target client

What will the ABC analysis results tell you?

After conducting ABC analysis, we look at the intersection of ABC and XYZ, identify the most interesting buyers for the company.

1) Those who got into groups A and B, provide the bulk of the proceeds. There should be as many of them as possible in the company.

2) AX and BX provide a significant volume of purchases with regular repeated calls.

3) AY and BY bring you good revenue, but return to you is not stable. Come up with bonus programs for them, regularly remind about yourself, report on promotions and new products.

4) AZ and BZ, despite the decent volumes of purchases, they return to you unpredictably.

5) Those who are in group C, it is worth revising and partially abandoning them.

6) With participants CX and CY work on increasing the average check.

7) B CZ the most uninteresting buyers for you get there. You shouldn't waste your managers' time on them, refuse to work with them.

ABC analysis: changing the business process of attracting buyers

What to do next with ABC analysis?

Raise the majority of buyers in the category, A

Based on the results of the ABC analysis, plan tasks for each of the target counterparties and measure their capacity:

Implement a buyer qualification process. It is important to understand that ABC analysis must be done regularly and based on its results, indirect criteria should be drawn up to determine the portrait of your target audience(for examples of the criterion, see the article). They will help you determine at the entrance whether it is suitable potential buyer for your target audience, is it worth spending time on it. Add this stage to yours.

Based on the results of ABC analysis, create additional fields according to the portrait of the target audience in CRM and prohibit moving from stage to stage without filling them in.

It is possible to carry out ABC analysis by individual segments: by partner and retail channels. It will be useful to look at this analysis for the product line.

This will kill two birds with one stone: you will be able to draw a conclusion about which products bring you the most revenue and which ones you need to give up. ABC analysis by segment of counterparties will allow you to find out what you need to offer them here and now.

ABC analysis: studying the product range

ABC customer analysis is just one of the slices for researching the current base. There is another approach - ABC stock analysis.

Traditionally, the importance of a product is identified by 2 parameters - sales volume and profit. Again, the Pareto rule is accepted as the general hypothesis. According to him, 20% of assortment positions provide 80% of profits.

A huge number of companies in the world have repeatedly carried out ABC analysis of the assortment. Conclusions almost always boil down to this ratio:

• 10% of assortment items (group A) account for 80% of turnover;
• 15% of assortment items (group B) account for 15% of turnover;
• 75% of assortment items (group C) account for 5% of turnover.

Assortment analysis

Taking all this into account, the entire assortment can be divided into groups according to the degree of importance:

• group A - the most significant goods that make up the most valuable and working part in assortment;
• group B - goods of the average level of significance;
• group C - the least significant goods. In fact, these are “candidates for relegation”. New products sometimes fall into this group. Their significance usually requires additional confirmation over time.

It is important to understand that the ranking of products by groups can occur according to different criteria depending on the goals.

Goal 1: reduce the assortment. In this case, the goods are analyzed in terms of sales volumes and profitability.

Objective 2: study profitability. The goods are distributed according to the level of profitability and the turnover ratio.

Goal 3: Reduce inventory maintenance costs. Here we analyze the assortment in terms of turnover rates and occupied warehouse space.

ABC analysis: looking at the assortment

In businesses of all kinds, experts note the effectiveness of such a tool as ABC assortment analysis.

An example for a cafe. Here ABC analysis is needed to optimize the menu. To do this, the share of each position in profit and turnover is found out. For this, a specific research algorithm should be used:

1. A special table is formed, in which data is entered on the cost price, selling price and the number of sales for a month / half a year / year for each product.

2. Using a formula, products are ranked on a scale from 1 to 100, depending on their share in profits and turnover.

3. Items are categorized into groups A, B and C.

If the position on this scale is in the range from 1 to 50, then this is category A. If the assortment group falls within the range from 50-80, then it is entered in B. Everything that is below the line - "80", is "non-liquid ".

The goods that ended up in groups A and B are significant and effective, as they bring profit and make up almost the entire turnover of the cafe. Group C must undergo "recuperation" or elimination. In the first case, it is necessary to stimulate demand. In the second, hopeless case (usually, when a position is not on the outsider list for the first time), such a product should be disposed of.

ABC analysis: supplementing it with XYZ analysis

In addition, XYZ analysis is used as a complementary tool and to obtain a clearer picture.

XYZ analysis is a tool that divides products according to the level of fluctuations in their consumption and the regularity of shipment.

A somewhat more complex method is used here, which involves obtaining the so-called coefficient of variation. Simply put - flow rate fluctuations. The coefficient itself shows the deviation of the flow rate from the average value and is expressed as a percentage.

The parameters are:

• volume (quantity);
• sum;
• the amount of the sold margin.

As a result, we get the same distribution of goods, but in other categories - X, Y, and Z. Products in these categories are characterized by the degree of stability of their behavior.

So in category X there are assortment groups, whose deviation varies from 5% to 15%. Sales for these items are easily predictable and easy to plan, since they have a stable consumption value.

In the Y category, everything that received a coefficient of variation from 15% to 50% turns out to be. Sales for such items are more difficult to predict. And here, most likely, we are talking about goods with seasonal demand.

Combining the results of ABC and XYZ analysis

Combining both types of analysis (ABC and XYZ) has 3 important advantages.

1. You can optimize your management structure commodity stocks based on the most reliable data.
2. You know the share of which product should be increased in your assortment matrix.
3. You understand how to reallocate personnel depending on qualifications and experience: who should deal with one product, and who should be transferred to another.

ABC analysis: monitoring migration

It is important to remember that ABC analysis will provide you with an “X-ray” snapshot of the status of your current customer base. You should use this tool regularly and then you will see the dynamics of the movement of counterparties and products from category to category, the so-called migration.

If a company operates in the B2B segment, then you can track migration across several sections.

1. Migration by buyer
2. Migration by product
3. Migration by managers
4. Change in the quality of the seller's portfolio by the volume of shipments by ABC
5. Control of the regularity of purchases by XYZ in the employee's portfolio

If a company operates in the B2C segment, then migration should be monitored by product. Initially, as a guideline for identifying categories, take the consumption rates of a particular product, depending on average salary.

ABC analysis: how to increase profitability

It should be noted the high degree of practicality of such a technique as ABC analysis. The company's marketing and strategic branding decisions also depend on the conclusions of this study. Let's give an example.

Problem

Oy-li's client was importing components for lighting equipment from China. The main buyers are manufacturers and wholesalers. The company was faced with the task of reducing the risks of dependence on a single Chinese supplier. Therefore, it was decided to sell the imported goods under its own brand. So it would be possible to get rid of the "dictate" of the supplier and change it, if necessary, without explaining this step to your counterparties.

Such a strategic and useful in every sense decision caused displeasure on the part of the mediators. They opposed him because they believed that their customers would go directly to the brand holder.

In order to understand who really is a significant customer and whether it will be possible to do without some of them in the future, an ABC analysis of suppliers - consumers of the importer's products was carried out.

As expected, the proceeds were distributed between two groups of buyers: manufacturers of lighting equipment - 60%, intermediaries - 40%. At first glance, in such a situation, abandoning the "rebellious" resellers would be tantamount to closing the business. They began to analyze who buys more often and more.

But the intermediaries were in the AY category. This meant that although they cover good volumes, their stability leaves much to be desired.

Experts at Oy-li found this situation alarming. And that's why. Reseller counterparties supply components to the same manufacturers. Logically, their purchases should also be uniform in order to meet the regular needs of manufacturing customers. However, this did not happen. Orders came in at unpredictable intervals. And sometimes just from case to case.

Solution

To understand what was going on, we decided to conduct an ABCXYZ analysis of the assortment matrix. And here a very interesting detail came to light. The study showed that intermediaries never purchased a strictly limited set of components from which the manufacturer could assemble final product... On the contrary, each time they made "chaotic" purchases: for completely different positions and irregularly. The obvious conclusion suggested itself: resellers used the services of other suppliers and simply ordered what they needed.

The study also demonstrated another unpleasant feature of working with intermediaries: the importing company had almost zero profit from interacting with them. The middlemen received too high discounts from the seller.

Result

Taking into account all the clarified circumstances, the Oy-li client was recommended to continue working towards its own branding, not focusing on the opinion of resellers. A set of measures was also developed that made it possible to reach new volumes with manufacturers.

Since all the solutions were implemented, the importing company almost immediately increased its profitability by 15% and had the resources to focus on working with manufacturers.

ABC analysis: conclusions

So, with the help of ABC analysis, customer base and assortment, which allows you to determine who and what brings the greatest result. The assessment will be even more informative, supplemented by an XYZ analysis, which will show the regularity of customer orders and purchases of a certain type of goods.

You can't do ABC analysis, see the results and calm down. It is necessary to make such cuts periodically. And on their basis, make efforts to ensure that some of the buyers or products migrate to more highly profitable categories. For this, buyers need to be stimulated - to offer promotions, bonuses, special conditions: if there is potential for customer development, this opportunity should not be missed.

It is also important to be able to part with some customers and some products. If they make rare and small purchases or are not in stable demand, then why spend efforts on them? It is better to direct actions to those groups that bring the best results.