Auto-generated product mix matrix

January, 2015
proto$ gen VIII mod 4B91B2 (Ultima)
The sales enterprise’s assortment matrix is generated by an automatic robot of dia$par that scans the consumer interests of millions of users through Yandex services. Everything that will sell and nothing that might get out of demand. Full recalculation several times a day for hundreds of goods groups and tens of thousands of items, without errors or corruption in the Procurement Department.

What’s the use of handsome and splendidly organized shops at prime locations if they offer nothing to buy?
No logistics, even if honed to the accuracy of an atomic clock, will save your business — unless the prospective client finds in your assortment the goods he needs at an affordable price.

Captain Obvious?

?Oh no.
From our practice we know many cases where corporate management, while sparing no effort on a lot of important (and much needed!) things, take Purchases (capitalized Purchases further understood to mean the group of assortment-related business processes — assortment matrix formation proper, pricing and, directly, purchases and logistics) for granted, with "market professionals" hired for a heap of money, phones ringing, goods receipts being typed, and truckloads of goods pulling up at the warehouse. In short, the office hums and clatters.

And yet there is nothing more important for a trade enterprise than Purchase.

A trade business manager is only well-advised to do anything other than Purchase if his assortment availability and price situation are at least very good.
Or perfect.

The manager of a trade company addressing other issues, however important in his opinion, while things are less than perfect regarding the assortment of required goods and their prices, is actually wasting his efforts. To put it simply, he’s barking at the wrong tree. (Of course if this isn’t about a fire or another visit by blood-sucking inspectors).

Our extensive practice of watching lots of money poured down the toilet has shown that there are no exceptions to this rule.

The base case solution to the problem of forming an ideal assortment comprises the following tasks:

  1. Forming the assortment matrix, i.e, the list of the goods, by category, that should ideally be in stock — taking into account the market needs in each individual sales region if the company has a sprawling geographical presence.
  2. Forming the price policy (pricing guidelines) and keeping the prices updated for each individual goods item — for each goods category and, generally each sales region.
  3. The practical organization of the purchasing business process, i.e. purchases from suppliers in the narrow sense. Bringing the actual assortment available as close as possible to the assortment matrix.
  4. The practical organization of the business process of replenishing shops" or sales outlets" assortment from the central warehouse / logistical hubs. This applies to geographically distributed network sellers.

Each section and its related issues will be illustrated with a series of Purchase cases, of which this is the opening text.

The concept of the assortment matrix is usually associated with the assortment of shops run by traditional retail companies, such as Magnit, Auchan, etc.

What characteristic features can be identified for traditional retailers" assortment?
The assortment changes very slowly. Do new "breakthrough" mayonnaises or beers appear very often? How often do there appear new categories of foods or individual revolutionary products that alter considerably the ordinary consumer’s diet? That is the question.
The prices are stable, barring not too frequent force majeure situations like embargoes on some products or the national currency plummeting.
The market shares among competing goods change very slowly if at all. Each category includes a minimal number of goods items in significant demand.
Demand is stable.
?Competition is minimal; virtually everything depends on the individual shop’s geographical location.

?The trick is the correct arrangement of goods in the sales space (not directly related to the assortment matrix) plus the optimal balance struck between sales-forming and profit-forming goods.

In general, there is nothing sophisticated about this, and the question is more than profoundly and extensively covered in business literature.

But the world is not limited to traditional retail with its essentially stable assortment, stable prices, terrific competition environment and guaranteed demand.

Some sectors are far less serene and favourable.
Just look at our beloved gadget retail.
?Here, dear raeders, we shall see complete nuts.

The assortment keeps changing kaleidoscopically. Whole categories of goods appear and die out all of a sudden.
The prices of the devices that generate the bulk of the sales keep falling rapidly — as a rule. But they may rise just as rapidly — driven by either the falling national currency or by wise and beneficial decisions of our government.
Any goods item that is scarce now due to high demand may stagnate in a month. This area features inter-species competition: thus, hardly anyone in the X5 food retail network has noticed the advent of a new brand of olive oil to bring down demand for mayonnaise — but this is quite common in the electronics market, with laptops losing their portion of demand to tablets. Digital cameras are rapidly dying out as smartphones gain ground.

Buyers" preferences are far more segmented; while three to five brands, the incumbent sales leaders, are enough to meet demand for vegetable oil, how many models of mobile phones will be needed? And the models are different every week.?

Competition is incredibly acute; the effect of Internet orders levels out the geographic remoteness factors (at least within an agglomeration), which is virtually inconceivable for trade in foodstuffs. Undertakings like the Utkonos ("Platypus") Internet foodstore, almost sure to run at a loss, are not considered; this is not and will never be a business.

This is enough for illustrating those harsh Martian conditions.

What about the assortment matrix here? The approach to its formation that has been working perfectly for decades in traditional retail will definitely bring forth garbage in this case.
Neither heaps of Harvard Business Reviews nor guru talk about ABC, XYZ and other XXX analysis will be helpful here (what a surprise!).

Among this market’s participants, the basic approach to forming the AM is a mere shot in the dark.
?That is, their high-paid purchasers and category managers (PCMs) are supposed to be competent enough in their goods categories to know what to sell. Anyway, nobody else in the company knows more about this.

This approach is not bad, in a sense — especially where nothing else remains.
But it inevitably precludes an objective assessment of both the assortment available and the purchasers" work quality.
In response to the management’s / sales personnel’s natural questions about assortment, the PCM will always cite tough excuses like: our suppliers have none, I refused to take at their bad prices, we have a cheaper and better analogue and so on: "Why are you messing me", "I know better what to sell", "they offered to supply this item much cheaper in a week", etc.

Category profit motivation common to PCMs, along with their presumably competent and rational conduct, lead one to expect that an individual PCM will make every professional effort to achieve an ideal assortment. Well, he is interested in this!

But the riot of human factors defeats strict logic. Which is demonstrated by a strikingly petty assortment in most electronics retail companies. NB: narrow doesn’t mean petty.

Why so?
Firstly, the very presumption of competence and reasonable conduct is incorrect. This is grim reality.
Secondly, "common profit motivation" does not motivate strongly in many cases — due to basic motivation building errors (a topic for a separate text) and because PCMs" profit percentage fee is based on their historic (meaning rather poor) results.
That is, an individual PCM receives a salary that has satisfied him/her from the outset. And if they need more, they can make it much easier in kickbacks than through really arduous work — studying their own category and the volatile demand situation and actively looking for things really needed by consumers from all suppliers.

What can we do about it?
?Here’s what.

We have a miraculous tool, called Yandex.Market, at our disposal.
It keeps a goods popularity rating for each category.

The goods mix rated at each specific moment is a good quality snapshot of the market situation — or buyers" interest, to be more precise.

The more goods from this ratings" top lines are present in your assortment (and adequately priced, of course), the more probable is the purchase made on your website. Or at your offline shop.
?This rule also works for buyers who do not use this rating (or even know nothing about it) and even for Internet non-users; due to its scale, the audience of Yandex is exceptionally representative of the consumer community and, accordingly, consumer behaviour in general.

Importantly, the Yandex.Market rating is not a direct projection of sales in quantitative terms (though not much different). It actually rates consumer interest.
?The fact of a goods item heading the Yandex.Market rating does not necessarily mean that it will lead your sales (though it is quite probable). That means that (a) you must have this item available and (b) it will certainly sell — if reasonably priced, of course.

Nor is the Yandex.Market rating a profitability projection; the first-ranking item may well sell at a loss, but having none in your goods mix will entail far greater losses. In all fairness we must note that if you’ve managed to scoop up some top-rated but scarce goods, you are welcome to profiteer. It will sell out.

Yandex.Market offers regional versions of its popularity rating, whose utility for network retailers cannot be overestimated.

In sum, the physical sense of the Yandex.Market rating for an individual goods category is as follows: the more top-rated goods are included in your assortment, the more probable is a prospective customer’s visit to your shop or website.
And what precisely you will sell him — that very item he came for or something similar but more marginal for your company, or either with accessories ten times as profitable in monetary terms as the basic item — will depend, in turn, on the quality of your website, shop and shop assistants. Good assortment will have done its customer attraction job.

Such are the theoretical fundamentals of using the Yandex.Market popularity rating to compose the assortment matrix.

?However, to consider just one top-rating item is not enough in practice. For each individual category, we must identify the Top X number: Top 3, Top 5 or Top 20.

The X logic is as follows: a number that is as small as possible but can cover 80 to 90% of sales.

In our above-cited traditional retail example about vegetable oil, the X number would be three.
The Top 3 brands from a hypothetical Yandex.Market popularity rating for a Sunflower Oil category would cover more than 90% of its sales.
A minimal X is required by practical logic: the fewer goods a PCM must keep available, the better is the result. Demanding a PCM to keep e.g. all the Top 50 goods available or die reduces his job to an absurdity.
As practice shows, with X higher than seven this tool loses much of its efficiency.
The limit of seven is also supported by neurophysiological studies of human short-term memory — similar to CPU registers.
To accept a lesser sales share for top-rated goods is the second worst alternative.

In sum, a typical goods category includes two types of SKUs:

  • Top X goods performing the role of consumer attractors and turnover generators; and
  • any set of analogous goods purchased at the PCM’s discretion (certainly with turnaround control), sold into the consumer flow generated by the first group — at a higher margin than the attractors.?

?It is clear that in practice both groups will overlap to some extent.

A goods category usually includes more marginal related goods (for instance, the Vacuum Cleaners category may include dust bags and brushes as well as the appliances). The goods category is thus extended to include a third type of goods — marginal accessories that generate a profit often many times higher, in monetary terms, than the margin received from the "basic" goods.

Clearly, for an efficient "attract with locomotives and earn on related goods" strategy it is vital to keep the "locomotive" assortment most attractive at any (!) cost.
?For which the above technique (we shall call it simply Top X) is of greatest help.

The use of the Top X technique produces good results with any goods category.
But the most impressive multiplicative effect is to be obtained in categories that includes goods central to the attraction of consumer interest.

Now a few words about the process.

At preset intervals, dia$par analyses the preset categories of Yandex.Market in the regions of interest to display its findings in the PCM’s working panel.
Still, it is not enough to display Top X ratings in the PCM’s panel, already full of numbers. The efficient Top X implementation practices will be covered in the episodes to follow.

Being somewhat previous, we can report that the use of this tool in Just Co for their main categories boosted their sales and profit by 30−40%. Out of the blue — simply due to a better assortment.
?After it was rejected (for reasons unrelated to the rational conduct of business) their purchases, assortment, and sales soon returned into their previous condition.

In the episodes to follow:

Being inside dia$par. Some stories
© 2020
U.S. Humanless Enterprises
contact us
U.S. customers:
Call me back
humanless service
We do NOT collect any visitor's
information, except given obviously,
and do NOT share it with third parties.