Printer Friendly

The SKU explosion.

SKU (pronounced "skew") is a handy acronym that stands for "stock-keeping unit." SKUs are the lowest level of granularity in the distribution channel: a SKU is a unique version of a product, one that is by definition not exchangeable with any other version. If a developer packages 3.5- and 5.25- DOS versions separately, each becomes a separate SKU; if a competitive upgrade carries a different price than the standard version, each is also a SKU.

What's important about SKUs these days is that there are so many of them. Upgrades, versions for new platforms, promotional packages, network and server editions, and multiple disk formats have transformed single-version titles into dozens of new SKUs. Utilities, templates, fonts, and add-ons further complicate the equation, while the growth of international markets--each with a set of language-specific versions-- will mean an even faster proliferation of SKUs in the near future.

No one has a true count of the number of SKUs in the distribution channel today--not to mention a count of all those products that sell through non-retail channels. But, based on data from PC Research, a new research service (11504 Links Dr., Reston, Va. 22090; 703/435-1025), there are at least 3,500 active SKUs in the channel today.

Of course, virtually all of these SKUs exist to produce revenue, increase market share, or save money (for example, by eliminating the need to provide dual media in one package). But it's becoming clear that the SKU explosion actually undermines the software industry's basic profitability. At every level of distribution, from developer to end user, the proliferation of SKUs imposes huge but hidden inventory management costs. Sales forecasts for low-volume SKUs tend to be little more than guesswork, so the distribution pipeline ends up with frequent stockouts, extra handling costs for special orders, and constant short-run, crisis-driven production runs.

These issues are especially critical for the reseller channel, where many of the key players simply don't have the money or systems savvy necessary to handle thousands of SKUs. It's instructive to watch how badly resellers fumbled even a simple issue like bar code standards.) In the long run, we expect that the survivors in the distribution channel will be resellers with hyper-efficient back room operations, not just razzle-dazzle marketers.

How bad is the SKU explosion? once again, PC Research's data offers some interesting insights. PC Research's Ann Stephens publishes aggregate monthly sales and inventory data from a growing universe of major distributors and resellers, including four of the largest wholesale distributors (not including Ingram). Using Stephens' November distributor data, we took a look at how the proliferation of SKUs is beginning to affect developers and resellers:
Sales and Inventory % of Total % of Total
Ratios by SKU Unit Sales Inventory
Top 10% of SKUs* 81.1% 51.9%
2nd 10% of SKUs 9.2% 11.8%
3rd 10% of SKUs 4.2% 8.2%
4th 10% of SKUs 2.3% 5.9%
5th 10% of SKUs 1.4% 5.7%
6th 10% of SKUs 0.8% 4.9%
7th 10% of SKUs 0.5% 3.8%
8th 10% of SKUs 0.3% 2.8%
9th 10% of SKUs 0.2% 2.9%
10th 10% of SKUs 0.1% 2.2%


* Ranked by revenue per SKU. Source: PC Research, Distributors Report 11/91. Sales and inventory figures represent a consolidation of data from four leading distributors.

* Revenue contribution: PC Research's data underscores the fact that a relatively small number of SKUs currently generate a lopsided percentage of total sales volume. Thus, 10% of SKUs produced 81.1% of sales volume for PC Research's four distributors. Or, to look at the ratio another way, 90% of SKUs generated less than 20% of distributor revenues. At the bottom of the chart, the numbers are even more troubling: 40% of SKUs produced just 1.1% of sales volume.

Considering these numbers, it's not surprising that distributors typically focus almost all of their promotional efforts on a small number of high-volume commodity titles, because nothing else really leverages revenues. (Distributors in other industries usually apply higher markups to low-volume SKUs, so these products make a respectable profit contribution; software resellers, alas, haven't fully mastered this trick.) Moreover, the numbers suggest why distributors and other resellers show so little interest in new, innovative titles: The potential dollars just aren't worth the hassle.

On a smaller scale, we see similar ratios for the SKUs that individual publishers and developers sell through the channel. For example, PC Research's database recorded 128 SKUs for WordPerfect in November, worth $9.2 million in distributor sales. Of these SKUs, just six produced 62% of total sales volume. Likewise, PC Research logged $65,081 in distributor revenues for Funk Software's 11 SKUs; of these, three produced 84% of Funk's revenues.

* High inventory overhead: The corollary to these lopsided revenue ratios is the disproportionate investment in inventory that minor SKUs require. PC Research's distributors required almost as much inventory to support the bottom 90% of their SKUs as they did for the top 10%. The lowest 40% were particularly costly: Here, 11.7% of inventory backed up 1.1% of sales. Although it's questionable whether distributors actually "invest" in inventory (since most pay slowly and insist on open-ended return privileges), somebody ultimately pays for the inventory carrying costs of all these low-volume SKUs.

* Low inventory turnover: Using PC Research's data, we also calculated monthly inventory turns (sales divided by inventory) for each group of SKUs. Overall, the distributors in our database turned over their inventory a respectable 10.5 times per month (median) for the top 10% of their titles, and they achieved between 31.9 and 5.6 turns per month for half of the SKUs in the top 10% (the 50% range on the chart below). But inventory turnover rates for low-volume SKUs were much worse; in fact, the bottom 40% of SKUs averaged only one turn per month.

A related measurement of inventory efficiency is fill rate--the percentage of orders that can be filled from inventory. PC Research's data doesn't directly measure fill rates, but the data does show that distributors at the end of November were completely out of stock for 411 SKUs, about 12% of all active SKUs. Predictably, stockouts occurred more frequently among low-volume SKUs.
Inventory Turns -Turns Per Month- No Inventory
 By SKU Median 50% Range
Top 10% of SKUs* 10.5 31.9-5.6 30 SKUs
2nd 10% of SKUs 7.5 28.5-3.3 33 SKUs
3rd 10% of SKUs 4.9 19.0-2.2 36 SKUs
4th 10% of SKUs 3.3 17.0-1.4 41 SKUs
5th 10% of SKUs 3.0 10.9-1.1 38 SKUs
6th 10% of SKUs 2.3 11.4-1.0 56 SKUs
7th 10% of SKUs 1.6 6.5-0.7 50 SKUs
8th 10% of SKUs 1.0 3.0-0.5 40 SKUs
9th 10% of SKUs 1.0 3.0-0.4 40 SKUs
10th 10% of SKUs 0.4 2.0-0.2 47 SKUs


* Ranked by revenue per SKU. Source: PC Research, Distributors Report 11/91. Sales and inventory figures represent a consolidation of data from four leading distributors.

In fairness, the data from PC Research's four distributors may not be completely representative of the entire reseller channel. We happen to know, for example, that Ingram tends to manage inventory better than its smaller competitors. But it's also clear that the explosion of SKUs has led to fundamental inefficiencies that good management alone can't solve. Moreover, these inefficiencies are likely to become an even greater burden in the future: our best guess is that the channel will have to absorb at least another 500 active SKUs during 1992 alone.

Nevertheless, we do expect that the chaotic growth of SKUs will begin to slow down eventually. Increasingly, software companies are asking their marketing and product managers to prune the number of new SKUs, and--to reduce inventory Costs--some publishers are beginning to cut back on elaborate four-color packaging for minor SKUs. In addition, manufacturing directors are fine tuning their own forecasting and inventory management systems, and many are exploring on-demand and just-in-time manufacturing technologies.

Finally, we expect that bigger publishers will revisit the question of delivering software electronically or through CD-ROM systems--not as an alternative to the traditional mainstream package, but as a more efficient vehicle for delivering upgrades, network packs, add-ons, and other marginal SKUs that the distribution channel probably will never handle effectively. None of these solutions will solve the SKU problem completely, but they'll all help.
COPYRIGHT 1992 Soft-letter
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1992, Gale Group. All rights reserved. Gale Group is a Thomson Corporation Company.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:distribution of software; stock-keeping units
Publication:Soft-Letter
Date:Feb 22, 1992
Words:1420
Previous Article:Awards: a European showcase.
Next Article:Electronic distribution: a French model.
Topics:


Related Articles
Benchmark survey: manufacturing operations.
How Microsoft tamed the SKU explosion.
Data points: the low-volume SKU problem.
ON TRIMMING THE SKU COUNT.
MERCANTEC DEBUTS MERCHANT CONNECTION FOR SOFTCART 5.0.
COMPANIES SAVE MONEY USING INTERMITTENT FORECASTING TECHNOLOGY.
MarketMile LLC. (Procurement).
Secrets Of Pankraton.
Secrets Of Pankraton.

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters