Looking askance 2
Looking askance 2
July 2003
The intangible meets the imponderable
I have just been reading the 2003 report on supply chain performance from the Grocery Manufacturers of America (GMA) [see note]. This is a survey of recent history and trends in the US food, grocery and consumer product industries. Prepared for GMA by the Roland Berger consultancy, it details the logistics experiences of 25 large companies or operating units. The respondents include giants like Hershey Foods, Procter & Gamble, H.J. Heinz, Kraft Foods and Pfizer.
Among the results is this classic Porter value-chain analysis of the companies' average logistics costs. These costs, says the report, rose 12 per cent from the previous year. They now represent 7.4 per cent of the organizations' net sales, instead of 6.6 per cent in 2001. The pressure on companies to cut these costs must be immense. Can BPM help?
Logistics cost chain
Looking at this diagram through my BPM monocle, I can see perhaps five areas of cost that might be amenable to business process automation. They are:
planning, forecasting and scheduling (including inventory management) - 2.5 per cent
information processing (including e-commerce) - 1.8 per cent
customer service - 2.7 per cent
procurement management (or "buying", as we used to call it) - 1.1 per cent
plant warehousing - 7.4 per cent.
I have not included distribution centre costs, since these are mainly to do with goods handling. For some reason, this seems not to be an area where people use BPM much, if at all.
Together, the five areas listed above account for 15.5 per cent of the responding companies' total supply chain costs. That equates to just over 1 per cent of their net sales (15.5 per cent of 7.4 per cent). Since none of these companies is running at a loss, this is 1 per cent or less of their cost of sales.
Many of the activities under these five headings will already have been streamlined and computerised. The scope for cost cutting will therefore be less than in a 'green-field' situation. We could possibly reduce their cost by an average of a tenth, if we're lucky. That is roughly 0.1 per cent of these companies' total costs.
This is not going to cut the mustard. Something that might cut less than 0.1 per cent of a company's costs is not going to head anyone's list of favourite cost-reduction programmes. These companies would be looking to claw back elsewhere in redressing the 12 per cent increase in their costs since 2001.
Clearly, BPM could not sell itself into these large companies on the back of its cost-saving potential. How else could it be made attractive?
The answer lies in the report. It identifies five main issues that these and similar companies must deal with:
improving the on-shelf availability of products
synchronizing data
increasing the flexibility of ordering and delivering
improving forecasting accuracy
improving internal cross-functional collaboration.
The notion of process runs through these like "Blackpool" through a stick of rock. So too does real-time monitoring and reporting. Industry-wide programmes like Collaborative Planning, Forecasting and Replenishment (CFPR) and Efficient Consumer Response (ECR) have these as important constituents. Services like the WorldWide Retail Exchange (WWRE) portal offer it too, as do in-house schemes like those in Safeway (UK) and Tesco.
This emphasis on enabling the new and the better, not just the cheaper, is confirmed by a different GMA report. Its Information Technology Investment Study 2003, found the top seven drivers of IT priorities for this sector to be:
improving trade customer service
data synchronization
optimizing business processes
greater data security
improved business agility
improved financial reporting, and
customer collaboration.
Cutting costs did not appear until number 8.
So that's clear. In these consumer product and food companies, BPM is a busted flush as a cost cutter. Its promise is as an aid to better response and to customer satisfaction. No surprises there, you might say.
Why is it, therefore, that nearly every supplier and consultancy trying to sell BPM keeps talking about "Return On Investment" (and meaning cost displacement)? It seems the only justification arithmetic allowed is one based on cost reduction or cost avoidance.
Either there is a huge amount of self-deception involved here or else some people are telling enormous fibs. How can an organization 'pay' for greater responsiveness, closer collaboration and other outward-looking objectives when its only permitted tender is shredded dollars? There must be other methods and monies in use to make the case, involving the dread 'intangibles', or am I missing something?
Answers on a postcard please, to BPMG Mansions, Letsby Avenue, Limply Stoke, Mummerset.
Close readings
I have also recently been reading my way through a collection of workflow case studies. This is nearly always rewarding, providing there is enough narrative meat on the factual bones. You can normally glean at least one insight or perspective from doing so.
Most of the organizations were using good old-fashioned document management or workflow, and none the worse for that. Few of the cases were about what I would regard as real BPM.
I judge this with two simple tests, the first technical and the other practical:
Is the organization using software that has a separable set of rules that users can manage and change easily and directly?
Is it applying the software to managing company-wide or intercompany processes?
Whether those rules are defined using Petri nets, pi-calculus or even the Kabbalah is not something I worry about. Users I have spoken to feel the same way. Telling them that BPM software must be based on a particular modelling method is straining at gnats if done sincerely and cynical sophistry if not. What next, some kind of "Milner Inside" label, to nudge users towards buying the 'right' kind of BPM-ware? This is a debate for standards committees alone.
Circumstances alter cases
One case that caught my eye was about an Action Technologies customer, International Truck and Engine Corporation, in the USA. The installation won a gold award in this year's "Excellence in Workflow" competition that WfMC, WARIA and Giga organized. A write-up of it also appeared in Line56 magazine, and has prompted these thoughts.
The article draws attention to the distinction between assigned and negotiated tasks. This is the difference between workers having to do exactly as they are told, F.W. Taylor-style, or having some discretion in how they work. Action Technologies' products are described as just right for the latter. (No surprises there either, you might say.)
This alignment with worker empowerment is ironic. The company's first product, in the 1980s, was The Coordinator. This was seen at the time as a cross between email and Procrustes' bed. Every message had to be one of seven types - note, inform, question, offer, request, promise and 'what if?' (which speculate upon an action).
You had to work to this model of communicative behaviour or else you couldn't use the software at all. This incensed some West Coast American users, who hated its infringement on their perceived liberty to send what they liked to anyone they wanted. (Some things never change.) They gave it the nickname of "Naziware, from Achtung Technologies" - not that they were mentioning the war in any way.
Like all the company's later products, The Coordinator embodied the Winograd and Flores four-part process model to track progress. The four parts are prepare, negotiate, accept and perform. It is the second of these that provides the discretion that International Truck and Engine value.
If nothing else, the change in the reputation of Action Technologies' products shows what happens when perceptions change. Workflow software typically enforces a 'one best way' view of life, something work-study engineers have lived by for over a century. Against that background, an engineering company like International Truck and Engine sees introducing user choice as liberating. Against the earlier background of unrestrained emailing, by contrast, the Flores-Winograd model was seen as repressive.
It is, I hope, a truism that the human situation affects, sometimes controls, how well a system will be received. ActionTech's prize-winning project shows this as clearly as one could wish.
Closer readings 
Also catching my eye this week was the fluent way with corpo-factoids displayed by the boss of a BPM software company. (No names; no pack drill.)
It is the essence of factoids that they are either unchecked or uncheckable. Everyday examples include the 'facts' that Eskimos have 40 words for snow, that the world's population can crowd on to the Isle of Wight and that The Great Wall of China is only man-made object visible from space. (Someone at the Pentagon has paid over the odds for camera lenses if that is true.) These are for the most part harmless miniature myths.
Corpo-factoids on the other hand can be damaging, because people are willing to base potentially important decisions on them. The 'fact'' that it costs 5 to 10 times more to keep a customer that to gain one is a prime example. Where's the research on this? Does it apply to all industries and all customers? Why are some banks so keen therefore to select out low net worth customers?
Another is the old chestnut that 95% of corporate information is held on paper. This used to be trotted out ad nauseam by salesmen looking to justify the sale of document management software. It may still be. I'm lucky enough not to have to sit through these presentations any more.
The author Norman Mailer described factoids as "creations which are not so much lies as a product to manipulate emotion in the Silent Majority". Or, with the corpo-factoid, the buying majority.
The factoid that this supplier of BPM-ware was peddling goes as follows: "Eighty years ago, 85 per cent of workers fell into the assigned task category. Now if you go to the Bureau of Labor Statistics in the U.S., more than 50 per cent of corporate workers are knowledge workers."
What is the source for this first statistic? How were the 'assigned' sheep in the 1920s sorted from the unassigned goats? Who decided what kind of job fell in which category?
Also, why choose that period? This was before the last World War and even before the Great Depression, so America's industrial and economic landscape was completely unlike today's. This is not even comparing apples with oranges.
The vagueness continues when you look at the Bureau of Labor Statistics' Web site. There is no mention of the term "knowledge worker" in its glossary or, apart from a book review, anywhere on the site. How has this company arrived at its figure of 50 per cent, therefore? (And why have people stopped saying "half"?) Will we ever know? Do they know?
This is the essence of using factoids, of course. It means never having to say, "I looked it up".
And what is a "knowledge worker", anyway. Have you ever met anybody who doesn't use knowledge in his or her job? Where are the ignorance workers? (Other than the newsrooms of the tabloids, of course.)
There are examples everywhere of how similar unexamined assertions are used in selling products. This is reasonable behaviour for suppliers - all's fair in love and marketing - but risky for the customers to accept unchallenged.
Ernest Hemingway famously said that an essential part of any writer's tool kit is "a built-in, shockproof crap detector". It should be in any user's, too.
Roger Whitehead