Automobile assembly line
A Volkswagen assembly line in 1960 at Wolfsburg (image via Wikipedia)

Workflow is a new buzzword (Google it), and its prominence has come about because it speaks to real needs and may deliver real benefits. Consider the situation of a typical publishing organization. This organization has had a workflow in place for many years; it was so thoroughly in place, so entrenched, that no one even thought of it as a workflow. It was simply how we do things here. It was only when things started changing around it, when new technology began to open up new possibilities, that workflow was viewed as such and analyzed carefully. It was soon determined that the introduction of a new technology would streamline that workflow; it might improve quality; it could deliver savings to the bottom line.

At first when new technology disrupted old workflow, the resulting situation was unwelcome; it was chaotic and costly. Most of the publishers who are reading this post have had the exasperating experience of having to live with both print and digital workflows side by side. It was a sign of better times to come when those workflows were eventually combined and formats, whether print or digital, were the endpoint of a common process. Whew! We have solved that problem once and for all.

Of course, as Kent Anderson’s excellent recent post describes, the number of new formats never ends. Yesterday we had to learn how to display the content of print publications on a desktop, today we wrestle with mobile formats. Yesterday the PDF seemed like the sign of an astute publisher (we have captured all the virtues of the print publication, but have reduced our costs and enabled new forms of discovery), today the PDF is the unwelcome cousin who shows up during the holidays whom you have to take in. Today’s hero is responsive design, but like all heroes it will come to disappoint us some day. No reason not to enjoy the jig of a hero while it lasts, but all things must pass. It’s not a bad idea to start thinking now about what you are going to do—and why—after you announce your mobile solution.

Most investments publishers make in technology go toward allowing the publisher to hold its ground, simply to keep up with the competition. The marketplace demands these investments, and I have yet to hear a good case for failing to make them. (There is a strong case that when a company reaches this point, it is a good time for the shareholders to sell out, putting the burden of the new technology investment on the shoulders of the new owner.) Other investments show a return: they reduce costs and thus contribute to the bottom line. In the book business today, the margins of many publishers have doubled, as publishers reap the benefits of digital formats without sharing them with authors (as one would expect, as there is little market pressure to do otherwise).

Once in a great while investments in technology enable a business to grow. Some easy examples: online bookselling opened up international markets; the Big Deal permitted the largest STM publishers to take a greater share of library expenditures; and print on demand created a revenue stream from the long-dormant backlist. It is noteworthy that augmented revenue brought about by technology usually only benefits a small number of companies, and often only one. To use the examples in this paragraph: Amazon for online bookselling; Elsevier, Wiley, and Springer for the Big Deal; and Ingram/Lightning Source for POD for academic titles. If you want to use digital technology to grow, you have to figure out more than how to get it to work; you need to have a strategy that uniquely enables your organization to find new customers or to sell more things to the customers you already have.

While publishers may discuss their spanking new digital workflows, it’s not out of place to ask how these investments have added to sales. Yes, costs are down and, thus, profits are up, but the top line? A company whose idea of vision and strategy is to examine its cost structure—and then to examine it again—is not going to be the one to establish a new paradigm, to open up a new channel, to bring more customers into the always embattled world of publishing—embattled because the low capital requirements of this industry invite a great many participants. The question every organization has to ask is how to move from the work stream to the revenue stream.

Monetization for new formats has been elusive. It was not so long ago that publishers tried to charge one price for print, another for Web versions, and perhaps yet another fee for some mobile formats. This made perfect sense: each of these versions required incremental investment, and it was only natural for the publishers to seek a return on that investment. The market has spoken, however, and it is increasingly rare for a publisher to reap any gain from the money that goes into multiple formats. It’s just the cost of business, and those costs keep going up.

I have long thought that publishers may find an opportunity in thinking more deeply about the nature of mobile computing, but now I am not so sure. Currently mobile computing is simply a cost to a publisher, which has to provide this capability without any augmented revenue. Mostly we think of mobile devices as small-screen versions of desktops. We read on a desktop, and we read on tablets and phablets and phones. But often when we are mobile we do not read at all. The small screen is hard on the eyes, and even harder on the fingers as you try to type on a tiny keyboard. It is no wonder that the mobile suppliers are investing heavily in voice-recognition and text-to-speech technology, as the mobile experience is often more aural than visual. What was it you were doing when you walked the dog today? Did you read a book or listen to music? Or perhaps you listened to an audiobook. Audio is in fact the fastest growing category of the consumer book business.

Unlike video, which, like text, presumes a user who is not moving around, audio potentially opens up a new use case for published materials. We listen to things when our sight is impaired in some way. Not necessarily in a medical sense but in a way that makes it inconvenient to read. So, for example, we listen to audio when we drive a car and have to keep our eyes on the road. The virtue of audio is that it potentially expands the workday by slipping in an audio stream when we are otherwise occupied. It is no wonder that so many STM publishers are experimenting with podcasts.

The problem for scholarly communications and audio, though, is that the densely packed information of academic books and articles is very difficult to absorb by ear. If it were otherwise, we would simply take a batch or articles, run them through speech synthesis technology and listen to them on an iPhone. What this means is that to tap into a revenue opportunity for audio, we have to do more than come up with a new format. We have to think about different kinds of content that researchers and clinicians find useful. This takes us away from the task of workflow management into new product creation. It’s more exciting, perhaps, but it is also painful, costly, and inconclusive.

And that’s the problem with the workflow buzzword. The workflows we reengineer now handle things more efficiently, but they still flow to the same place. Publishing is not production. One test of an innovation is whether people will pay for it.  Have we passed the test?

Joseph Esposito

Joseph Esposito

Joe Esposito is a management consultant for the publishing and digital services industries. Joe focuses on organizational strategy and new business development. He is active in both the for-profit and not-for-profit areas.


22 Thoughts on "Workflow without Strategy is a Dead End"

Getting better authors can increase sales and workflow can play a major role here. Conversely, new products can be big losers. So new products is just one possible growth strategy, and a risky one at that.

Just two unrelated observations: 1) OA publishers are faced with investing in multiple formats too even though they are giving away their product (and, unfortunately, many library-based IRs do not make this investment); and 2) there should be a real market for audio for academic articles and books among those athletes who do distance swimming or running; they could be multi-tasking.

An efficient taxidermist is still a taxidermist.

The best line in this post is “Publishing is not production.” We tend to be so focused on the mechanics of making things that we forget the publishing is about risk attenuation, the priority system achieved through editorial and peer review, and branding that reflects both. The production systems are necessary, but not sufficient.

Actually things like editorial and peer review are part of production, something I call cognitive production. Cognitive production is the dominant form of production in America today, in the straightforward sense that more people produce thought than produce things. And cognitive production systems have workflow, although it is very different and far more complex than physical production. In fact I call the management of cognitive production systems “chaos management” because of the nonlinear nature of the workflow.
See my little essay on this:

Boy. This could lead to some great new job titles. Chief Chaos Officer anyone?

Cool! I had not thought of that title. The basic idea is that cognitive production systems are subject to what I call “issue storms” which if not properly managed can quickly consume a lot of resources. We increasingly live in issue driven work environments, which are very different from physical production settings. Traditional workflow analysis typically does not recognize the role of issue storms, so they are managed on an ad hoc basis, if at all.

For example, I once looked at a big city building permit system that had bogged down. They were renovating downtown but it was taking a year to get a major permit. It turned out that the average application was going back nine times for revision, so instead of processing 2000 Apps a year they were doing 20,000. The sources of all this revision turned out to be simple and fixable once identified.

So the key to chaos management is early identification and resolution of issue storms. And yes this needs to be someone’s job. I suspect that publishing workflow is also subject to resource draining issue storms, leading to delay and frustration.

The publishing world that I work has spent a great amount of time in article submissions, editorial systems, and the review process. But the black hole continues to be fulfillment, renewal processing, invoicing, and customer service. Often the solution is to outsource many of these functions, but given the luxury of double digit margins, losing a few % on the black hole is reasonable and some publishers are worse than others. Applying some modern workflow techniques to the black hole of publishers could easily pick up additional margin. Most publishers are still using the fulfillment system that they installed in the print heyday, but now operate with more workarounds than standard procedures.

“Most investments publishers make in technology go toward allowing the publisher to hold its ground, simply to keep up with the competition. The marketplace demands these investments, and I have yet to hear a good case for failing to make them. (There is a strong case that when a company reaches this point, it is a good time for the shareholders to sell out, putting the burden of the new technology investment on the shoulders of the new owner.)”

I’m really interested in the parenthetical second line here–it seems like an afterthought, but does anyone have any additional resources on the subject? I’d like to learn more about this argument.

Preparing scholarly communications to be absorbed via speech is not as simple as running it through a speech synthesizer. Images need to be described. Tables need to be read in an order that makes sense, with column heads repeated. Navigation needs to be created so we can re-read tricky passages and get back and forth to the sidebars. Those of us who structure our books to be accessible to people who are blind or have other print disabilities could offer some advice. But absorbing this kind of information auditorily would still be difficult for those of us who are “sight dependent.”

I find this whole premise that scholarly communication needs new paradigms and products to be wildly overblown. A journal article is basically a letter and a letter is a fundamental logical form. Assuming that the digital realm or the Internet is somehow going to change this basic form is to confuse cognition with physics.

David, I disagree. Of course scholarly communication needs new products to improve. The traditional format of the science journal article has some really serious problems. For example, try to reproduce a recently published scientific article in the lab, even if you have all the equipment and money in the world. According to many studies, today your chance is less than 30%. This is only one big problem that needs to be addressed, and there are others.

Moshe, I consider this to be a completely bogus concern. If you want to reproduce someone’s work start by calling them on the phone. There is no reason a journal article should convey everything needed to know for replication. People are promoting a nonsensical standard. (By the way, if the work was federally funded try the research report. They tend to be much more detailed.) Journal articles are merely long abstracts and that is all they need to be. They are the beginning of communication, not the end of it.

I think you’re presenting a self-contradictory argument. First you state that the “whole premise that scholarly communication needs new paradigms and products [is] wildly overblown”, and when presented with a clear example of a need for a new type of communication, you call the concern bogus, but then admit that it is a need that must be served through an extremely inefficient and archaic means of communication.

It’s fair to argue that these sorts of things aren’t necessarily the purpose served by the research paper. But it’s not fair to argue that these needs don’t exist, nor that we can’t do better to fulfill them than a phone call, particularly in an age where the “phone” part of your phone is now just another app, and likely one of the least used apps on the device. Perhaps at one point science was small enough that everyone could call one another, and techniques were simple enough to be easily described over the phone. That is no longer the case.

Do you really think it’s reasonable for a researcher in Beijing to call a researcher in Latvia to ask for methodological details? Is that a trivial activity in itself or an undertaking that may require multiple interpreters. Worse is the position of the lab with the groundbreaking technique. They may end up spending hours of every single day on the phone explaining the same thing, one by one, to thousands of labs around the world. Wouldn’t it be easier to write up the details once and make that available to the research community?

The research paper is indeed a highly evolved form, honed through years to serve particular purposes. But like in the natural world, evolution doesn’t stop even when one reaches an apex state. The environment continues to change and so does the organism. If those needs exist (and I’ve certainly argued that they do ( then the paper might change to meet them, or if not, other forms of communication will be developed.

There is no contradiction in my position. My point is that wanting journal articles to solve the replication problem is a bogus concern. Replication is complex business, as you yourself have argued at length. It is best done through personal contact, probably over a lengthy period of time, hence my “phone call” metaphor. As I recall you give the example of people physically traveling from coast to coast.

There may or may not be a need for new forms of communication here but they will not be journal products, because the magnitude of the required communication is far too great. The whole concept of journals solving the replication problem (if there even is one) is a mistake. The journal article is not all there is to scientific communication, but a lot of these new product schemes act as though it were. Scientific communication is a complex, highly distributed process, within which journal publishing plays a very important, but a very narrow, role.

I find this whole premise that scholarly communication needs new paradigms and products to be wildly overblown.

There may or may not be a need for new forms of communication here

There is no contradiction in my position.

Again, there’s a difference between arguing that communication needs aren’t changing and new ways of sharing information don’t need to be adopted, and arguing that journals are the wrong place for them to happen.

The whole concept of journals solving the replication problem (if there even is one) is a mistake.

So you think that making detailed protocols publicly available (Nature Methods, Nature Protocols, CSH Protocols, JOVE) and making datasets publicly available (GigaScience, Scientific Data) doesn’t help reproducibility? Yes, there are some fields and methodologies that require detailed instruction and expertise, but there are many that can be communicated in a straightforward manner, and the journal provides a reasonable method for doing so.

I called for methods journals more that three years ago:

But these are about summarizing important new methods, not about supporting replication. (Nor are new journals new products in the sense thatJoe is calling for, not as I read him.) Not unless you want to mandate that every discovery article be accompanied by a methods article, and even then the methods article is not likely to provide the detail needed for replication. Same for data posting by the way.

The issue tree plays in here, see What it shows is that the number of sentences increases exponentially with level of detail. And the level of detail required for replication is likely to be several orders of magnitude greater than that provided by journal articles. Journals cannot solve the replication problem.

David, there are bogus and imaginary concerns in the current discussion on the scholarly communication, but the reproducibility of published articles is not one of them. It is a very real problem that seriously affects primary users of these articles (practicing scientists).

>> “If you want to reproduce someone’s work start by calling them on the phone.”
Just ask any practicing scientist you know how effective this approach is.

Too many Davids around here. Of course the same could be said about Andersons, Michaels, and Phil/Phills.

I will be happy to comment on this blog just to diversify you with “Moshe” 🙂

Moshe, I am not claiming that replication is a bogus concern, just that thinking journals can solve the problem is. See my reply to David C. above. The journal article, as a logical form, simply cannot provide all the information needed for replication of complex scientific work. It is a simple matter of scale, such as six pages versus sixty pages.

I also have my doubts about the seriousness of the replication problem, which has the trappings of a fad, but that is a different matter.

I was chairing the STM eProduction seminar on 4th December and saw Kent the day before at the Innovations seminar. STM eProduction dates from 1999 and was conceived in 1997 or earlier – hence the odd title. It now attracts about 80 registrants. No-one at the seminar would consider that publishing is equated with production and I am very puzzled by Kent’s aphorism or sound-bite. It is a pity Kent could not have come to the seminar. When I set it up the idea was that it should be accessible to senior management because production spend is the big spend in publishing and yes the work flow is most important. BUT alas only a very few top people find time to come but they do pronounce about work flows at the drop of a hat.


Comments are closed.