JoAnn Hackos, Comtech Services, Inc.

At the September 2009 Best Practices conference, Maria Brownstein of Sybase and I spoke about the concerns some managers have raised about productivity. Based on the summer 2009 Productivity Survey conducted by CIDM, we found that while nearly half of all respondents reported increases in productivity with the implementation of DITA and XML-based authoring, nearly the same percentage reported productivity declines.

How do we account for the productivity issues that seem to have emerged with the implementation of new tools and processes in our organizations?

With close analysis of the survey data, we learned that productivity decreases seem to be focused on organizations with two or fewer years of experience in implementing DITA or other XML-based authoring and content-management systems. As groups gain work experience, they experience improvements in productivity. Given the startup costs and the amount of training required to develop reusable topics rather than static books, the initial productivity declines are understandable. However, we also found some groups still experiencing productivity decreases after three years of implementation.

Based on the survey analysis and discussions with Maria Brownstein and other senior managers, we think we know why.

In response to our inquiries about productivity changes, we received an insightful comment from Bob Ryan at Motorola.

I don’t think we have any hard data regarding the impact of XML/CMS to our productivity—positive or negative. That said, my guess would be that if you polled the writers, the net result would be a negative hit to productivity.

I believe that the primary reason for this is that we viewed XML/CMS as a tool change rather than a paradigm change. Rather than examining our processes to determine how we could maximize the potential of these new tools, we examined the tools to see how we could make them fit our old processes.
—Bob Ryan, Motorola

Bob points out a critical factor—how we implement change, particularly change of the magnitude we find topic-based authoring and content reuse to require.

Unfortunately, we don’t know the job titles of those responding to the survey (should have asked that question—in retrospect). We suspect that individual contributors who find their working processes changed significantly are more likely to believe that their personal productivity has declined. Considering that they are being asked to write in structured topics, use an XML editor and a content management system, plan their content in terms of optimal reuse, and collaborate with other authors in the development of a body of content, their impressions of work to be done (and the work itself) most certainly have changed.

Clearly, the writer’s job changes dramatically in a topic-oriented and content-managed environment. My conclusion, however, is that these changes are for the best. Better planning of content, more collaboration among authors, more disciplined writing, and addition of semantics at the topic and element levels of topics—all of these activities take time. But the time needed also results in added value to the outcome. The individual author may have more to do to plan and develop topics, but, over time, the author should have fewer topics to maintain and create because he or she is now part of a content-development collaborative team.

Let’s look at productivity in broader terms than an individual’s work. We measure productivity not by measuring an individual’s output but measuring the output of the organization as a whole. For individuals, we measure performance against goals. In that individual performance affects the group’s productivity, they are, of course, interrelated.

The writers may take more time to complete individual tasks, like writing new topics, because they must plan and collaborate with team members. They take more time performing their most difficult tasks. At the same time, the tasks of maintaining topics and creating final deliverables by reusing topics is made much easier and faster. Overall, the productivity of the team will have increased, especially after the time and cost of learning is completed.

If the writers and others on the team, including those supporting the tools, are well trained and embrace the new ways of working, productivity is most likely to increase after the initial investment. In fact, the greater the investment in good training, well-constructed information architecture, and effective project and change management, the more likely that the investment will pay off and pay off more quickly.

Figure 1 illustrates the importance of the initial investment and the paradigm change that results in long-term savings.

The Success curve illustrates an organization that invests early and strongly in a paradigm change. Once the curve falls below the original labor costs line, the savings are continuous and may in fact increase. The Failure curve illustrates an organization that has not invested in a paradigm change, perhaps simply converting whole books into XML ditabase files without making any changes to process, architecture, or project management. This organization never recovers its costs.

How then do you ensure that you see productivity increases with a move to DITA, topic-based authoring, and a CMS?

  • Develop a strong change-management strategy
  • Institute productivity measurements before you institute your change process
  • Ensure that you make a strong investment in information architecture and your reuse strategy (reuse will be the key to your productivity increases)
  • Understand that training must include more than tools training; this is not a tools project
  • Train for project planning, topic-based authoring, minimalism, reuse strategies, and collaboration (at least)
  • Measure productivity carefully all along the way as you institute the change process

We’ve seen this work amazingly well in organizations that we’ve guided through the process. However, we find that some organizations and some CMS vendors consider the investment in the process changes we recommend to be too costly. I’d ask them to consider the cost of seeing productivity decline after a substantial tools investment. For productivity really to increase, a paradigm change is essential.