Metrics Mania—Writing

Dawn Stevens, Comtech Services, Inc.

Whether or not you have specialists, such as graphic artists, editors, and formatters, supporting your information development, you’ll find that the bulk of your project budget goes toward the initial creation of the content by your writers. In fact, often the amount of time allocated for the specialists is calculated based on the writing time – editing might be allocated 25% of the writing time, for example.  Thus, your writing metric is one of the most important to capture. Yet, it is also the one with the greatest variation and the one where you’ll likely get the greatest pushback:

Writing is a creative art, not a science. You can’t rush a masterpiece. How can you measure creativity?

Although there’s likely a little dream to write the next great novel in every technical writer’s head, the reality is that your technical information probably does not make a riveting plot line. Creative writing expresses the writer’s thoughts and feelings in an imaginative and unique way; however, technical writing expresses facts and instruction in a consistent and repeatable manner. That’s not to say technical writing is not a creative process, but it is a predictable creative process that can certainly be measured and used to anticipate future performance.

What to Include

Beyond the time spent putting pen to paper (or fingers to keyboards), your writing metrics should include time for:

  • Research.  It seems obvious that the process of writing should include the time spent learning about the subject matter to be written about. In fact, it’s a safe assumption that for new information the bulk of writing time is this learning curve, rather than the actual act of composing sentences and paragraphs. So it’s surprising to hear of people who haven’t accounted for this ramp-up time; two days into a 10-day writing assignment, they are looking for 20% completion of pages, instead of recognizing that there will be little to show while the initial research is being completed. This research might include meeting with SMEs, installing and using the product, searching the internet for information, or attending training classes.
  • SMEs. If you rely on experts to develop your information products, you need to include their time in the writing metric, even if they aren’t answering or directly billing to you. Remember your metrics are not only an indication of how much you spend, but also how long it will take to complete the project. You must account for the time it will take SMEs to provide written information if required or to meet with the writer to explain the information that needs to be written and answer questions as development continues. If SMEs are reviewing information prior to an official review cycle include that time as well; however, time spent reviewing an official release is likely better allocated to your editing metric.
  • Editing and draft cycles. Writing time needs to include not only the time required to complete a first draft, but all the time required to respond to editing and review comments as the document proceeds through the workflow. Be sure to not only include these hours, but to save them for the appropriate place in the project. It is a common mistake to allocate all the writing time in the budget to the writers when making the assignment. When they then use all the allocated time to create the first draft, which they will, there is no time in the budget or schedule to take care of the changes.

As previous columns have indicated, your writing metrics will likely include many other activities, depending on the responsibilities of the individual writer; for example …

Although you might try asking writers to track such activities separately, the reliability of these numbers decreases. You will likely get fairly accurate numbers that reflect the amount of time that person worked on a specific piece of information, but ask that person to pay attention to how much time is spent taking screen captures or formatting the document, and you’re more likely to get guesses that differ between people or even between days for the same person. It’s better to simply take into account that these activities are considered part of your writing activity.

Dependency Factors

Writing metrics can vary dramatically depending on the following factors:

  • Complexity of content. The more difficult the subject matter is to understand, the longer it will likely take to translate that information into appropriate language for the user. Compounding the issue is that frequently technical writers are primarily trained in writing, not in technical subject areas. If the writer doesn’t understand the information, you can be sure that either the resulting product will reflect that, or that it will take longer to write as the writing time incorporates the writer’s learning curve. Although a certain level of research is included in the writing baseline, highly complex projects will require more training and research to bring the writer up to speed.
  • Availability of resources. Although occasionally, the writer is the subject matter expert, often the writer is instead dependent on the subject matter expert for information. You’ll need to factor in the amount of time spent managing that relationship, waiting for answers and sending reminders. Similarly, content based on products that are under development still require access to requirements documents, prototypes, and betas in order to write something resembling reality. Your budget and schedule must be based on the availability of these resources. Finally, consider whether the information will be written from scratch or based on existing content or content written by someone else (a non-writer, for example). Generally, it is faster and easier to work from existing information than to start from scratch.
  • Stability of content. If the subject matter is highly subject to change during the writing stage (for example, a product under development), you need to not only plan that you will receive more review comments and requests for change than normal, but also that those requests will likely come late in the development process. Although you want to be involved as early in the overall process as you can, to minimize the impact of this factor, pay special attention to what you can reasonably accomplish during the early, less stable development phases. It is counterproductive to write full drafts that you know will require complete rewrites by the end of the project.
  • Other responsibilities. As mentioned earlier, if your writers are wearing multiple hats, those responsibilities are most often buried in your writing metric. If responsibilities, such as creating graphics or taking screen captures, are not a normal responsibility and therefore not factored into your baseline metric, you’ll need to account for them in your planning if the project requires them.
  • Writing aptitude. As with all disciplines, you must take into account the experience of the writer, not only in writing, but with the style guide, the technology, the process, the extended team, and so on. Be open also to the assertion that experience is not the only factor that affecting writing ability. Many experts agree that while you can teach good mechanics so that a message is adequately conveyed, there is an inherent writing aptitude that can’t be taught. If you are fortunate to have such naturally talented writers, you should account for that in the projects they are assigned to, rather than factoring their speediness into your overall average.

Interpreting Your Writing Metrics

When writing metrics vary from your estimates, and they will, first check the validity of your assumptions, particularly scope. One of the most common reasons for writing to take more, or less, time is that the amount of information being produced is more or less than expected. Some writers love to provide all the information they know about a particular topic, regardless of whether that information is needed by the intended audience. If you see writing trending high, check that the writer is keeping your scope definition in mind. Conversely, writing hours trending low could indicate a problem with information availability. Scope is smaller because the writer is unable to get the information that might be required.

Keep in mind, however, that tracking writers against a general average writing metric is bound to result in individual variations to the budget and scope. Although you can learn a lot about your organization and overall productivity by looking at general trends in most areas, tracking writing metrics at the individual level is much more effective in predicting schedules. Writing time tends to vary so much between individuals that if you use a department average for each person, you’ll likely find that handoffs to editing rarely fall when you expect. As a result, instead of a steady flow of information to editing, you face dead periods and mad rushes. You need to set guidelines – is editing handled on a first-come, first-serve basis? If something is turned in late, do the editors drop everything to help catch it up in the overall schedule? You’ll have a much more accurate schedule if you take into account the individual writing trends of your team.

Keep in mind also that a measurement against an average might be disheartening to a slower writer who turns over consistently clean copy, and artificially ego-boosting to a fast writer who ultimately requires a lot of editing. In fact, you cannot effectively interpret and evaluate your writing metrics, without considering the corresponding editing metrics at the same time.

Writing

Editing

Interpretation

High

High

If both writing and editing metrics are higher than you expect, you clearly have a problem. The writer is taking a long time to create content, but that content is not acceptable to the editor. In most cases, the issue lies with the writer, not the editor. Ask the editor for trends in the writer’s work, and provide coaching for the writer to improve in those areas. Explore whether the writer is hanging onto his or her work too long, but without noticeable gain in quality. If you suspect this to be the case, encourage the writer to turn work over to the editors earlier.

High

Low

When writing takes longer than expected, but editing is less, you may or may not have an issue to worry about. The numbers imply that the writer is a good self editor, taking the extra time to ensure that the content meets your company standards. However, you’ll need to evaluate if the combined writing and editing time is within your expectations, and if so, look at factors such as who costs more and who is more in demand. If you find bottlenecking at your editor, having a writer who is capable of doing a good job of self edit could help manage workflow. However, if there is a lot more to write, you might need to encourage the writer to forego some of the self checks and hand things over in a more timely fashion.

Low

High

When writing is quicker than expected, but editing takes longer, you might not feel the pain in your overall schedule, but you still should address the issue. The implication of these metrics is that the writer leaves a lot of cleanup for the editor. Perhaps the writer is unaware that he needs to raise the quality of his work, or perhaps she is simply lazy or relies on the editors as a crutch. Again, work with your editors to show the writer issues that he can improve, and share her metrics so she knows that she can take more a little more time to improve her content before editing. Keep in mind also that if you allow the trend to continue, you might find resentment and frustration building from the editors, who tire of always commenting on the same thing.

Low

Low

If both writing and editing take less time than expected, you likely have nothing to worry about. However, it’s a good idea to check that the output is meeting the expected scope of the project. For example, the low numbers might indicate that the content that’s written is well done, but further exploration could reveal that it’s incomplete.

 

If you find a general trend that your writers are always very close to their allocated budgets, you might consider reducing the time allocated to see what happens. Most writers tend toward perfectionism, so even though they have an acceptable draft well within their budget, they will use the rest of the budget making minor tweaks that are irrelevant or unnoticeable to the majority of your readers. If they have less padding, they’ll use less time. You might be able to improve your overall department metrics by taking away some of that “spare time.”

Tracking Pitfalls

Measuring writing productivity using the traditional metric of hours per page might send the wrong message to your writers. It indicates that the best writers create the most pages in the shortest period of time. Who is the most productive writer – the one who writes 8 pages in a day or the one who writes 20 pages in a day? The reality is that you can’t tell from that measurement alone. It truly can take longer to write concisely, to consciously provide only the information that the reader needs. Although you’ll likely still keep hours per page (or topic or screen) metrics, try to relate that information to qualitative measurements as well. These measurements might include both internal and external concerns. For example, as already discussed, the writing metrics should be coupled with the corresponding editing time. Unfortunately, the most meaningful qualitative metrics are difficult to capture and even if available take a while to manifest – whether or not the final product meets the needs of the intended audience. Consider tracking things like number of calls to customer support that could be addressed through better information products, and certainly keep track of any customer input related to the information.

In addition, the hours per page metric is a common metric to use when estimating, but it is only as good as your estimate of the number of pages that need to be written. Using this type of estimate requires that you not only have good writing metrics, but also good metrics for estimating the number of pages. For example, you need metrics that reflect the number of pages required to provide instructions for completing a task or to describe a product feature.

We use cookies to monitor the traffic on this web site in order to provide the best experience possible. By continuing to use this site you are consenting to this practice. | Close