DITA Metrics: Developing Cost Metrics

By Mark Lewis, special to The Content Wrangler

Table of Contents

Introduction

Cost of Metrics Overview

Cost of a Project

Cost of a Topic

Cost of User Guides Without Topic Reuse

Cost of User Guides With Topic Reuse

Cost Comparison: User Guides With and Without Topic Reuse

Cost of a Reusable Master Topic

Cost of User Guides With Reusable Master Topics

Cost Comparison: Topic Reuse Versus Reusable Master Topics

Conclusion

About the author

Introduction

You’ve read all the papers (and attended all the webinars) on return on investment (ROI) for XML and you get it. You’ve already concluded that moving to the Darwin Information Typing Architecture (DITA) will likely save you tons of time and money. But management says, “Prove it!”. This paper helps you determine the cost portion of the ROI calculation. What are my costs now? What will my new costs be with DITA? And what is the difference—my savings? This white paper is the first in the DITA Metrics series. The series will discuss cost metrics, reuse metrics, and a reuse strategy. This paper is the first in the DITA Metrics series. It describes one model for calculating the cost of a DITA project. After doing some content analysis on your own documentation set, you can customize this cost model to suit your documentation project needs. In the end, you should be able to speak the financial language of managers and prove to them in dollar signs the value of moving to DITA.

To benefit from this article, you should have at least an intermediate level understanding of DITA including topic structure, elements, conrefs, child maps, and filtering/conditional processing.

For your convenience, we’ve provided a downloadable PDF of this article.

Cost Metrics Overview

The cost to develop content and reuse percent values are standard components in many ROI calculations. We need the cost and reuse values for a DITA project to determine DITA ROI.

This paper focuses on the cost of content creation and introduces various levels of reuse into the model. We’ll begin with a deep dive into the cost of creating DITA topics and then incorporate the cost of unique content, identical content and similar content.

Over the course of the series, we will discuss the following components of our cost model:

• Cost of content creation (in this paper)
• Cost of content analysis and inventory
• Cost of review and project management
• Cost of filtering
• Cost of publishing
• Cost of content maintenance
• Cost of converting legacy content to DITA
• Cost of translation

Cost of a Project

Let’s show how our model is used to determine the cost of creating user guides for three models of a fictitious personal digital assistant (PDA). PDA One has a base set of features. PDA Two has all the features of PDA One, plus several additional features. PDA Three has all the features of PDA One and PDA Two, plus several features unique to PDA Three. This documentation project contains lots of identical content and unique content.

The first version of our documentation project is relatively simple so that the associated cost model is also simple. Later, we’ll introduce more of the content reuse and conditional reuse features of DITA, and show how to incorporate these into the model. Gradually, we’ll increase the complexity of our project and our cost model.

Cost of a Topic

The first step is design the content creation component of the model.  Traditional cost metrics focus on the cost of a page [1]. Since pages are similar to topics, we will start our discussion with determining the cost o creating DITA topics. Costs are expressed in terms of content creator labor hours.

Table 1 shows the cost of creating Task and Concept topics. For simplicity, we exclude Reference and other topic types, but you can easily customize the model to include them if needed. The scope of these estimates is a topic and does not include time for project/publication level activities such as designing the document outline, user task analysis, project management, implementing context-sensitive help, testing, status meetings or design meetings.

Table 1 (Cost of a Topic)







*Includes time to learn the product through interviews or research.

** Screen shots/images: Includes time to create sample data that would be shown in screen shot, capture the image, convert the format, name the image using naming conventions, and store it in the repository.

*** The elements listed here are optional. For example, Concept topic > screen shot, or Concept topic > feature description.

Now we have an approximate cost range in hours for creating Task and Concept topics that we can use in our model.

Cost of User Guides Without Topic Reuse

Table 2 through Table 5 shows the cost of developing a user guide for each of the three PDAs in DITA without taking advantage of topic reuse.

Table 2 (Cost of PDA One User Guide)

Table 3 (Cost of PDA Two User Guide)

Table 4 (Cost of PDA Three User Guide)

Table 5 (Total Cost All User Guides Without Topic Reuse)

This is really a worst case scenario that is not realistic because moving to DITA and not reusing any content is highly unlikely. But this simple project is a good starting point for the model and a base to which we can add reuse features of DITA. As you will see, the reuse features that you incorporate can be different for each documentation project.

Cost of User Guides With Topic Reuse

Now we’ll incorporate the cost of reusable topics (reusable in more than one user guide) in our cost model.  Reusing topics is nothing new. This feature has been available in help authoring tools for more than 10 years.

Table 6 through Table 10 shows the cost of developing a user guide for each of the three PDAs taking advantage of reusable topics. Some topics for the PDA One user guide may be reused verbatim in the PDA Two and PDA Three user guides because the topics are identical.

Table 6 (Cost of Reusable Topics)

Table 7 (Cost of Topics Unique to PDA One User Guide)

Table 8 (Cost of Topics Unique to PDA Two User Guide)

Table 9 (Cost of Topics Unique to PDA Three User Guide)

Table 10 (Total Cost All User Guides With Topic Reuse)

Cost Comparison: User Guides With and Without Topic Reuse

Although the scenario without topic reuse is highly unlikely and not realistic, just to be thorough, we are showing a cost comparison in Table 11 through Table 13.

Table 11 (Total Cost All User Guides Without Topic Reuse)

Table 12 (Total Cost All User Guides With Topic Reuse)

Table 13 (Savings)

For over a decade, significant savings have been achieved reusing topics in multiple publications.

Cost of a Reusable Master Topic

The project is simple when topics can be reused verbatim in multiple publications. But what happens to our model when there is sufficient variation in our products that we cannot write a single topic to describe a given feature? Perhaps the product screen shots are different, an extra note or warning is needed, or a button has a different label. For all three user guides, some topics are similar. Most of a similar topic is the same for each PDA and can be shared. So, if all three versions of that content are included in one topic, then all versions of the user guides may be published from this topic. Using filtering metadata, content that is unique is marked as belonging to a specific PDA. When the user guide for a specific PDA is published, content that is specific to the other PDAs is filtered. The filtering feature in DITA is also known as conditional processing and it is what allows us to create and use reusable master topics.

Table 14 shows the cost of creating reusable master Task and Concept topics.

Table 14 (Cost of a reusable master topic)

It would complicate our model to incorporate conrefs to a large variety of content types. Therefore, we are limiting the incorporation of conrefs in our model. We’ll discuss conrefs later, but for now a simple example is to reuse feature descriptions or screen shots that were created in Task topics by conref’ing them in your Concept topics. This would reduce the cost of the Concept topic by several hours. You will be able to customize the model to incorporate your use of conrefs in your project.

Now we have the estimated cost for reusable master Task and Concept topics that we can use in our model.

Cost of User Guides With Reusable Master Topics

Now we’ll incorporate the cost of reusable master topics into our cost model.

Table 15 through Table 16 shows the cost of developing a user guide for each of the three PDAs taking advantage of reusable master topics.

Table 15 (Cost of Reusable Master Topics)

Table 16 (Total Cost All User Guides With Topic Reuse and Reusable Master Topics)

Table 17 compares the cost of creating unique topics to reusable master topics.

Table 17 (Cost of Unique versus Reusable Master Topics)

Reusable master topics cost significantly more to create, but with proper planning they can be used in multiple publications such that the overall cost of content creation for your publications drops dramatically.

Cost Comparison: Topic Reuse Versus Reusable Master Topics

Table 18 through Table 20 shows a cost comparison of a project taking advantage of reusable topics to a project that is taking advantage of both reusable topics and reusable master topics.

When filtering and reusable master topics are used, the number of unique topics and reusable topics that have to be created is drastically reduced. This is where the cost savings of using DITA really leaps ahead of conventional help authoring tools and technology.

Table 18 (Total Cost All User Guides With Topic Reuse and Reusable Master Topics)

Table 19 (Total Cost All User Guides With Topic Reuse)

Table 20 (Savings)

Conclusion

“What is the cost of DITA?” is the question that is the focus of this paper.

Chances are that if you are reading this paper, then you’ve read the papers on the promise of XML, the ROI of DITA, the clear savings in translation costs alone, and you get it. However, you are being asked to justify the cost. How much is it going to cost to develop and publish content using DITA? Knowing percent reuse is important, but equally as important is the upfront cost? And, the savings are in more areas that just translation. There are savings in both content creation and content maintenance due to content reuse and conditional reuse.

The cost to develop content and percent reuse are often discussed separately. But to know the cost of a DITA-based documentation project, both of these values need to be accurately incorporated into a cost model. In this paper, we focused on cost metrics. Later in this series we will focus on reuse metrics.

For a given project, we must know the cost to create content without reuse and the cost with reuse. The difference is the savings. The cost to create content with reuse is equal to the cost to create the unique content and the reused content. That’s the upfront cost. Knowing this will allow us to more accurately calculate the cost of translation as well. What must be translated is reduced to unique content and reused content.

In Part 1 of this paper, we covered the following cost metrics:

• Cost of content creation
  
  • unique content
  • reused content
• Cost of a Task topic
• Cost of a Concept topic
• Cost of user guides without topic reuse
• Cost of user guides with topic reuse
• Cost of a reusable master topic
• Cost of user guides with reusable master topics

The most important things we accomplished in this paper included determining the cost of creating a DITA-based topic rather than a traditional page. We then incorporated conditional reuse/filtering and determined the average cost of creating a reusable master topic. We observed that the biggest savings resulted when reusable master topics are incorporated. The flexibility and diversity of conditional reuse in DITA differentiate it from typical help authoring tool technologies and offer greater savings in not only content creation, but also content maintenance.

The ultimate question.

In Part 2 of this series (coming soon), we will look at other costs that need to be accounted for in a DITA project:

• Cost of content analysis and inventory
• Cost of review and project management
• Cost of filtering
• Cost of publishing
• Cost of content maintenance
• Cost of converting legacy content to DITA
• Cost of translation

Other papers in the DITA Metrics series (coming soon!):

• DITA Metrics: Cost Metrics – part 2
• DITA Metrics: Reuse Metrics
• DITA Metrics: Reuse Strategies for Minimizing Cost

Join the DITA Metrics Group

Interested in learning more about DITA metrics? Join the DITA Metrics group on The Content Wrangler Community. The group is lead by the author of this paper, Mark Lewis.

About the author

Mark Lewis works for YOU, the tech writer, and the information architect. Mark has received STC awards for Distinguished Chapter Service and the Florida Technical Communications Competition. Currently, he is the DITA Product Manager for Usability and a product Evangelist for Quark. He provides product direction and user experience designs for Quark’s structured authoring products allowing everyday authors to create reusable content without knowing the details of XML syntax. Mark presents at conferences and other industry events on topics including: object oriented design methodologies (for non-programmers to help jumpstart their understanding of structured authoring), information architecture, and the promise of DITA and XML. Mark manages The Content Wrangler Community Groups – WritingOBJECTively and DITA Metrics. Send questions or feedback to mlewis@quark.com or hyperwriters@hotmail.com

————

1. For more information on traditional cost metrics, see Lasecke, Joyce. “Stop Guesstimating, Start Estimating. In Intercom, February 2006 issue, Society for Technical Communication. Return to article

Similar Posts:

• Transitional Text: Do We Really Need All that Glue?
• DITA Maturity Model Contest – Win an iPod Touch, Free Tickets To DocTrain East
• The Power of the Crowd: Finding DITA Resources and Information

Tags:Adoption, DITA, Lessons Learned, White Papers