Business Analysis: Models for Everyone

 

I’ve been anxiously awaiting the time to write this post because it is about something near and dear to my heart – analysis models! Models are by no means new to the business analysis profession, but I’m thrilled to say that the use of them has become much more prevalent – almost mainstream – over the last five to ten years. Given that, you would naturally expect the PMI foundational standard in business analysis to have a home for them, much like they are in section 4 on Requirements Elicitation and Analysis of Business Analysis for Practitioners: A Practice Guide.

As described in the practice guide, models are visual representations that efficiently arrange and convey information in a concise manner. Models can be in the form of diagrams, tables, or structured text. Models are used in planning, analysis, design, construction, training, coaching, and many more roles. There are many types of analysis models and the following table shows the ones we found to be most prevalent, and are therefore covered in the practice guide. The models are grouped by categories in this table to help you think about the kinds of models that exist more precisely than just listing them. Most projects, programs and portfolios will need to use at least one model from each of these categories. In fact, the table of models is sort of a model in itself because it organizes information for better understanding!

Table source: PMI, Business Analysis for Practitioners: A Practice Guide

The when, why, and who of it

As analysts, models help us create order out of information so that we can find inconsistencies, gaps, and extraneous information. Analysis models also help us to represent information so that others can consume it. Our business stakeholders, construction teams, testing teams, executive teams, and managers all use models to understand what the product needs to do and why.

But wait, what about agile?

Sue’s recent post highlighted how business analysis is applicable in both predictive (such as waterfall) and adaptive (such as agile) life cycles. As she points out, models are useful in both life cycles! The choice of models, formality of their representations, or timing of their use might vary from adaptive life cycles to predictive life cycles. But frankly, those same things vary from any waterfall project to any other waterfall project also. There is no “one size fits all” approach to modeling. In agile approaches some scope and goal setting models are more likely to be created early in the project to get a lay of the land. However, more detailed models to identify acceptance criteria are used just in time to plan iterations as part of elaborating backlog items.

All industries

If you work outside the software or IT arenas, you might be wondering whether analysis models are even applicable. The short answer is absolutely! However, some models are better than others. For example, some form of process flows or use cases can be used to describe how users will use or interact with anything – be it a shopping center layout, an insurance claims process, a city intersection, a turn signal on a car, or a rocket ship. On the other hand, some models such as display-action-response models are only useful if there is some kind of a user interface to be specified.

I once met some engineers from a company that builds heavy machinery. They didn’t work on software, but rather they were in the product design department for the physical machines. Once introduced to analysis models, they quickly adopted business objectives models, process flows, and feature diagrams to improve their requirements and design work.

What to expect

PMI’s foundational standard in business analysis will explain analysis models in general, describe why we create them and how we analyze them, and offer a visual template for many types of models. Business Analysis for Practitioners: A Practice Guide offers more depth on when you use each of the models, how you use each of them in relation to requirements, and an example of each model. The two publications in combination will provide a wealth of knowledge for those just getting started in the modeling space or a quick lookup reference for the more experienced professional.

Like I said earlier in the post, there are so many different types of analysis models. I’d love to hear what the community thinks about models. What are your favorite models? Are we missing any key ones that you’d like us to consider for inclusion in the standard? Do you really like or not like any of the ones I listed here?