(Re)defining Business Analysis

In the last couple of months, I’ve started the early planning process for version 3.0 of the BABOK® Guide. Before everyone starts asking:

  • The release date is sometime in 2012, which has always more or less been the plan. As we get closer to that time we will communicate other information. As with 2.0, you can expect a 4-6 month period between the 3.0 release and updates to the exams.
  • I expect it to be a significant upgrade, with some new tasks and possible reorganization of KAs, but not nearly as big as the 1.6 – 2.0 change. Most tasks will be the same, techniques will still be in their own section, and so forth.
  • Don’t ask about volunteering yet. The call for participants in the core team will be published in BA Connection when we’re ready to kick off.

So, what’s changing? Well, one of the things I’ve spent time on to date is the definition of business analysis. In version 2.0, the definition is:

Business analysis is the set of tasks and techniques used to work as a liaison among stakeholders in order to understand the structure, policies, and operations of an organization, and to recommend solutions that enable the organization to achieve its goals.

I felt that there was room for improvement in this definition. It’s wordy, it’s a little complex, and it includes things that aren’t strictly accurate. For instance:

  • It says that “BAs work as a liaison among stakeholders”. But do we always? I worked on one project where something like 90% of my analysis work was interpreting a government-defined standard for data transmission and figuring how a software application should translate that into billing information. I wasn’t really liaising anything. Also, as you move up the enterprise, there’s less liaising and more researching and recommending. The liaison role is important, certainly, and many BAs focus a lot on it. But it’s not necessary…because you can do business analysis without it.
  • “understand the structure, policies and operations”…what about strategy? What about IT? I can (and we do) claim these are implicit, but the risk of any list like this is that you may find that you’ve left something out.
  • “enable an organization to achieve its goals”. Two problems here. One is that it implies that everything a BA does must fit with the overall organizational strategy. That’s not wrong as such, but it may not always be practical. Does every single requirement really tie back to an organizational objective? Plus I have another objection–organizations don’t have goals. People have goals. The leadership of the organization may have goals, but the organization itself can’t.

The key to any good definition is that it should include everything that is necessary and nothing that is not. In other words, everything in the BABOK® Guide should in some way be traceable back to that definition, and we should also strip out anything that is “aspirational”. That means removing anything that talks about what a BA “should” do or know unless it can be proven to be necessary. It should also be applicable to everything from maintenance or continuous improvement to enterprise-level work.To see the kind of things we’re looking at, check out the Competency Model (pages 6-7) which gives a glimpse of the roles that the definition should encompass.

Also, I intentionally avoided using words like “facilitate” or “support”. There were quite a few “support” tasks in version 1.6 and we removed every last one of them from 2.0, either by narrowing the scope of the task or by redefining the nature of the work. When you get into “support” in a professional body of knowledge, you’re actually crossing the line into another profession that is actually responsible for the outcome in question. For instance, as a business analyst (job title) the project manager, developer, or tester may ask me to help them out by doing something. That’s fine, and as a team member I should help other people–but in that case I’m assisting you in performing your role, not doing business analysis. As a business analysis professional, if something is an outcome of business analysis I and only I am accountable and responsible for it.

(In practice, this is not always true. Many business analysts are too junior to have final accountability for the ultimate analysis product, which may rest with more senior analysts or even managers. However, we’re talking about business analysis here, not your job or mine, and the BABOK should encompass everything needed to reach that final result).

So, after a lot of effort, I came up with the following (not final) definition.

Business analysis defines the capabilities that enable an organization to create value for its stakeholders.

There’s still validation to do before this becomes final, but I think it captures the essence of the profession. Comments are, of course, welcome.

BA Insight Relaunch

So, after a number of years, I’ve finally gotten around to relaunching BA Insight.

BA Insight will serve as a place to talk “unofficially” about business analysis and other topics of interest to me. Over the last couple of years I’ve felt a certain pressure to make sure my posts on IIBA were in keeping with my official role. Some people seem to get uncomfortable when the guy responsible for the BABOK Guide expresses a firm opinion about things that disagrees with theirs. Most times, that stems from a concern that their views will be dismissed by IIBA. I’ve had a lot of practice separating my personal views from my work for IIBA, though. Nevertheless, I think it’s best that I express those views on a “personal” channel.

I’ve pulled in many of my old posts from the IIBA SLT Blog, deleting those that were pure status updates or the like. Official IIBA stuff will continue to be blogged about on the IIBA Community Network, and of course some things will be cross-posted.

This site will also serve as a home for my business. Yes, I work for IIBA but it’s as a contractor, not an employee, and it’s not my sole source of income. I don’t do work that would conflict with my IIBA responsibilities, but I am available for other short-term consulting engagements or for side projects that can be done part-time. Drop me a line if you want to know more.

Other than that, I look forward to getting back into the stream of blogging again.


Thoughts on 3.0

So, it’s been a month now. That’s enough of a break, right?

Here’s my (very early) thoughts about what version 3.0 might look like. A big consideration for me will be that it shouldn’t be too much longer than 2.0. I can see adding a chapter on lifecycles and methodologies to explicitly show how the BABOK maps against them, for instance, and other materials that might bulk it up a bit, but I’d be very reluctant to see it go over 300 pages. So that means that it would stay at about the current level of depth and detail on the topics it covers.

First, the obvious–whenever it comes out, we will run surveys to see if there are any important changes in which techniques are generally accepted (either more so or less so).

Then, there may be some restructuring of tasks. Nothing like the 1.6-2.0 transition, to be sure, but I think there’s some opportunity to structure things a little more consistently and also a couple of things that might deserve to be tasks in their own right.

However, I think the biggest changes will come from various efforts now underway. We’re doing analysis on BA competency models, and version 3.0 will emphasize those qualities that we find are most important to being effective as a BA. We expect to see work done to align the BABOK more closely with other critical standards. And thirdly, the field itself will change, as we see how Agile methods, BPM, outsourcing, and other current hot trends play out over the next few years. (My personal guess is that at least one or two of those will peak this year or next, but we’ll see).

As always, though, we’re open to feedback. Let me know! And in the meantime, I have plenty to keep me busy.

Kevin Brennan, CBAP
VP, Professional Development


Change x4

I should mention that my job around here has changed too. As you might have noticed in the posts below, I’ve been signing off with a new title.

We’ve merged the Body of Knowledge and Professional Development groups into a single group. I am now heading up the new Professional Development team as the VP of Professional Development, and I’m now responsible for the development of IIBA standards (yes, plural) the EEP program, our webinars, and developing other educational opportunities and resources for our members.

I’d like to thank Cherifa Liamani, our former VP of Professional Development, who will be staying on as part of the PD team. I’ll be working a lot with Julian in his new role as well.

And finally…version 2.0 is now complete. We have a little production work left (such as verifying the list of contributors, creating the table of contents, and similar) but the main body is done and has been sent to the IIBA Board of Directors for approval. Again, thank you to everyone who contributed, participated in reviews, commented on it, or who has contributed to shaping ideas around the profession of business analysis over the last few years. We’ve been working on this for four-and-a-half years, and I am both happy and strangely sad to see it go. Sure, there will be other revisions and other standards, but my life just won’t be the same without people asking me “when is the BABOK coming out?”

Kevin Brennan, CBAP
VP, Professional Development


Requirements and Solutions Are Not Software

Don’t get me wrong, they can describe software. But they don’t have to. One of the things we’ve been trying very hard to communicate is the idea that the BABOK describes a much broader, general process for identifying solutions that meet business needs.

Let’s give a concrete example. Let’s assume that your organization has assessed your existing HR operations and come to the conclusion that they’re not working very well. Remember, “solution” is not “software”–in this context I’m talking about the entire HR department! At this point in time all that is known is that the HR department is viewed as being ineffective and you need to fix it.

Our business need, then, is to “fix HR”. But what does that mean? Enterprise analysis talks about the work required to understand what we mean by a “fixed HR”…in other words what are the business benefits that accrue from fixing HR? How much value does fixing HR generate for our organization? In order to get there we are likely to spend time doing solution assessment and validation as well, because that talks about how to identify problems in a current solution that we want to fix. Once those activities are complete, we have a business case for fixing HR. However, we have a challenge–we have a lot of very different options for this. We could outsource the department, build or buy a new HR system, train the employees in new practices, or some combination of all those things.

So what do we do now? We start defining stakeholder and solution requirements! But these requirements certainly are not a description of a software application. Instead, we’re going to be looking at the legal and regulatory issues we face, the workload the department has, the processes it has to execute, and the way it integrates with the rest of the organization. We have to define these requirements in order to assess the solution options I outlined above.

Let’s say we do that, compare the options, and decide that the best option for our particular needs is to build our own HR system (for some reason we have very unusual HR needs). Our solution is still probably not the HR system because we also need to understand other issues, such as how our HR processes will be affected. In BABOK terms, the processes and the software are both solution components. We can define requirements for the solution at whatever level is appropriate, but eventually we proceed again to perform solution assessment and validation. The allocate requirements task actually addresses figuring out which requirements are implemented in our business processes and which are in our software.

Once we decide that the software is going to do a particular task, we get to go back to eliciting requirements and analyzing them. We go back because once we’ve decided that the application will do something, we probably need to drill down into what that is (as opposed, for purposes of this example, to processes or business rules that may be left to the discretion of staff). For instance, if staff access the HR system over the internet, there are security requirements that may have to be met.

This goes on until you’re done, for whatever value of “done” you have in mind (the BABOK intentionally doesn’t draw a hard line between “requirements” and “design” other than to say that a requirement is something that meets a stakeholder need).

Certainly a solution can be a software application, and a requirement can describe a characteristic of that application. But when you read the BABOK, don’t assume that it is. A solution is something that meets a business need, and the solution scope is the full set of capabilities that are needed for that purpose. And as you can see from the example above, you also shouldn’t assume that the solution is going to be completely defined in a single pass through the knowledge areas. In many cases you’ll cycle through them a number of times before the solution is fully defined.

Kevin Brennan, CBAP
VP, Body of Knowledge


Version Three

Yes, I know.

As I found when we put version 1.6 out the door, there are issues with version 2 that we just won’t have time to fix, or which it’s too late to fix (because the impact it would have on other work underway is bigger than the benefit of the change). Looking back on my life in IT, this is a feeling I have every time I release something.

Yesterday I spent a little time pulling together a proposed outline for BABOK version 3. I won’t post it up here because it’s too early to be talking in public about this stuff. We might very well find after version 2 rolls out that these aren’t the right changes to make in any case, based on the feedback we get.

With version 2 the BABOK shrunk from 60-odd tasks to 32. I think we can actually reduce that a little further, and I also think there may be one or two tasks that should be added. In addition, there’s some tasks that should move between KAs.

Don’t get me wrong, the changes I’ve identified so far are a lot smaller than the changes from 1.6 to 2. I could describe them all in a blog post, unlike the 1.6 changes which will take a 15-page document to address at even a high level. With the exception of a new task or two, these changes wouldn’t affect the scope of the BABOK at all. They’re mostly about reorganizing things a little to deal with some persistent issues we’re encountering or misconceptions that are affecting people’s ability to understand what the BABOK actually says.

Finally, yes, we are getting very close to being able to announce the release date.

Kevin Brennan, CBAP
VP, Body of Knowledge


CBAP Exam Advice

My last post sparked a question about the CBAP exam (as talk of the BABOK release usually does). I think it’s best to answer it here where people will see it.

Our existing policies around writing the exam are not changing. If you apply for the CBAP exam and are approved, you have one year from the date you are approved to write the exam.

This means that you can apply under 1.6 and write the 2.0 exam, if you so choose. If you have applied within the last few months, you will certainly have the option of waiting for the 2.0 exam. There’s no reason you should wait for the 2.0 exam, but you could if you wanted to.

The exam itself will cut over to version 2.0 on a specific date, which has yet to be determined but will be at least three months after the release of version 2.0. Everyone who writes the exam after that date will be tested against BABOK version 2.0, no matter which version was in place when they applied.

If you are currently working towards writing the exam (even if you’re still in the application stage), you still have plenty of time to submit your application, get approved, study for the exam, and write it under 1.6. If you have not yet started this process I would suggest that you assume that you will be writing under version 2.0. You can still apply–as I said above once you’re approved to write the exam you have a year, and the version change will not affect that.

Kevin Brennan, CBAP
VP, Body of Knowledge


BABOK-SOA Mapping

“Solution” = “Service”

Kevin Brennan, CBAP
VP, Body of Knowledge

Structural Metaphors

So last night, as I lay in bed trying to get to sleep, I was thinking about the BABOK. Yes, that’s sad. Anyway, what I was thinking about in particular is the tendency of people to look at it as describing a process. I fear the only way to dislodge that thinking is to replace the model with a different one familiar to BAs.

I think it would be a closer approximation of our intent if readers looked at the BABOK as the set of use cases that constitute business analysis. Like use cases, they have goals, pre-conditions (the inputs) and post-conditions (the outputs). Also like use cases, but unlike processes, there are no rules telling you when to perform tasks in the BABOK, when to loop back to previously executed tasks, and so forth.

Activities in a process flow are strongly connected. An activity cannot begin until its predecessor is complete, and if it runs into difficulties then the process must specify how to handle that. A use case, on the other hand, has no real connection to other use cases (leaving «include» and «extend» aside for a moment). It can be performed whenever its preconditions are met. If there’s a problem with one of the inputs, there’s no specified process one follows to fix it.

There are, of course, key differences. The BABOK doesn’t tell you how to perform tasks, just what the task is. The model is not at all perfect. It’s just a better model than thinking of the BABOK as a process.

Another way to look at it is functional decomposition. As you read this, please keep in mind that this is an extremely informal description of the BABOK. Don’t try to parse these words too closely.

At the top level we have the subject matter, which is business analysis. Informally, business analysis is about analyzing how organizations function in order to help stakeholders understand how to accomplish organizational goals and objectives. We can break that down into two major subtopics. One is the analysis part. The other is the dealing with stakeholders part.

On the analysis side, we break it down into three KAs. That itself breaks down into three KAs: one for defining the problem, one for defining solutions to that problem, and one for making sure that when a solution is delivered it actually solves the problem.

On the stakeholder side, we also have three KAs. One involves getting information we need for analysis, one involves providing analyzed information to people who need it, and the third focuses on organizing the work we need to accomplish to perform all the other KAs.

Each KA is then broken down further into 4-7 tasks that detail what is required to do those things in the context that business analysts operate in. That context, by the way, is what makes business analysis distinct from other roles in its “family” like Product Management, IAs, Usability Professionals, and so forth. At the very high level I described above, you could apply the same structure to all of them, but each would perform somewhat different tasks and use different techniques to define the problem, validate the solution, and so forth.

Finally, there’s the Underlying Competencies, which describe the soft skills and general knowledge that helps you do all the things above more effectively.

I could write about how techniques fit into this but this post is already long enough.

Kevin Brennan, CBAP
VP, Body of Knowledge


Techniques in BABOK v2

Once again, a big thank you to all who participated. We received over 1,100 responses to the survey. On the whole, I believe that respondents were generally more experienced and had a wider range of exposure to different methodologies than the industry norm, which means the results are very likely to align with actual industry practice. We did not find major differences in the results based on experience with different methodologies.

In analyzing the survey, we combined some techniques into a single description where they are used for essentially the same purpose in business analysis. Some changes were made to the list as well based on our assessment of the value of a given technique. I also excluded some techniques from this list that only support a single task in the BABOK, as they will be described in the text of those tasks (for example, requirements prioritization techniques). More detail on the process and on the categorization of Primary and Secondary techniques can be found here.

Finally, I can assure you that the BABOK will include a strong disclaimer to make it clear that the IIBA does not intend or desire that this listing be viewed as a recommendation on our part as to the value of a given technique. We’re not saying that this is what you should be doing; we’re saying that this is what most BAs are doing and so if you know these things you’ll be prepared to work in most organizations.

Minor changes to this list in the final cut of v2 are still possible, but it will look very much like this.

Primary

Primary techniques are ones that over 80% of BAs use occasionally or regularly.

  • Brainstorming
  • Business Metrics and KPIs
  • Business Rules
  • Data Dictionaries and Glossaries
  • Data Flow Diagrams
  • Data Models (ERD, Class Diagram)
  • Interviews
  • Issue and Defect Tracking
  • Non-functional Requirements Analysis
  • Organizational Models
  • Process Models (includes Flowcharts and Activity Diagrams)
  • Requirements Workshops
  • Scenarios, Use Cases, and User Stories

Secondary

These are used regularly or occasionally by more than 50% but less than 80% of BAs.

  • Acceptance Criteria
  • Architectural and Compliance Frameworks
  • Benchmarking
  • Document Analysis
  • Environmental Analysis (includes Market, Industry, Competitive)
  • Estimation Techniques (for Solution Scope as distinct from BA tasks which are covered in the BAP&M KA)
  • Event and State Model
  • Focus Groups
  • Functional Decomposition (includes Work and Product Breakdown Structures)
  • Gap Analysis
  • Impact Analysis
  • Interface Analysis
  • Lessons Learned and Retrospectives
  • Observation
  • Prototyping
  • Requirements Risk Analysis
  • Root Cause Analysis
  • Scoping Diagrams (Context Diagram, Use Case Diagram)
  • Sequence Diagrams
  • Storyboards/Screen Flows
  • Structured Walkthrough/Inspection
  • Survey/Questionnaire
  • SWOT Analysis

Kevin Brennan, CBAP
VP, Body of Knowledge