The Agile Process Showdown at Jama: Scrum versus Crystal or Maybe a Hybrid?

At Jama, we work with a wide range of companies using different software development processes. And, despite all the evangelism by the process gods in our industry, if there is one thing we’ve learned over the years it’s that no single process is the best practice for all companies. What works for one company, may not work for another.

Too often, our industry gets caught up in labels, and “Agile” is a great example of that right now. Agile is as trendy a buzzword as “innovation” or “Web 2.0”, but if you lined up 100 different developers and asked them what Agile means, you might get 100 different answers. But, is that really the point – are you Agile or not? Agile itself isn’t a single process – there’s different flavors such as Scrum, Crystal, XP and others. And, as the Agile Manifesto outlines, the underlying concepts of Agile are common across all flavors, such as breaking things into smaller projects, welcoming changing requirements, delivering frequently and having frequent conversations with the customers and users of the application you’re building.

So, the question isn’t so much about whether you’re Agile, but which techniques within these Agile methodologies will work for your team? How do you build great products that are safe, reliable and customers love to use? That’s what it’s about.

Like a lot of dev teams, we’re constantly looking for ways to improve our own process internally in building our product, Contour. So we decided to blog about our experience with Agile techniques, not as experts or evangelists (there’s plenty of those), but simply to share what’s working, what’s not, what we’ve learned through our own experimentation and retrospectives.

Over the next several months, we’ll be incorporating different techniques and enhancing the ones we already use to optimize our process at Jama. Also, we’ll leverage what we learn to enhance the features within Contour to support a range of techniques – for Agile, traditional methods, CMMI or other practices our customers need.

Last week, as an interesting exercise, we had two of our senior developers do a showdown of Scrum vs. Crystal Clear – two of the more popular Agile methods. Derwyn representing Crystal Clear and Frank representing Scrum. What ensued was a bloody, mixed-martial-arts-style cage match of epic proportions. Well…not really. Actually, it was a civil meeting over lunch with our team and both came armed with notes from their favorite books and brief presentations on the realistic techniques that we can weave into our process at Jama.

Scrum versus Crystal Showdown

The net result of our discussion was that we’re not purists of any one specific methodology. There’s a blend of techniques that we’re leveraging. In fact, one of the principles Crystal teaches is this sense of evolution - that your process needs to fit your team, and not the other way around. Point Crystal. Oh wait, we’re not keeping score…

Here’s the rundown of what we’re doing at Jama (click image for .pdf):

Agile process at Jama

Our development team works within the same office, so we have the advantage of being able to do daily stand-ups and we use a large whiteboard with sticky notes for managing tasks that are active for our current iteration (a.k.a. sprint). We use Contour to manage our product backlog which includes new feature requests, defects, requirements and tasks.

As we do our planning, we group these together into releases and shorter sprints. In context to labels, I guess you could call our mixed approach “Agile Hybrid” – and really the only way to top one buzzword is to add another to it, right? In all seriousness, our approach is rooted less in theory and more in what we realistically see ourselves adopting and having success with - and it evolves.

Techniques we’re doing now:

  • Daily stand-up meetings
  • Frequent delivery/iterations
  • Product backlog
  • Whiteboard with task cards
  • Requirements documentation (yes, it has a role in Agile)

Techniques we’re not adopting:

  • Use of points or velocity score
  • Self-organizing teams
  • Pure pair programming (we do side-by-side programming at times)

What we’re experimenting with:

  • Blitz planning
  • More thorough retrospectives with each release
  • Implementing burn-down charts

In future posts on our use of Agile, we’ll share more examples of the techniques we’re experimenting with and we’ll tackle the often-asked question, “What’s the role of requirements management within Agile development?”

Share with us your own experience: How has your process evolved? Which techniques of Agile have you had success with?  Email us directly or feel free to post a comment here.

If you’re new to Agile, here’s a few of our favorite resources to help get you started:

Innovation Design Keynote at Autodesk University – shifting from “What if?” to “What else?”

Earlier this morning, I was among the 10,000+ attendees at the Autodesk University event in Las Vegas and took in the keynote address that included Carl Bass, Autodesk’s CEO, Jeff Kowalski, Autodesk’s CTO and Tom Kelley, the general manager of IDEO, one the world’s premier product design firms.

A couple of key take-aways emerged.

1.  From What if to What else?

The Autodesk executives showcased several amazing examples of innovation, from buildings to bridges to surf boards to cities (yes, complete 3-D models of cities)  – many of which were designed by Autodesk customers in the room.  They shared their vision for how software (like theirs and others) with enormous computing power will enable us to shift from asking “What if” questions to asking “What else” questions.

Before, an architect, designer or engineer might ask, “What if we change the shape?”, “What if we change the material?”, “What if we change the user interface this way?”  With today’s tools, we have to imagine all the different What if scenarios ourselves.  The human mind can only process 7 bits of information at a time, and software today is still a relatively passive experience, reacting to the inputs and commands we give it.  Regardless of how smart we are (or think we are), the software tools we use to build things, should and will soon become much more proactive.  The apps will assess all the possible variables and combinations for us (including ones we probably wouldn’t have thought of) and offer us back options to choose from that provide the best design, the best experience.

An example was given with Boeing, where they were already able to do in 6 days what used to take 8 years, and analyzing exponentially more options and identifying the safest, most reliable solution to their engineering challenge.   Essentially, the true power of software will be that it will enable people to digitally conceive, design, and experience things before they are real – which has enormous benefits in terms of speed, costs and quality.

2. Innovation & Design - At Work Together.

Tom Kelley of IDEO in his presentation made the point that innovation and design aren’t mutually exclusive disciplines, that they’ve blended together – whether you’re developing a new eco-friendly building, a new bike, an entire city or a new software application – design isn’t superficial as some might think – if you want to innovate you have to design.   And, design is as much about identifying the right problem as it is engineering the right solution.

Tom touched on his 2 favorite personas from his latest book, “The 10 Faces of Innovation” – the Anthropologist and the Experience Architect.

He joked about how there is general skepticism around the anthropology work that is needed in the innovation process, of how clients often ask, “Can we just skip that and just do the real design and engineering work?”  He had a nice way of summarizing the value of each role.

The Anthropologist is great at observing customers and “finding the opportunities that are hidden in plain sight.” What he called “Vuja de” or the opposite of Déjà vu – meaning looking at something and seeing it differently with new eyes.  He gave the simple example of how IDEO designed a new kid’s toothbrush for Oral B.  The pre-existing assumption was that we all know how to brush our teeth.  We’ve been doing it for centuries, so it would seem logical that kids are just smaller versions of adults, so just take an adult toothbrush and design a smaller and skinner version of it, right?  Wrong.  The anthropologists at IDEO observed that when kids brush their teeth, they can’t hold the brush with their finger tips the same way adults do; they hold it in their fist.  So, the smarter, better design is to actually make the handle bigger and softer so the kids could easily hold it while brushing.  Simple, but brilliant, and in plain sight.

These kinds of opportunities exist all over in our field of work in software application design – we just have to be open to looking for them.

The value of the 2nd key persona, the Experience Architect, is that they think beyond the product or service or user interface and focus on the customer journey from start to finish.  Paired together, these 2 personas are critical to the process of building great products, services and experiences.

3.  Software Design - We must aspire to the Wet-nap user interface.

Even though Autodesk’s tools primarily apply to the design of physical products, these same principles apply to our world of software design and development.  We should challenge ourselves to put our anthropologists and experience architect hats on everyday and look at our work with new eyes.  Too often as developers, work can be very specification-driven.  The logical answer to the problem is believed to be understood and thus documented in a static requirements specification doc, and now go engineer it.  But, innovation doesn’t happen that way.  Continue to ask ourselves, “What else is possible?”

As Tom Kelley put it, we should aspire to the “Wet-nap user interface” even when developing complex, sophisticated products.  The directions on the back of a wet-nap are so simple, they read, “Open and use.”  When we can build software that is that simple for our users, that’s when we know we’ve nailed it.

For more photos on Facebook: John’s Autodesk University album

Customer Spotlight: Welcome to GrapeCity.

GrapeCity, an award-winning software development firm, offers a variety of services to help its clients develop new products and achieve their business goals.  GrapeCity’s client list includes global brands such as Microsoft, Accenture, Sony, Procter & Gamble, Intel, Mitsubishi, AT&T and Johnson & Johnson.  

GrapeCity specializes in two kinds of projects. First, they work with companies that have ideas for new applications, but have not had a chance to thoroughly define their vision and requirements. Through GrapeCity’s standard processes and iterative development models, GrapeCity helps its clients develop new products that meet their vision.

Second, they proudly work with what they call “burn victims”, organizations that have experienced the pain of previous project failures.   And, can’t we all relate to that?

Usually these failures result from poor processes, a lack of understanding of the vision of the product, or too much emphasis placed on technical coding.  In these cases, GrapeCity reviews the status of projects, pinpoints the causes of failure, and provides remedies to rescue projects hanging on the edge and bring them to successful completion.

With offices around the world, GrapeCity chose Contour because it needed a powerful solution that was Web-based and easily adopted by team members working in different offices, time zones and countries.

GrapeCity now uses Contour to manage the requirements of all their development projects.   With Contour, the learning curve for new users is very fast – in just a few hours of training, they are ready to go.

“In today’s day and age, teams are geographically distributed all over the world. There is a clear need for a requirements management tool that has a rich yet lightweight client that users can use to collaborate. Jama clearly gets it. They understand how a requirements tool should work, ” says, Devinder Singh Josan, Vice President of Engineering at GrapeCity.  Learn more about GrapeCity’s products & services >

Want more info on Contour?

Watch the guided tour video to see how fast, easy and effective Web-based requirements management can be.

What’s the value of requirements management in a down economy?

You open a newspaper, turn on the TV, surf the Web – the news is dominated by the economic downturn.  Is it a worldwide financial crisis? Is it a bona fide recession?  How long will it last?  Who knows.  All I know is that you can’t escape it.  In one way or another, it affects us all.

As a smart business, the bigger questions are:  How can you thrive during tough times?  Do you cut spending? Do you freeze budgets?  Do you take a wait and see approach? Isn’t that what other companies are doing?

If you listen to the financial wizard Warren Buffett or follow the contrarian path of several innovative companies, they do just the opposite.  It makes sense really.  When others are cutting back because of economic fears, they see this as an opportunity to create value by increasing investments in areas that can fuel growth and create operational efficiencies for their businesses - from technology to processes to new product development.

As Mr. Buffett likes to say, “Be fearful when others are greedy, and be greedy when others are fearful.”

So, how does this relate to our world of software product development?  More specifically, what’s the value of requirements management - or any tool for that matter - in a down economy?

In theory, an investment in a requirements management tool will create efficiencies, eliminate costly errors and ensure higher quality products.  But is that just theory, or does it hold up in the real world?  The only way to find out is to explore it for yourself.

To help though, we’ve done the math using an ROI model developed for Software Quality Engineering (Stickyminds.com) by Richard Denney, a well-respected industry expert.  Based on a team of 10 active users and 10 stakeholders, with 2,500 requirements (500 of which are implemented in the 1st year) and industry averages for a few other variables, the model shows you can achieve $300,000 in cost savings with a requirements management tool in the first year versus not using one at all; creating a 4:1 cost-to-benefit ratio.

Now, requirements management won’t make headline news, but the return on investment both immediate and long-term can be significant - if done well.  And, if you think about it, that’s exactly the kind of unsexy, penny-wise investment that a Mr. Buffet-like investor might consider.

Make the penny-wise case for RM at your company:

Download the Requirements Management ROI Analysis (.pdf)

Learn about the Contour Team Package and how to save 40%

Note: If you’d like the Excel version of the ROI model to play with the variables, just email me.

The 5 Signs You Know You Need Web-based Requirements Management.