Interview with Naresh Jain on emerging Agile trends in India
InfoQ spoke to Naresh Jain, chair of Agile India 2013 about emerging Agile trends in India. Naresh talks about Product Discovery, Lean Startup, Continuous Delivery and much more, from an Indian context.
InfoQ. What has changed from last year with regards to Agile adoption in India ?
Naresh Jain: One year is a very short time from the point of view of an organization's behavioral change. So I don't see any drastic changes, but for sure, many organizations I've visited in the last year are moving beyond Scrum and seriously exploring more XP practices, some are even experimenting with Kanban. Lean Startup has made a small impact. I hear CxOs talk about validated learning, value & growth hypothesis, and sometimes even about pivoting. They are pushing their teams towards continuous delivery and recognizing the inverted testing pyramid problem. Most large organizations have habits and they are still struggling to break the habits. Some have even realized what mess they've got themselves into, when it comes to "being agile" over "following agile."
Another interesting thing I've seen (for some time now) is that customers are demanding projects to be executed in fixed-price, fixed-time Agile fashion (whatever that means). And companies are struggling to grasp this new trend. At the surface this certainly sounds oxymoronic. Net-net, I think it's a good thing, it's getting organizations to question their existing business model.
Lastly, I see a large number of Scrum certification companies in India have gone silent and few big players like PMI, APMG, ICAgile, etc. have joined the race. It will be interesting to see how 2013 unfolds itself from the certification point of view.
InfoQ. Are more project teams in India adopting continuous delivery in their development process ? What challenges do they normally face ?
Naresh Jain: Disclaimer: I don't have enough data to give you a holistic view. But I'll answer based on several companies I've interacted over the last year.
I see a wide variety of pricing (working) models being used by the Indian software industry and their implications on Continuous Delivery practices:
- Fixed-Price (New Development, enhancements & testing projects)
- Time-and-Material (R&D, customization, implementation, pure 3rd party testing, maintenance & support projects)
- Staff-Augmentation (Random set of tasks)
- Retainer (Small enhancements and bug fixes)
- Quick n Dirty (startups building e-commerce, mobile apps, web apps, integration projects, websites, etc.)
Out of all these, only the last two models seem to be applying some continuous delivery practices.
Teams on existing projects have a hard time even thinking about continuous delivery for the following reasons:
- They don't believe in it and don't see the value in it.
- Management's perception of higher risk with continuous delivery.
- It might disrupt the existing business model for many companies.
- Management lacks the trust in the team's ability to pull this off.
- They are still very heavily structured in the waterfall model with different functional departments (Sales, Business Analysts, Architecture, UI, Development, Testing, Database Administration, Release, Operations) handling different parts of the project lifecycle.
- Even if the organizations are willing to try, the inverted testing pyramid becomes a serious bottleneck
- Most customers don't want frequent software deployments as they've been disruptive in the past.
InfoQ. Being largely an offshore and outsourcing centre is it difficult to apply the Lean Startup principles in projects done in India ?
Naresh Jain: Traditionally, the offshore teams did not have to worry about customer development or value hypothesis. They never had to worry about “Are we building the RIGHT product.” They had to focus on “Building the product RIGHT.” For decades we believed that we can separate these 2 concerns, sequentialize it and then work in isolation. .i.e. We can first spend enough time to nail down the exact requirements and then have a (cheap) team offshore build it exactly according to specification.
Alas, to everyone’s surprise, we realized how fundamentally flawed this model is. These 2 concerns are a very integral part of defining and building a successful product.
Lean startup principles have really hit the nail on the head. Today we understand the value of really short Build-Measure-Learn cycles (or as Kent Beck puts it Learn-Measure-Build cycle), the importance of pivoting and continuous delivery. I’ve found the amount of collaboration, the depth of buy-in and the efficiency of the teams extremely high when working in this model.
Even Product Discovery (read as User Centered Design) practices like Personas, User Goals, Story Maps, Low-fi prototypes, etc. have been proven extremely valuable to offshore teams.
So yes, it is certainly difficult to apply Lean Startup principles but extremely important esp. when working in an offshore (outsourced or not) environment.
IMHO the biggest roadblock for teams embracing these principles and practices is simply lack of awareness and exposure. I’ve worked with organizations in implementing these practices and they tell me that they’ve tremendously benefited from these principles and practices. Some even claim that they’ve benefited more from these practices than Scrum and Agile.
InfoQ. Has there been an increase in the adoption of Kanban in the last few years ? How does it scale with distributed or offshore teams ?
Naresh Jain: For donkey’s years, Indian companies have been using some perverted form of Kanban. I remember leading a maintenance project back in 2003, where we started with XP practices and after several “inspect & adapt” cycles, our process evolved to a simple kanban style of working. We limited our work in progress, we focused on one piece flow, we kept everything visible, we automated the heck out of things to measure and improve the flow, we had the concept of class-of-services and so on.
I would like to believe this has been the nature of many offshore projects, esp. maintenance, support and implementation projects. They might not have the same rigour and thought process gone behind the evolution of their process, but they do end up with something on those lines just because of the nature of work.
Having said that, I clearly know working collaboratively in short cycles has been missing. Applying these principles to new product development or major enhancement projects has been missing. In the last two years, I’ve seen companies streamline the process, form cross-functional teams, focus on bottom-up process improvements and so on. Agile, Scrum in particular, should get the credit for having brought about this change.
Now teams & companies feel that Kanban is the next logic step in their process evolution.
Also in the past, there has been very little branding and awareness regarding Kaban in India. But many local tool companies are marketing their tools with a strong emphasis on Kanban support. This is slowly increasing the awareness.
Regarding scaling Kanban with distributed & offshore teams, I don’t see any special issues. A large number of tools make it relatively easy to bridge the gap. In addition, there has been some amount of experience working in this fashion, which does help.
- Last year the conference program was built 100% via open submission system. In spite of a meticulous, 4-month long, open review system, quality of some of the sessions was poor. Also because of the open submission system, we were not able to attract many big names. This time, we’ve over 25 Agile/Lean luminaries from around the world; each presenting 2-3 of their greatest hits. The conference is paying and flying each of those speakers down. To encourage other practitioners to present their first-hand experience, we’ve also accepted about 20% of the sessions via an open submission system. This has ensured a very high quality program. We believe, it would be far better than any other Agile conference world-wide.
- Based on the participant’s feedback, we learned that 3-day conference was a bit too long for the participants. They found the conference scope to be very broad. So we’ve split the conference into 2 parts. First 2-days Management & Leadership focused and the last 2-days is focused on technology/delivery. This would lead to a more focused audience and will appeal to a large group of diverse practitioners.
- Also this year we’ll co-locate GuruPLoP, the first Indian conference on Patterns & Patterns languages of Program, taking place on March 3rd and 4th, 2013.
- There was a huge demand for more focused hands-on workshops. So we are hosting 14 hands-on workshops, which is separate from the conference itself.
- Last time we had 7 parallel tracks distributed on different floors and the participants were lost. This time we’ve 4 parallel tracks and an open space. All rooms are on the same floor in a central area. The room sizes are much bigger compared to last conference, ensuring comfortable seating capacity.
- Each day we expect 650-700 participants. Plus another 100 participants will attend the paid workshops. So over the 4 days we should see 1500 unique participants. This is more than twice compared to the 2012 conference.
- Unlike last conference, this time the sponsor area is right in center of all the halls. Food would be served in the same place and an open space is planned in the same area. This will ensure a significantly higher opportunity for the sponsor to interact with the conference participants.
- Each part of the conference has a dinner reception allowing participants to network with other participants and speakers.
- Except the room layouts, most participants loved the conference venue last time. For 2013, we are hosting the conference in Hotel Sheraton, a brand new luxury 5 Star hotel. A much better venue compared to the last one.
Overall we want the participants to have the best conference experience they’ll remember for years to come.
Tom Gilb & Kai Gilb Jan 26, 2015