Lean ICT4Law: Applying the principles of the lean startup entrepreneurship model to legal A2J projects

One of my UC Berkeley iSchool ICT4D students had written this piece a few years ago, but I find it a good reminder of the business thinking we can apply to implementing technology projects for access to justice.

By Seema Puthyapurayil (@seemahari)

Last week, I stumbled upon an inspiring article about the Indian sanitary pad revolutionary: Arunachalam Muruganantham, who has developed a simple machine that can make low-cost sanitary pads. Arunachalam’s invention was aimed to help the 88% of Indian women, who were using dirty cloth, leaves and even ash during their menstrual cycles. Not only do these machines have the potential to improve women’s health and hygiene, but they can also empower women in small communities to become entrepreneurs and create and distribute their own brand of sanitary napkins. If you google his name you can find a ton of articles about this man and his great story (there is even a documentary film made about him!), so my focus in this blog is to highlight what I think is a big factor in Arunachalam’s success: applying the lean startup principles of customer validation and minimum viable product (MVP) to a development project.


The Lean Startup

I took a class at the Haas School of Business last year called the Lean Launchpad which helps students convert their nascent project ideas into full fledged startups through the lean startup model. The lean startup provides a scientific approach to creating and managing startups and getting a desired product to customers' hands faster. The method teaches you how to drive a startup-how to steer, when to turn, and when to persevere-and grow a business with maximum acceleration. The method encourages startups to embrace the just-do-it approach by eliminating all processes and focussing on developing a product that users need and will use. It forces entrepreneurs to build the minimum viable product or MVP which has the minimum number of features required for the product to be deployed and tested with users. The MVP doesn’t even have to be the product itself, it can be a prototype or an alternative that still allows users to understand the core value proposition. For example, the MVP for a presentation can just be speaker with a whiteboard. The model encourages entrepreneurs to think about the question “Should this product be built?” instead of focusing on the age-old “How to build this product?” question which makes entrepreneurs hung up on the solution without thinking about whether users even need such a product at all. The build-measure-learn model brings valuable feedback into the new product development process and helps entrepreneurs make mistakes early, through which they can iterate, refine or even pivot from the initial product idea.

Arunachalam’s Lean Development Journey

Without even having received a formal education, I think Arunachalam implemented what most entrepreneurs are now learning in business school: using the lean startup method for developing new products. The key to Arunachalam’s success apart from his grit, passion and unwavering dedication, I believe, was his manner of building an MVP, testing it out with users and iteratively refining the product. He recognized the problem when he learned about his wife using really dirty cloth during her period. He built a small prototype of the sanitary napkin with a wad of cotton, asked his wife to use it and demanded feedback. When she said that he would have to wait for a month to get her period and then get feedback, he immediately approached his sisters and medical students. When even that failed, he decided to try it himself by carrying around a football bladder filled with goat blood with holes punched into it! User empathy in design at its finest! He started out with the most basic version, an MVP, of the sanitary pad which is a wad of cotton and aggressively tested it to get feedback. Even though Arunachalam’s goal was to not make profit, he makes sure that there is actual need for his product and that it will be easy to use for his target audience. He tests this out by deploying his machines in the most rural, illiterate female community in the state of Bihar in India, with the rationale that if it can work in Bihar, it could work anywhere. His pad making machines look skeletal and rudimentary but he intentionally keeps the machines simple, so that they can be easily operated and maintained by women who have little to no knowledge about operating machines.

Another big contributor to his success, is his personality trait of being fearless, a trait required by entrepreneurs who want to follow the lean startup model. Getting out of your comfort zone and testing MVPs with users is not easy and often may lead to rejection of the initial product idea. Arunachalam was shunned by society and abandoned by his family, but he continued on with his mission and kept asking his target users for feedback embrace the uncertainty. He believes that being illiterate also forces him to stay curious and ask many questions "If you act like an illiterate man, your learning will never stop... Being uneducated, you have no fear of the future." which also ties back to the lean startup concept of asking 5 Why’s to every problem to get to the root of the problem. The MVP model for Arunachalam was that it forced him to make mistakes early and allow validation of the actual need before building anything.

LEAN ICTD



I believe that the ICT4D world could hugely benefit from the lean startup methods of starting with the questions of “Should we build it?” and not “How do we build it ?”. Even though most ICTD projects invest heavily in ethnographic research and user needs assessment, which sometimes may only focus on “How to build the product right?” or “How do we build the right product?”, still starting with the assumption that an intervention is going to be built. The principles of continuous customer validation through the use of MVPs can allow practitioners to test out whether an intervention is needed at all in the first place. This can help avoid the trend of ICTD projects failing to stay in use after the pilot stage and ensure that risks to adoption and use by the user groups are identified very early in the development stages, saving considerable funds from being invested in the wrong initiatives.

The lean startup model also encourages entrepreneurs to start with a business model canvas and refine it continuously as the project progresses. The business model canvas is considered as a live document that needs to be updated every time the team discovers new insights or when round of user testing with the MVP is done. This ensures that the entrepreneurs have sustainability and profitability on their mind throughout the new product development process. Even though the goal for ICTD project is not to generate revenue, a live business model canvas ensures that the practitioners are forced to think about key partners, customer segments, cost structure, channels to reach the target user groups etc through the product development phase.

7 comments:

  1. Dear Seema,

    Thanks for your great, inspirational blog post! Arunachalam is a hero and 100% committed! His own testing really impressed me.
    I really love the idea of a Minimum Viable Product and focusing on whether a project should be built (by deploying it to users at an early stage) rather than obsessing over how to build it from the very beginning. It makes total sense and I'm kind of surprised to hear that this is not the way start-ups usually operate.
    One quick question about the model: What do you mean when you say that the lean startup approach is scientific? It's a model, but how is it scientific?

    ReplyDelete
    Replies
    1. Hi Christina, Thanks for your comments. The lean startup website calls the method scientific: http://theleanstartup.com/principles and Eric Ries describes the science behind it in his book, but I haven't read it.

      Delete
  2. This comment has been removed by the author.

    ReplyDelete
  3. Dear Seema,

    I am really really really impressed by your blog post, I mean really. I didn't know the Arunachalam's so thank you for sharing. Like Christina, I am very impressed about his own testing.

    I totally agree about your idea of using the lean startup methods for ICT4D projects (but also for projects in developed countries too!).

    About that you should look at the TED video "Build a tower, build a team" (http://www.ted.com/talks/tom_wujec_build_a_tower). It is about the "marshmallow problem. Its goal is simple: teams of four have to build the tallest structure with sticks of spaghetti and one marshmallow. The marshmallow has to be on top. It is a race of eighteen minutes.

    The reason why I am speaking about this example is that some very serious researches have been made about it and the first advice to build the tallest tower is to use an iterative process. If you try directly to build a very big tower, you have a lot of chance that the structure collapse at the end when you put the marshmallow at the top. On the contrary the study shows that if you begin with a little structure, and make an iterative process to make this first structure bigger and bigger (with the marshmallow at the top) you have much more chance to build a very tall tower : prototype matter! (just like in the lean startup methods you described).

    ReplyDelete
  4. I'm glad you liked it Antonin! Loved the TED talk. Thanks for sharing that. Great comparison to the building the tower. I completely agree!

    ReplyDelete
  5. Great post Seema! You compressed entire lean launchpad course in one blog post! Kudos on that.
    You might also want to look at Tapan Parikh paper on user centered design approaches:
    http://people.ischool.berkeley.edu/~parikh/projects/CAM/papers/parikh-thesis.pdf

    ReplyDelete
  6. Kudos for compressing entire lean launchpad course in one blog post!

    You might want to look at Tapan Parikh ' s paper on user centered design approaches :
    http://people.ischool.berkeley.edu/~parikh/projects/CAM/papers/parikh-thesis.pdf

    ReplyDelete