Six Cups Of Tea

By Nirat Bhatnagar, WASH for India

“The first time you share tea with a Balti (Baltistan people), you are a stranger. The second time you share a cup of tea, you are an honored guest. The third time you share a cup of tea, you become family”. – David Oliver Relin

This quote from the book Three Cups Of Tea by the now discredited writer Greg Mortensen rang true as I sat down to write about my experience on the recent field trip to Davengere with Vijay, Vasu and Krishnan. Over the last few weeks, I’ve been working with Arghyam to observe and document different aspects of this campaign to generate demands for toilets in rural Karnataka and recommend strategies for scalability and advocacy.

Last week, over 2 days, I had gone to the field villages three times. And because the cups of tea were so small, ended up having six (or seven) rather than three 🙂 While I did not end up being part of the families of Davengere, I came back with a far more detailed understanding of what Arghyam and the district government and the various partners are trying to do through this project.

Let me separate my post into five overlapping buckets looking at different aspects of this campaign the way I’ve been able to observe them so far.

A) The Context

The overall context is that Davengere district needs to build 140,000 more toilets and have them used consistently to become open defecation free. The district population is about 220,000 households (or a million people) spread over 1,000 villages of differing sizes. According to the discussion with the District Support Unit for Sanitation Head, Mr. Basavarajjappa, they’ve managed to construct some 80,000 toilets over the last 20 odd years and want to construct the remaining 140,000 in the next two years! If I were an outsider, this would seem strange and perhaps a classic “supply side infrastructural push”. My fear would immediately be whether those toilets will be a) Linked to user demand b) Of high quality and c) Thus used.

In this context, what Arghyam is doing – helping generate user-demand for the toilets that district wants constructed and then later follow it up with behavioral nudges – seems perfectly placed in these larger goals of the district. The government also seems to feel the need for such inputs and seem to recognize that the biggest bottleneck to better sanitation seem to be “people’s mindsets”. Mr. Basavajrajjappa was honest enough to admit that he did not have clean scalable answers to this. He, like many others in the government, feels CLTS is quite effective but requires extremely high resource intensiveness.

In terms of the user-side of the context, my interactions were rather superficial given that I did not understand the local language and very few of the people (less than 2-3%) spoke Hindi or English.

However, the one thing that did stand out as it often does for me, is how people’s mental models do not include a toilet to be an inherent part of their house. We saw several houses that seem to have been recently constructed for INR 250,000 or thereabouts. But people had not spent an additional 4% to construct toilets. Cognitive sciences talk about this bias called “anchoring”: people should have been framing the construction of a toilet when they build a house as just 4% extra and not as an additional INR 10,000 – which independently is a large amount of money. But there’s wasn’t any evidence of this having happened even in houses that were less than 5-6 years old by the looks of it. Clearly, some opportunities to anchor the toilet as a natural part of the house had been missed in earlier campaigns. But this is definitely a communication lever to keep in mind in future campaigns.


Unfinished toilet in Davengere for which the NBA subsidy seems to have been paid.

There were some very interesting examples where toilets had been left unfinished and were just lacking minor elements such as a door. But due to current guidelines of the Nirmal Bharat Abhiyan (NBA), people need to finish the entire toilet to get the payment from NBA. However, some of these toilets at least had painted numbers on them, which seemed to give each toilet a unique ID and also seemed to suggest that the subsidy money had been disbursed already. The explanation offered was that the local contractors were “seeking to make profits by doing this” and these toilets were not the result of end user demand.

B) The design of the IEC campaign

At its core, the Arghyam campaign is about creating “buzz” with a toilet event in the village to which people are personally invited; an IEC van making announcements, the Swachhata Dhoot going door to door giving invitations and urging people to come, and school children carrying out a rally, urging people to attend an event on the evening of the second day. The theme of the entire communication is that of a “responsible man” constructing toilets for the women in his house and fulfilling his responsibility.


The central “stage” in a village where the event takes place.

The entire thing unfolds over two days for each village and involves about 3 hours of standardized activity (the van, the rally, and the evening event) and as much as 7-8 hours of customized activity (the Swachhata Dhoots going door to door). To serve as a reminder to people who are eligible to claim the toilet subsidy and hence should participate actively in the campaign, there are also walls in the village that are painted with their names. There are two calls to action: a paper slip that people need to fill to express interest in constructing a toilet and second, a formal application form that needs to be submitted to the government. The campaign facilitates the collection of these forms as well.


The IEC van and the generator that powers it waiting pre-event.

If one analyzes the inputs (costs, management time, & design time) that have gone into the campaign, the centerpiece of the campaign is the evening event which comprises of 3 films that are shown to attendees, a street play that seeks to evoke empathy in the husbands / fathers / sons for the “women in their life” who live without toilets, and finally the collection of demand slips and the application forms that formally induct people in the toilet construction and subsidy process. This evening event at all the target villages takes place at a central location with a makeshift ”stage” where people are supposed to gather.

I saw a large chunk of these IEC components for the first time on the trip and felt that they were well designed in general. The play was a hit amongst the participants and I could see all – women, men, & children – smiling and nodding while watching it.

Of course, there are tweaks that can be made here and there.

  • For instance, it might be the case that the current set of films is a little too long. Certain aspects of it could be reduced for instance using the 5-minute animation film instead of the ten-minute version.
  • It is important to start and finish with a “bang”. In the first village we went to, the program started with a 10 minute long film that wasn’t part of the pack. I could see people leaving midway while this was going on. It is very important that if extra content is being added to the program, it should come AFTER the planned IEC program and that it be approved with respect to content & theme by someone at Arghyam or Centre of Gravity.
  • There is some merit in having form collection / slip collection immediately after the CEO film so that it becomes the call to action. In the second village that I visited, this worked really well.

C) The implementation of the IEC campaign

Arghyam is program managing the implementation of this campaign across 9 teams who will cover 100 villages in the district over two rounds of engagement, lasting a few weeks in all. While the design of the first round is described above, the second round is in the process of being designed and will start in a few days from now. It is definitely a challenge to adequately staff, train, and execute a campaign with 9 different teams in 2-3 months. And kudos to all involved for doing a good job.

I’m going to focus my observations on the participation of people in the evening IEC event and their engagement with the campaign. The background is that in both the villages that I went to, about 50% of the attendees for the evening event were children and the remaining 50% were adults. Only about 10% of the adults were women. It was not very clear how many of the participant households had or did not have toilets.

I feel there are five aspects to optimal participation by people in this campaign

How people get invited

  • Currently, invites are being handed out to households the previous day, but no “RSVP” occurs. Could the Swachhata Dhoot go around collecting thumb-prints or signatures from people confirming their participation, asking them to respond very honestly about whether they will come or not?
  • Perhaps the van can keep distributing the flyers as it goes around the village in the morning of the day of the event, and also perhaps the school children can be given flyers when the school rally happens.

Who comes to the event

  • We have seen not more than 60-70 adults at both the IEC events. The number of children is about 80 – but this is largely irrelevant. These 60-70 adults comprised of about 60 males and 10 females. Assuming that out of these 60 adults, about 66% are from households who don’t have a toilet, we’re getting representation from about 45 households without toilets. This is about 25% of the total households without toilets in a village. This number needs to increase through the invitation process.

Participation by women

  • There are hardly any women coming to the event. From what we’ve observed and discussed with the DSU, this is because women and even the “decent men” don’t want to come out at night to a location where there might be “drunks etc”. Could the invite be tweaked to get “husband-wife” pairs to attend?

Optimal “user-experience” at the evening IEC event

  • It is very important to place the event so as to maximize attendance – bringing it closer to the part of the village where the proportion of households without toilets is higher.
  • The location should be chosen so that it’s not continuously disrupted by traffic. In the first village, there was a flow of tractors and trucks regularly leading to severe distractions.
  • There should be adequate seating for at least 200 people. Otherwise, the men tend to form a ring around the tarp that is placed (which seats mostly children). This ring of men leads to a) Women not coming close and b) People floating in and out. Sitting and seeing this is a more comfortable experience and also has a higher “cognitive commitment”.
  • The sound: The speaker on top of the van is a conical directional speaker. I’ve observed that the sound levels drop dramatically as one moves orthogonally to the speaker. Yet, in both the instances, the van was kept to one side facing straight. With a little tilting of the van, the sound experience will improve even with the same decibel levels.


  • It will be very important to ensure that conversions happen from exposure to the campaign, to forms being filled and toilets being constructed. So there is a funnel optimization problem here: i.e. an increased collection of slips -> to forms submitted -> to pits dug ratio. It might be worthwhile to announce an inter-Swachhata Dhoot contest to maximize the conversion numbers. The total “prize money” could be somewhere in the order of INR 50,000 (broken in 4-5 prizes) – enough to incentivize better performance.

Bridges to the next stage

  • It’s going to be very important to ensure that subsidy payments reach on time and that the DSU / CEO are getting a daily report (enabled by the software that Arghyam is in the process of designing or selecting) around key metrics of the project.
  • A helpline / outgoing calling process could be put in place that periodically calls the households who have submitted slips to ensure that they are carrying through with the process.

D) Next steps

I’m looking forward very keenly to more data on the application form numbers from the first round of this campaign, and also the follow-up phase a few weeks later. Similarly, I’m really looking forward to the kick-start of the Final Mile campaign that begins very soon.

E) Meta Observations 

Most people in the WASH sector have two implicit mental models that I feel need to be questioned a lot more. The first mental model is that most urban slums don’t have space and need community toilets. The second mental model and the focus of my observation is that “villages need household toilets”. This mental model is enshrined in the subsidy policy of the government and the focus of all campaigns I’ve seen so far to generate demand in rural India.

However, field visits to Davengere make me believe that this is a flawed mental model. Why? Let’s unpack this mental model into its three underlying assumptions

  • All rural households have space to build a toilet as an augmented feature to their existing house: This is simply not an architectural possibility in many cases. I saw on this trip that several houses, especially the ones more proximate to the village center are packed edge to edge like townhouses with a 1-2 meter empty stretch in front where the household carries out activities such as washing dishes / clothes / lazing in the sun. Building a toilet without major architectural changes to the existing house is simply not possible. In some proportion of the houses, they may have space in the backyard, which is where the toilets can get constructed.
  • In case, they don’t have space, they’d be willing and able to build a toilet some distance from the house: There are some cases where people build houses on empty land 10-20 meters away from the house (almost like an outhouse of the earlier days) but this is not universal at all and is going to become increasingly as rare as the per capita land availability shrinks even in rural India.
  • A toilet is a toilet is a toilet: People believe that the probability of a household using a toilet is going to be the same no matter where the toilet is in their household. However, I don’t think this is true – people tend not to use toilets that are placed closed to the kitchen (for obvious reasons) and those placed in the front (for aesthetic reasons). Would we want a toilet to be the first thing that greets visitors to our house? Why do we assume that villagers would? Thus, one explanation of unused toilets in India could be that given the overly “supply side” process of construction, people didn’t really have a say in where the toilet got constructed and now don’t use it.

In any case, even if a village has 15-20% of such houses that fall into the above categories, alternate provisions should be made to the policy. I would love to explore more whether this is one of the reasons why the form collection rates have been slightly lower than we were all hoping for.

PS: I’ve received no satisfactory answer yet as to why the cups of tea are so small in rural Karnataka. Scarcity of water, tea, or money doesn’t seem to be adequate answers for this behavior. And there are several secondary questions: Given that tea drinking is a social event, do smaller cups mean shorter social interactions with accompanying effects on the social fabric, or do people stretch out their small cup of tea to last the same time as someone drinking larger cups of tea somewhere else? What does this kind of “stretching out resources to last longer” behavior for small ticket items mean for purchase habits of people for large ticket items such as toilets? But these are all musings for another blog post 🙂


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s