Why p-commerce is the new e-commerce in Bangladesh

The Innovation Dividend Podcast, EP 3

The Innovation Dividend explores how innovation in society and government are paying off. Over the coming weeks, we’ll be exploring how ‘policy frontliners’ are innovating in real time in the COVID-19 pandemic and asking which of these changes and “raw learnings” might become part of our new normal. You can see the backstory by Kal Joffres here, Podcast EP 1 and Podcast EP2.

In this third episode, we interview Anir Chowdhury, who is at the frontlines of Bangladesh’s COVID-19 response. He shares perspective on how you get essential goods to people through e-commerce when the majority of your country isn’t online: introducing p-commerce. Also: he updates us on the effort to build a social safety net for 5 million people and how getting intra-government buy-in for projects has changed in the age of coronavirus. This interview with Anir Chowdhury was recorded on April 24, 2020.

• You can subscribe to the podcast on Spotify or Apple.

[00:00:00] Anir Chowdhury: I have seen cabinet members, permanent secretaries, and high level officials being very open to trying out new ideas. There is a 10-fold jump in terms of attitudinal shift, that “let’s try this out.” We are in the middle of this massive, unprecedented crisis. So, if you try out something new, I mean, it can only help us.

[00:00:21] Kal Joffres: Hi and welcome to The Innovation Dividend, the podcast that explores how innovation in society and government are paying off. I’m your host Kal Joffres and over the next few weeks we’ll be exploring how the COVID-19 pandemic is unleashing some interesting and unexpected sources of innovation.

[00:00:36] We’ve been following Anir Chowdhury. He’s at the front lines of the COVID-19 response with the government of Bangladesh, and for the last 13 years he’s been leading the formation of an innovation ecosystem in Bangladesh, institutional reform, and a service innovation fund. He’s also a member of the Prime Minister’s National Digital Task Force.

[00:00:55] Anir, thanks for spending the time with us.

[00:00:57] Anir Chowdhury: Thank you, Kal.

[00:00:58] Kal Joffres: Last week you were starting the second week of a two-week initiative to build a new social safety net for 5 million people in Bangladesh. Can you tell us a little bit about what’s happened since then.

[00:01:09] Anir Chowdhury: Yes, so we underestimated. instead of 5 million, now we are now looking at maybe 10 million or so. We also underestimated how many times we would have to do this . We would reach out to these 5 million people. Now we realize that not only the number is going to be larger, like more like 10 million, but we may have to do this multiple times.

[00:01:28] So we’re now planning to do this at least three times. So once this month, and then May and June. And then we’ll have to replan if it needs to continue further. So it’s a drastic shift in terms of size, in terms of repetition, in terms of perhaps more use of technology. Previously we thought that technology would be used to do some targeting.

[00:01:50] We were looking at multiple options, but now we need, we know that we have to use the technology on a sustained basis. So we may need a bit more robust architecture.

[00:02:00] Kal Joffres: And how did you come to realize that you needed to reach out to more people.

[00:02:04] Anir Chowdhury: We looked at the previous safety net databases. I don’t think we looked at all the people who would become jobless. We did a better calculation in terms of what are the occupation categories that will no longer have work. Right now, they’d live on daily income. We also are in a quandary in terms of how many people are going into the fields to do harvesting.

[00:02:31] Targeting has become more difficult. I think I may have mentioned this last week as well, that about 12 million people moved, from the cities to the village homes. So that also becomes a very challenging task in terms of where you’re targeting whom. So they used to have a different address.

[00:02:51] Now they live in rural areas, and a lot of the neighbors don’t know them. Sometimes a political bias comes in, so you may be missing out on people. So now we’ve taken another week to look at the numbers more closely. So targeting is still a challenge.

[00:03:09] I’ll maybe talk about one or two innovative initiatives that I have seen in the last week of how the district administrators are targeting the poor.

[00:03:19] So one district administrator decided to do a full sale targeting. She decided that looking at different databases which are disjointed, not connected, huge gaps in them. She wanted to do the whole survey of 325,000 households. In her constituents in her district. So how do you do that in a week?

[00:03:42] She took an innovative approach of using old technology, paper and new technology, computers and internet. So what she decided was she printed 325,000 copies of household information, collection forms, and she distributed them to all the households. The instruction was that each of the households would have to find somebody who would be able to fill out this form.

[00:04:08] And they were able to do that. So it was on paper writing by maybe a, 12-year-old or maybe or by a 15-year-old, who goes to school so the parents cannot write or read, but the child fortunately can read and write, so they fill out this form. She’s already collected about 250,000 of these forms into a central location.

[00:04:28] And then again, she distributes these forms to about a hundred volunteers, who have come forward with internet connection. So these are high school goers and college goers and university goers who are sitting at home, waiting for something interesting to do.

[00:04:45] These are more connected, maybe middle class, upper middle class kids who have now emerged as the new data entry personnel. So she obviously doesn’t have the money to employ anybody. So she printed those out. That cost money, but after that, it’s all free of charge. So a hundred volunteers have come forward and they’re feverishly entering this data into a central database.

[00:05:10] So now, one question, obviously that comes to mind is whether this is valid data or not. That’s something that we have obviously have to check. So we need to check this against the household database. The household survey that we did last year. Unfortunately, that data is still on paper.

[00:05:26] It has not been processed into a fully digitized database.

[00:05:31] Kal Joffres: No. To have financial aid is important, but how are people getting the necessities right now?

[00:05:36] Anir Chowdhury: That’s an interesting space right now. Some of the shops are still open, so some of the grocery stores are open. People make weekly visits to them. Some of the super shops, as we call them, department stores, they are open and some of the bazaars in rural areas, the open markets, they’re still open.

[00:05:54] So for the last two decades, almost any e-commerce in the last five or so years, it’s been growing fairly rapidly, but still, it covers a small segment of the population. It’s a largely urban phenomenon, or at best a peri-urban phenomenon.

[00:06:11]90% of the e-commerce transactions were for luxury items, so things that you don’t need in your daily life. So people would buy electronics, people would buy maybe an expensive handset phone, luxury clothes, that kind of thing.

[00:06:27]So what has happened in the last, I would say, last month or so is that we’ve seen two huge trends. Business models are just coming out of woodwork, new business models. So luxury item sellers, the eCommerce companies, have now become daily necessity sellers.

[00:06:47] The second thing is that people now want to make electronic payments. So, we saw about only 5% of the payments before being made through electronic transactions, 5% mobile financial wallet transactions, and 95% cash on delivery. Now that has changed. Now we’re seeing about 50%. Of the people making electronic transactions because again, this is happening very, very quickly.

[00:07:15] We’re talking about weeks to this trend, this trend as it’s emerging. So now companies are overnight changing their business models without hiring expensive consultants and doing competitive market analysis. And this shift towards mobile financial wallets, 5% to 50% in a matter of weeks, is also dramatic, obviously 10 times.

[00:07:38] That 5% has remained 5% for many, many years. So what is also happening here, I think, is that many of the consumers are looking at e-commerce as a way to do a trustworthy transaction.

[00:07:53] So e-commerce, as you know, is built on trust. So when you order something. You’re not seeing the seller. The goods will arrive after a certain time and that you have to make the payment early and sometimes you don’t want to. You want the goods to arrive first and then pay them, and that’s why the 95% cash on delivery. Now as we are moving towards 50% cash on delivery, less than 50% cash on delivery.

[00:08:18] And possibly this number will actually become less and less in the coming days and weeks. It basically indicates that people are trusting e-commerce more, that they are trusting the goods that will arrive for them.

[00:08:29] Kal Joffres: What happens to the people who don’t have advanced phones or don’t have internet connectivity.

[00:08:33] Anir Chowdhury: Well, Kal, given that about 60% of Bangladesh population who carry a cell phone only carry a feature phone, not a smartphone. We really have to think about what they will do with that phone as far as commerce is concerned. So how they will order food, or medicines, or daily necessities. So obviously, since they don’t have smartphone, they don’t use internet. If they don’t have internet, they cannot get onto a traditional, what we call e-commerce sites.

[00:09:04] So we have a new concept, called p-commerce, phone commerce, using old technology. So think about at a time when we used to be able to order over the phone in many countries, not in developing countries and in developed countries. So let’s say 20 years ago. So that was how the early version of eCommerce, before Amazon or eBay used to work. So we are looking at similar things in Bangladesh. So a person with a feature phone calls into the 333, the 333 that’s been repurposed many times. And there is a new branch of 333.

[00:09:38] So they dial five after the dial 333. And five takes them into an e-commerce branch and all that does is it asks a couple of questions in an IVR and it captures the phone number of the person. So the phone number also, has the relative location associated with the call.

[00:09:57] Somebody picks up that number. A volunteer. We have about a hundred or so volunteers who work in a call center, distributed, not in a specific location, but distributed. So a person in that call center picks up the number and then calls the caller, and finds out the specific needs of that person in terms of daily necessities, medicine, food, and so on and so forth.

[00:10:22] And then that call, that list is then forwarded to another volunteer. We have about 4,000 such volunteers across the country who do food deliveries. So this second volunteer is in the location of the caller and can pick up the food, daily necessities from a supplier. It could be a super shop, one of those Western-type supermarkets, or it could be one of those very small Eastern-type grocery stores, and then delivers that.

[00:10:51] So the combination of a call from a feature phone picked up by a call center operator who’s a volunteer. Forward it to another volunteer who is basically a procurer or deliver of that list of food or medicines. So that combination is what makes p-commerce work.

[00:11:11] It’s a series of dots that need to be connected, and that’s what we’re doing. It’s a pilot right now we’re seeing whether this works. We have delivered about 2,000 orders in the last, one week . It seems like both the backend is able to take the load. The front end is also a combination of all these technologies and human agents are making this work.

[00:11:33] Kal Joffres: Could you tell me a little bit about how these P commerce networks came about? It sounds like there are several different groups of people and systems that are connecting together.

[00:11:41] Anir Chowdhury: For the last, six weeks or so. We have been working with a lot of volunteers across the country. Many of them have come from the digital centers that we run, about 5,000 digital centers that we run across the country. Each digital center has about two entrepreneurs, digital entrepreneurs who sell different types of services to citizens, about five to six million citizens every month.

[00:12:04] So many of the volunteers have come from these digital center entrepreneurs because they don’t have the centers running right now. So they are looking for ways to make an income or ways to contribute to society. But then there are other volunteers from Boy Scouts, from local volunteer organizations who are trying to make a meaningful impact in their society in this time of great need.

[00:12:27] It’s scouting across the entire country, keeping our radars on, using the volunteer network we have already that brings in other people into this. So that’s what has created this list of about 4,000 volunteers as an initial count of people.

[00:12:43] Kal Joffres: What are you learning from the operations of that pilot? Are there things that you pivoted early on as you were running it, other lessons that might be useful for other countries that are thinking about how to do something like this?

[00:12:55] Anir Chowdhury: We’re learning quite a few things. One is obviously that it’s possible to do these kinds of innovations, in a very rapid manner when your back’s against the wall. But it also requires many actors and many stakeholders to come together and offer their brain power towards the same solution. So what I have seen in the last over a decade of working with the government and with many private sector providers is that often innovation is triggered in the mind of one person, or maybe a small group.

[00:13:28] But the implementation of the innovation on large scale really requires huge amount of teamwork, collaboration, and public private partnership. And that’s what we are seeing in reality today.

[00:13:41] So the actual platform is the platform of the government for e-commerce that aggregates about 16 different e-commerce platforms.

[00:13:49] In the last one month, that platform has grown to about 112 partners. So again, we’ve worked on it for two years, 16 partners. In one month, it grew to be about 112 partners. And the combined intelligence combined efforts, connections in terms of, who can deliver what in what locality, what kind of product, is all coming together into one, huge platform right now.

[00:14:15] So that, that is one big lesson that public private partnership, collective intelligence of operators from different areas have to be harnessed to make an innovation go to scale. Otherwise it remains the lab and it remains a small impact and it really does not help tens of thousands or maybe tens of millions of people.

[00:14:35] A second thing is that business models need to be adaptive. So the ones that we have seen are adapting. The e-commerce companies are seeing, the ones that are able to adapt are surviving. And the ones that are not able to adapt that look to probably disappear within months because they don’t have deep pockets.

[00:14:51] The operators who are delivering goods, food, medicines, and other necessities, many of them are operating at maybe 25% or 30% the number of people they used to have. So we are, we are seeing a dramatic rise in the number of orders, but a dramatic reduction in the number of people. So this is a huge challenge right now.

[00:15:11] And, what we did was we had the Ministry of Commerce recognize e-commerce as an emergency service. So this was very helpful because we did the emergency service, the people who are delivering the goods are not able to move around.

[00:15:26] The police will stop them, the army who are, ensuring social distancing. They’re stopping them and they’re making them go back or, finding them and all of that. So what we did was we used an app to register the verified delivery agents. So this app now actually takes in information from the e-commerce companies and the logistics companies, and then makes them verified agents who can move around and they can be, again, recognized by the police and the army personnel and they can move about. So that one app now has about 11,000 verified agents who can move around the country to deliver this. So that again is a law and order issue that we have solved with technology.

[00:16:10] One thing that we’ve been looking at, which we have not been able to be very successful with, is large movement of food across the country.

[00:16:19]So e-commerce primarily is a consumer phenomena. So I am interested in ordering something. It may be a large grocery list, but still it won’t be more than maybe a 20 kg or 40 kg worth of goods. But what about hundreds of tons or thousands of tons of food that need to be moved around the country?

[00:16:40] So whether we could use any of these technologies, we’ve been exploring that. We’ve had joint meetings with the Minister of Agriculture who was very worried about perishables in different parts of the country, which are not moving to the areas where the consumers live, primarily the urban areas.

[00:16:58]The Ministry of Food which is worried about making sure that there isn’t a food scarcity across the country. So both ministries — Ministry of Agriculture which produces, and the Ministry of Food which ensures the movement of goods and the food happens are both worried.

[00:17:14] And we’ve been exploring whether this e-commerce platform that combines many, many platforms across the country, whether these can be used for large scale food movements, we have not found a good solution there. So anybody listening to this, if you have solutions for large scale food distribution in the country, so coming up with a new supply chain where many of the intermediaries are completely missing from action.

[00:17:38] So they’re sitting at home, they’re not able to move, they’re scared. They don’t have authorization to do any of these things. They don’t have the right safety measures. I would, I would welcome these thoughts and ideas from other countries, which are using maybe technology or many, any other innovations to move large amounts of food across the country.

[00:17:58] Kal Joffres: Suppose I’m a junior person in the civil service today, and I have an idea around how to address one of these critical challenges, has the situation democratized the ability to get ideas out in government ?

[00:18:09] Anir Chowdhury: This is something that we have been exploring for the last few years.

[00:18:13] We’re now looking at people coming to us with ideas from within civil service and from also outside of civil service.

[00:18:20] So just yesterday, we were making a presentation in front of the ICT Minister of how the new subsidy distribution system would look like, which was developed within about three days. So we’re looking at the prototype for that which brings in two ministries, again unprecedented, as it brings in the Food and the Agriculture Ministries together to deliver these subsidies across the entire country.

[00:18:43] Actually three ministries, the Ministry of Relief was also part of this opening. So two ministries were connected to this one platform. Now, during the meeting, we had about 90 people in the meeting from many different ministries and from the field administration. So one of the technical persons said that he wanted to make a presentation.

[00:19:05] So obviously we said, go ahead. And he came up, he actually presented something fascinating. So he has now developed with many of his colleagues from different districts that they’re not even located in the same, same districts or five different districts — technical personnel, civil servants. They have developed a system that is probably far better than the one that we developed in three days.

[00:19:30] Obviously, a lot more effort went into it. It’s highly secure. It can deal with food, dealers, food procurers. The one we were developing was basically just to handle subsidies. It did not connect to the food dealers, did not connect to the local markets and all of that, they actually had developed.

[00:19:49] So this is just one example and many such examples that are happening from within civil service, from outside of the civil service, new innovations. People are coming on their own saying that we have developed this. We have, we’re seeing face recognition technologies, several of them, which can do biometric authentication when touch-less.

[00:20:09] So you don’t want to have fingerprints, you don’t want to have iris, all of which require some kind of contact with your finger or with the face to some kind of device. And that is dangerous right now. So face recognition is wanting technology that I have seen just coming out of woodwork in the last one month. Companies, startups have come to us and said, “Okay, you can use this face recognition technology for identifying people who have come for subsidies. We have not used them yet, but these are things that are happening and we’re just putting them to pilot testing in different locations around the country and seeing which one makes it into a scale up and which ones don’t.”

[00:20:45] Kal Joffres: It sounds almost like for a particular challenge, you might have different solutions that are being tested in different places. Is that right?

[00:20:54] Anir Chowdhury: That is absolutely correct.

[00:20:55] Kal Joffres: Traditionally buy-in, getting people on board has been really, really important to get things done in government, especially when you’re working across different parts of government. How do you do that now that everything is happening over phone calls, over video conferencing, and you have the social distancing.

[00:21:11] Anir Chowdhury: Buy-in is a complicated matter within the government. So buy-in is often associated with buying or procurement. So what gets procured is what gets buy-in. But in this war situation, procurement, the standard procurement processes are on debate. So we still have to innovate. We still have to create new solutions.

[00:21:38] We have to put them to practice. So the standard procurement processes, which take a request for proposal responses from industry, and then a procurement process that may take months. That does not apply. Then there is another kind of buy-in, which is outside of the procurement process. It’s somebody in the government thinks that something is a good idea and says, okay, let’s try this out.

[00:22:03] But that second method is not very standard. It’s not very, I would say it’s very rare. Now we’re seeing an emergence or, or an increase of the second method. So somebody saying, okay, that looks like a good idea. Why don’t we just try this out? So I have seen cabinet members, permanent secretaries and high level officials being very open to trying out new ideas.

[00:22:29] I want to say that that’s the, that has become 100% practice within the government. But if I had to put a number, let’s see what this works, if that is that issue, if it was maybe 5% now I’m seeing that 50% so again, the 10-fold jump in terms of attitudinal shift — that “let’s try this out”. We are in the middle of this massive, unprecedented crisis. So, if you try out something new, I mean, what’s the, what’s the big deal? I mean, it can only help us. So that, I think that is a very interesting shift in mindset that I’m seeing. I consider it a war situation, right? So we are really at war with an invisible enemy.

[00:23:10] So during war, you repurpose whatever you have at home, and you are open to accepting new things because it may really help you. It cannot really harm that much. So that’s the shift that we’re seeing within the government, that government officers, the high level officials who are making decisions, are open to new ideas, are open to trying out new things.

[00:23:33] And, as you said, as you pointed out very rightly, that we are trying out multiple solutions to a problem. And that’s a good thing because we don’t know what will work really well. A different district administrators are finding themselves in different types of challenges. So at the surface, we may be categorizing that challenge as one thing from the center, but it’s really, really not one thing.

[00:23:56] So just the case of targeting the people, we need to, for subsidies, that challenge is very different in a location that has a lot of water bodies [are] very different. That is coastal, very different than hilly area. […] It has a lot of factors. So when we set targeting, in the center, we may be saying it’s a monolithic single concept, but for the district administrator, or maybe even the sub-district administrator targeting may mean something totally different.

[00:24:27] Kal Joffres: It also sounds like you’re saying that because there is less buying of things. Is that there is less of a need for buy in to get things done.

[00:24:36] Anir Chowdhury: Probably true because our buying, buying or procurement is often stifled by the theater of audit. So we often don’t get the best product because there is a procurement process, the public procurement rules and the public procurement act that govern our behaviour in terms of what we can take from the industry, from outside of the government because most of the products actually come out to the government. Most of the innovations come outside of the government, and we don’t always have the luxury or the opportunity to pick the best product because audit requirements drive all our decisions, all our thinking sometimes.

[00:25:16] But here I’m seeing that we are thinking outside of audit. We are being bold, we are being innovative in terms of what we want to pick, what we can pick. So the buy-in tied to buying is less important right now. This will continue to be the case when the lockdown is lifted, when the Coronavirus goes away.

[00:25:40] That, I don’t know, but I certainly hope that some of this citizen-centricity, some of the speed that we’re seeing, many of the collaboration that we’re seeing right now continues.

[00:25:51] Kal Joffres: I’d love to hear a little bit about what you’ve learned over the past week. What are some of the things that have happened that have surprised you or that came up as unexpected and, and turned out to be lessons for you.

[00:26:01] Anir Chowdhury: Maybe three things I’d like to point out, some maybe will have lasting impression in my work and my colleagues work. Some, maybe just a usable during this time of crisis. First is that we have realized that technology is not a panacea.

[00:26:19] So we may be looking at the lens of technology, but here we’ve had to go back to old technology. We’ve had to go back to phones. We’ve had to go back to human agents, a large pool of volunteers, as you said, democratizing the process both in terms of creation of technology, but utilization of technology. So technology and human interaction.

[00:26:44]It’s an interesting play that we are seeing that is making the most practical solutions happen. The second is, and this we’ve known for a while, but it was again, a stark reminder that the most useful innovations, the most effective innovations come from the field, not from the center. The people at the center I think are aware about problems, but don’t always come up with the best solutions because they don’t always know the needs of the people.

[00:27:14] They don’t always have pulse on what specific contexts really requires. So that’s why we are seeing such proliferation of innovations from the field. And because we are open from the center, we’re open to these innovations, the buy-in discussion that we talked about earlier. We’re open to trying these different innovations out.

[00:27:37] We’re seeing more success in my mind, at least. The third is that things can happen really fast. Just one example from last week, we were discussing a new idea within the team of how we can use phones. In combination with internet and TV to include the excluded. We needed to have this discussion with the two Ministries of Education, and would have to set up meetings with them separately.

[00:28:05] I was thinking that we would take a few days to set them up separately. Our ICT Minister decided that we should have a joint meeting. I don’t recall in my 13 years of working with the government of Bangladesh that I’ve had a single meeting with the two Ministers of Education ever. So he approached the three ministers in these two ministries.

[00:28:26] Within 16 hours, we had a meeting, which lasted for about two hours. We had four ministers, including the Minister of ICT. We had a number of permanent secretaries of these ministries. We had probably about 10 or so director generals of these of the directors under these ministries. And within two hours we were able to come up with the idea that could potentially work.

[00:28:52] So this basically reminded me again that things can happen really quickly and often the silos that exist within the government, that exists because of physical distance that exists because of perhaps egos. That exists because the protocol does not exist to have two ministries and the ministers of those ministries have a meeting without the prime minister presiding it. Here nobody presented, thus the senior-most minister was the Education Minister. And she said, “Okay, I’ll just be a participant, and we’ll all just discuss and we’ll all look at each other’s ideas and move. Just make a joint decision.”

[00:29:35] So it was very collaborative, and two ministries came together with one solution and we’ll try out this solution. But the learning from there is that the potentially impossible can happen.

[00:29:47] Kal Joffres: Anir! Thank you so much for joining us again this week.

[00:29:50] Anir Chowdhury: This has been a particularly, I would say, enlightening series for me because I’ve had the time to reflect and in the feverish pitch that I’m working in right now, that is often very important. So I really thank you for giving me the opportunity.

[00:30:06] Kal Joffres: Next week Anir will be joining us again and we’ll be digging deeper into changes that are happening in the education system in Bangladesh.

The Innovation Dividend explores how innovation in society and government are paying off. Over the coming weeks, we’ll be exploring how ‘policy frontliners’ are innovating in real time in the COVID-19 pandemic and asking which of these changes and “raw learnings” might become part of our new normal.

You can subscribe to the podcast on Spotify or Apple.