Teller + Orange Money!

We are so excited to share that we officially completed our pilot with Orange Money and are ready to launch a full rollout. Introducing MoMo –– a mobile-money assistant that helps  people open an Orange Money account, check fees, find agents, and pay bills. Check out our TV spot below!

Thoughts on Building Hybrid Bots

Before I get to specifics on building hybrid bots, I wanted to spend some time discussing what makes a successful bot experience. For customer service-focused bots, I believe the keys to success are: (1) narrowing the scope of your bot as much as possible and (2) making sure you nail the bot-to-human handoff. Let’s dive into each piece.

Narrowing Your Scope

You may not realize it but the bar for customer service is actually very high. Aside from a few exceptions, traditional customer service works pretty well because humans are usually pretty intelligent and can easily process millions of different commands. Additionally, people don’t come to customer service to ask simple questions. If they have a simple question, chances are they attempt self-service or Google it and find an answer (reports confirm this).

Given these two simple facts, if your bot is supposed to fully replace a human CSR (customer service representative), it is inevitably going to disappoint customers¹.

But don’t worry — there are still plenty of use-cases where bots are well suited! Need to walk someone through a long form, like an insurance checklist? Bot. Want to send proactive reminders or receipts? Bot. Want to simplify a time-consuming process, like contacting your elected officials? Bot.

What do these examples all have in common? They only do a few things but they do them really well. That is the future for most consumer bots.

Making sure your bot maximizes its strengths is only half the battle. Eventually, your bot will still get confused and that is when you will need to figure out the best way to have a human take over.

Building a Hybrid Bot

There are several factors to consider when designing a ‘hybrid bot’ system for escalating to a human CSR:

  • What channel is the bot on?
  • Do you need to provide a chat interface for the CSRs? Do they already have a CRM that you need to integrate with?
  • How should the hand-off work in the user flow?
  • Should the user be aware of this passover or should it happen behind the scenes?
  • How fast can the CSR respond to a user who is waiting?

The answers to these questions will determine how you proceed in building a hybrid bot. Here are some options and ideas based on my experience researching and implementing them.

Option 1: Facebook Messenger

Facebook Messenger is an ideal channel for bots for two reasons: (1) there is a messaging interface built into the Page admin panel and (2) there is an API to pause the automated responses. Here is a basic flow for the bots I have built:

  • A user triggers the fallback intent or escalates directly (e.g. “Speak to a human”).
  • Your middleware calls the FB handover protocol to pass control to the Page admin interface. Alternatively, another “hack-y” workaround is pausing responses for this user for a specified amount of time. This requires an actual business process discussion and making sure someone will help out the customer in a timely manner.
  • Finally, I generate an automated email for the client with the user’s name, the trigger message, and a link to the conversation inside FB.

Option 2:

For bots that are not on FB Messenger, offers another quick way to build a hybrid bot². It offers a lot of functionality to build a bot that can respond to a user or escalate to a human (including a built-in option to send an email notification). Configuring this is pretty straightforward but here is my basic process and some considerations:

  • Use’s modules to construct your bot flow³.
  • Create an ‘Await Human’ module with some basic text (e.g. “Please wait while we connect you with someone on our team”)
  • For your ‘Bot Statement’ module, add a connection to your ‘Await Human’ module that triggers when someone says the words ‘Help’, ‘Manager’, ‘Agent’, etc.
  • Note: obviously, keyword based routing is very simple. For a more advanced configuration, I’ve used a webhook to pass a custom variable that also triggers the ‘Await Human’ module.

Option 3: Custom build

If you are looking for something more advanced, such as integrating with an existing customer support tool, or have additional privacy/security requirements, a custom build is probably the way to go. One workflow is found in this article. However, I personally feel that this integration is overly complex and would end up being very pricey.

The key to a custom build would be making sure you have four pieces: (1) A front end for users to message (which could be web chat, SMS, FB Messenger, etc.) (2) The NLP engine (3) CSR chat interface to respond to escalated messages⁴ and (4) The middleware to connect everything.

Due to distinct business requirements, I find that nearly all of my projects end up being some form of a custom build (but hey — that keeps things interesting).


I hope some of these ideas are able to help you with your next chatbot project and if you have other suggestions or examples in this space, I would love to hear about them. If you are interested in talking more about bots, especially in the finance and banking space, please get in touch.


  1. It took me a long time to realize this but I strongly believe it now.
  2. was recently acquired and is not allowing new signups. Hopefully that will change soon.
  3. Or you can be really creative and use their webhook module to pass messages to another NLP engine like Dialogflow,, Lex, Watson, etc. Proceed at your own risk.
  4. One option I am interested in exploring is Front.

Introducing: The Finance 150 Project!


After months of research, curation, and translation, we are excited to announce the launch of The Finance 150 Project. At its core, this project is aimed at providing simple and colloquial definitions of core financial terms. We have provided definitions in eight languages, and we hope to add several more after launch. More information on the background and purpose are on the project page. 

We know that providing definitions isn't a silver bullet to solve financial inclusion. But we strongly believe that explaining products in simple, transparent language is a prerequisite for any financial institution hoping to attract a new, potentially underserved market. If you work for a financial institution or financial literacy organization, please consider making our content available to your customers. 

The Finance 150 Project is a work in progress, and we hope we can leverage the financial communities help in expanding and improving the content. Please contact us with any feedback, ideas, suggestions, or business inquiries. 

Leveraging Chatbots to Drive Financial Inclusion

Quick author’s note: I originally wrote this paper with the DFS Lab team as part of my application for a grant from the Bill and Melinda Gates Foundation. I recently found out that my application was successful so I wanted to share this ‘Theory of Change’ document for anyone else interested in financial literacy and mobile-messaging applications. I know I still have a lot to learn about this space so please reach out if you want to discuss this paper or Teller.

Improving Access to Financial Services Through Mobile Messaging Applications

In recent years, the rise and spread of mobile phones has provided a valuable tool for people to connect with each other, access information, and interact with businesses in new ways. In the same period, however, banks and other financial institutions have been slow to embrace mobile technology to provide widespread access to financial guidance and services.

We believe that mobile phones, and especially mobile messaging applications, provide the perfect channel to increase access to financial services and also deliver financial coaching and guidance. Coupled with the ability to understand ‘natural language,’ mobile messaging applications can allow everyone in the world to have their own personal financial coach in their pocket. In the not-too-distant future, these financial coaches will not only have a full picture of an individual’s financial situation, but will be able to proactively provide tips and manage their financial services and accounts.

“We believe that mobile phones, and especially mobile messaging applications, provide the perfect channel to increase access to financial services and also deliver financial coaching and guidance.”

In this paper, we discuss the state of financial education and literacy, why banks and financial institutions are the ideal organizations to provide messaging-based solutions, and how Teller’s technology can enable these financial institutions to improve access and use of their financial services.

Section 1: Financial education is proven to raise economic outcomes

Two of every three adults worldwide are financially illiterate, meaning they lack the knowledge and skills to use their financial resources effectively for lifetime financial security. This staggering share of people is disproportionately concentrated in developing countries, where formal financial services and financial assistance programs are often lacking. Among the financially illiterate, women, the poor, and lower educated respondents are more likely to suffer from gaps in financial knowledge.1

Financial literacy only provides an educational descriptor of a population. To actually see why it matters to general well-being, we need to go one step further. Numerous studies have shown a strong relationship between financial literacy and positive economic outcomes. In one of the first studies in this vein, Hilgert et al. (2003) document a “strong relationship between financial knowledge and the likelihood of engaging in a number of financial practices: paying bills on time, tracking expenses, budgeting, paying credit card bills in full each month, saving out of each paycheck, maintaining an emergency fund, diversifying investments, and setting financial goals.”2

Furthermore, as household income and wealth begins to increase, numerous studies demonstrate marked improvements in both health and happiness.34 Building better financial habits and accumulating savings is one small, but important, tool to increase general quality of life.

Section 2: Banks and financial institutions should provide coaching programs

One strategy to improve financial literacy is to simply improve access to financial services and encourage usage. According to a 2015 S&P Study on Financial Literacy, “adults who use formal financial services like bank accounts and credit cards generally have higher financial knowledge, regardless of their income. Even poor people who have a bank account are more likely to be financially literate than poor people who do not have a bank account.”

Even more importantly, academic work by the DFS Lab team has shown that access to formal financial services can improve income and well-being and help bring people out of poverty.5

Clearly, greater access to financial education and financial services offer a straightforward path to better economic outcomes for the financially uneducated and underserved. At the intersection of these two spaces (education and services) lie formal financial institutions such as banks.

A financially educated population not only offers these positive externalities, it is also actually good business for banks. Financial education can be the difference between a profitable loan repayment and an unprofitable loan default. Financially savvy customers are also more likely to use additional banking products. So why haven’t there been wide scale efforts by banks to tackle financial illiteracy?

Historically, providing a wide-scale, in-person financial coaching program would encounter the following obstacles:

  • Location dependence — a bank would only be able to serve an audience in its geographic vicinity
  • Labor dependence — financial counseling would require educated financial counselors or volunteers
  • Time dependence — requires that bank customers be free and available during the financial counseling sessions
  • Language dependence — the bank must provide support in multiple languages
  • Fear of the unknown — research has shown that for many, formal financial services can seem intimidating, inaccessible, and only for the wealthy.6

All of these factors contribute to a very high cost for educating and serving the most at-risk customers: women, the poor, and the uneducated.

Section 3: Teller is the first truly scalable financial coaching tool

How can banks provide financial education at scale and also address the many challenges posed by existing models of educational programs? The answer is through mobile messaging applications such as SMS, Facebook Messenger, and Whatsapp.

In the past few years, messaging applications have grown exponentially, and now have over four billion monthly active users.7 Adding in SMS, messaging apps are available to nearly everyone in the world. After all, unlike smartphone mobile applications, even the most basic feature phones are capable of sending and receiving messages. In this same time frame, two very important developments have taken hold:

  1. The opening of messaging platforms for developers — As of two years ago, it was impossible to build a messaging application ‘inside’ one of the popular messaging platforms. However, more recently, the major tech companies have recognized the same potential that we are highlighting right now: businesses and other organizations will want to communicate with users through messaging. Accordingly, many of the major messaging platforms including Facebook Messenger, Line, and WeChat have already opened the gates, and many others will follow in the next few quarters.
  2. The rise of ‘natural language understanding’ engines — Just a few years ago, if you wanted to build any kind of chat-based software, you would need to hire PhD’s to help you build the artificial intelligence backend of your application. These days, major tech companies are racing to release their cutting edge ’NLU’ tools to give developers access to the same technology that powers IBM Watson and Amazon Alexa.

With these new NLU tools, and messaging apps to deploy them, businesses are exploring brand new ways to provide services. These new services are commonly called ‘chatbots.’

Teller is a messaging platform, or ‘chatbot,’ that uses natural language understanding to help banks and other financial institutions, on-board, educate, and provide continuing support to their customers. A potential bank customer can chat with Teller to learn step-by-step how to open a bank account, set savings goals, build their income, and more.

Teller directly addresses many of the previous hurdles to chatbots.

  • Location agnostic — users can message their financial institution from anywhere in the world
  • Automated — because the entire system is automated, a chatbot can talk to ten or tens of thousands of customers at once (with only minimal staff needed for escalations)
  • Time independent — users can start a conversation, and if they get busy, resume hours or days later whenever they have time
  • Language independent — Teller will be available in multiple languages such as English, Hindi, Swahili, and others
  • Low barrier to use — talking to a real person in a bank can be intimidating, while chatting with a virtual banking assistant through a messaging app is much more approachable

Because no humans are involved, and the technology is hosted on cloud services (like Amazon Web Services), the cost for providing these tools is extremely low. Teller is currently exploring pilots with major financial institutions in both South Asia and Africa. These financial institutions are leveraging the Teller technology to scalably provide quick and convenient customer service and education to their customers. They are also excited at the prospect of using Teller to expand their reach to new customers.


Existing research, including and especially work done by the DFS Lab members, indicates that increasing financial literacy and access to financial services provide an achievable solution to improving economic well-being. Banks have historically been unable or unwilling to take on this challenge based on their own internal cost-benefit analyses. However, recent developments like NLU technology and the opening of messaging applications have created an opportunity. Teller allows banks to cheaply and scalably provide financial coaching and guidance to new and existing customers. With the help of Teller, financial institutions will finally be able to offer quick, convenient, and personalized financial guidance to anyone in the world.

1. 2
4. to-Health-and-Longevity.pdf

Teller at Digital Banking 2017

Super excited to attend my first Digital Banking conference, especially since it's in Austin! I can't wait to meet everyone and learn more about the innovative technologies coming to the retail banking world. 

If you are attending and want to meet up or grab a coffee, please let me know by sending a message to See you there!


(Link) Can Chatbots Promote Financial Inclusion?

Since attending the DFS Lab in Sri Lanka, I've been thinking a lot about how chatbots can be used to promote financial literacy and access to financial services. I believe that given mobile penetration around the world and the proliferation of messaging apps, chatbots present a great opportunity for formal financial institutions to reach new customers. Many of the same thoughts are echoed in this write-up about chatbots being used to promote financial inclusion. 

Attracting Millennials Through Online Support

I found this article on LinkedIn and am really glad I came across it. While I had an intuitive sense for much of the findings in this report, I was glad to see it confirmed through this research study. One thing really stuck out to me: 

  • Online before Phone Support: >60% of the respondents that eventually call the support line for their bank first look online at the banks website or do a search on Google.

It's clear to me that customers actively try to answer their questions themselves if they can (or conversely, really dread calling customer support). This is just one of the reasons that I feel like a chatbot solution is perfect because it provides customer support answers without the time and effort required to call your bank.