Q&A: Ananya Agrawal

Embracing Interactive Voice Response (IVR) Technology for Youth-Based Life Skills Training.

July 15, 2022

As an educator and social entrepreneur who strongly believes in the value and potential hidden in grassroots, Ananya Agrawal has always been sensitive towards leveraging her design and technical education for social impact. In her latest endeavour as the founder of Dreamverse Learning Lab, an edtech consulting studio, Agrawal gave a new face to the Call-a-Kahaani project, which is a story-based participative network that models community innovation and entrepreneurship around Interactive Voice Response (IVR) technology to engage youth with life skills learning.

After learning about her project, which was successfully launched during the height of the pandemic, we reached out to Agrawal to share her story in our inaugural Q&A series.

1. What inspired you to become a part of the Call-a-Kahaani intervention in 2020?

Having spent the year 2018 in a media dark zone in Arunachal Pradesh, modelling a maker lab, and much of 2019 understanding the scale of the problem of quality education at a policy think tank , I had built considerable grounding and inclination to work in the education sector, especially in the life skills space. Mid 2020, the knowledge that school closures, due to the pandemic, were affecting more than 1.5 billion learners worldwide with almost 24 million at the risk of dropping out of school, was more troublesome for me. As a designer who is biased towards taking action – research and ideation was the next best step that I could think of.

For the first prototype, I made a few sketches and tested out a sample content piece with my housekeeper’s kids. This was done in a frugal manner, unsure of how I should take it forward. In August 2020, I also started looking for organizations to collaborate with when, in a moment of serendipity, I stumbled upon Udhyam Learning Foundation and got to know that they had recently started prototyping this product called ‘Call-a-Kahaani’, and were looking for a consultant to work with them on the same. Over time, I ended up owning the idea end-to-end, and the organization supported me to grow it into an independent entity under my own organization, Dreamverse Learning Lab, officially registered in early 2022.

2. How did you arrive at the idea of using Interactive Voice Response (IVR) technology as a way to teach learners important life skills?

Around May 2020, I had already collected a lot of data and information in the form of newspaper clippings, articles, conferences on the COVID response to education, etc., in order to understand  how learners were being reached out to and where the gaps in response were. One of the key insights that I had derived back then was that a lot of content was being broadcasted, but its learning impact was still questionable given the limitation of data capture mechanisms, especially for mid-to-low tech mediums like TV, Radio, or SMS.

It was during this parallel research project, when I was exploring an accessible job portal for migrant workers, that I stumbled upon IVRs as a technology. I found it intriguing that one could broadcast content with such ease, using a telecom network, without the need for the internet and can be accessed 24/7 by the end-user. Not just that, the platform had the ability to record data on key presses, which could be easily used for quizzes, surveys and polls for learning, and even voice recordings, agent calling, and conference calling – which can be leveraged for receiving qualitative inputs.

The platform could also be seamlessly connected to a chatbot interface like a WhatsApp or contextually created Webapps, where one could add extended functionalities for users who have better access to the internet. However, the IVRs-based education platforms that I came across back then, were few and were either using just the broadcast feature to deliver content or just the agent calling feature to solve student or teacher doubts regarding subject-based content. I saw a lot of possibilities for using IVRs as a scalable data-driven medium for learning life skills, especially in the current scenario, where as of June 2020, 70% of rural India was still struggling with a lack of internet.

3. You started the platform in the midst of the COVID-19 pandemic, how did you navigate this period during the launch of your platform?

We started with a remote-only way of engagement which is definitely not the best possible way to work on learning, as it knowingly or unknowingly ends up replacing the teacher or facilitator. However, given the limitations of the situation, we tried to make the best of what was possible during this time. The first approach was to work with Udhyam’s existing audience in polytechnic institutions like Haryana ITIs, tapping on the core reasons Udhyam wanted to experiment with this medium.

The next approach was reaching out to other NGOs like Those in Need and Teach for India, Hyderabad, which were happy to plug in our content to their work. With these NGOs we experimented with a variety of on-the-ground strategies, such as using a chain reaction approach, where each volunteer had to tell 5 other people about the program, which was a simple but effective way to reach out to more than 2000 people within 1.5 months.

The third approach to reach out to learners remotely, was through community radio advertising for which we prepared jingles, promos and even full-fledged shows with the support of organizations like Lalit Lokvaani and Waqt ki Awaaz, which were broadcasted as well as narrowcasted (weekly focus group engagements in their region of work) with their audiences. This was also the time when we realised the power of facilitator-supported blended engagements on learning outcomes.

The last but most important approach that we tried during the pandemic, was engaging our learners on the platform itself. This was done via co-creation campaigns like Gully Tales, or story-sharing campaigns like ‘What’s Your Kahaani’ and ‘Apni Beat Par’, or introducing features on IVR like ‘Talk to Us’, where learners would directly get connected to one of the team members and have an open-hearted conversation.

4. Call-a-Kahaani has been able to deliver its content to learners without the need for Internet access; how has it been able to do so while ensuring the learning experience is both engaging and interactive?

A key engagement criteria is having the content on the platform to sound as human and relatable as possible. The voice artists who help us with our content are therefore encouraged to use voice modulations, sound words and dramatic music to keep the audio real and fun. We also actively worked on building a persona around the navigation, and content, as well as indicative music to communicate the beginning or ending of a content piece, a quiz, etc. to build familiarity.

To segment our content better, starting October 2021, we have introduced region-specific, time-bound programs where we make it a point to conduct listening-in calls with relevant stakeholders at the start of the program, to be able to understand the regional context better and tailor-make our vernacular content accordingly.

Understanding nuances of voice-based user interfaces like voice assistants, apart from radio platforms, have helped us design these small but important interactive elements better. The key press feature has always been fun to experiment with as an element of interactivity; it can invite the user to make a choice to direct how the story proceeds, or even conduct a small quiz or survey to test recall. Key presses are always an essential part of our voice bot design, as even if a user is too shy to speak up, if the content is able to spark enough curiosity, they can easily send key press inputs as a response.

While our earlier experiments were done inhouse, using Kaleyra as a service provider, starting Oct 2021, we are taking the support of the Gramvaani organization for our tech, where they have introduced us to another interesting way of building human connections; they publish user recordings back on the platform in a way that allows users to listen to each other as well as share what they have to say. This is a great way of building a voice-based peer discussion forum, and we are working on facilitating meaningful conversations leveraging the same.

5. A particular EdTech trend that has seen significant adoption in 2022 has been the use of immersive learning technologies to introduce real-world scenarios; what impact have you seen in the learning outcomes of your students with IVR?

Our content design started with publishing relatable, real stories of grassroot entrepreneurs to build an “i-too-can” attitude in our learners across India. These stories were interspaced with situational questions where we encouraged the learners to pause, reflect and make a choice, to be able to see how a simple growth-minded approach in decision-making can lead to positive outcomes in life. Most of the learners found these stories ‘inspiring’ which made them feel ‘confident’ about making their choices. Some learners also spoke about how the ideas shared in the stories were something they would like to try, or how they gave them exposure to what was possible, especially to the girls in rural areas whose exposure to media and opportunities is still limited.

However, we didn’t want to just bank on inspiration and wanted to shift to a more actionable learning outcome. In our next pilot, specific to the Odisha region in India, we tried to create dedicated modules around a specific mindset focus. For instance, a module that focused on trying new things was built like a game which we called ‘inspiration hunt’, where our characters would encourage learners to figure out what was being created, and the answer would be revealed in the upcoming episode.

Hence, the real-world context over here meant engaging in real world activities with us, to figure out what inspired them to learn and work. We observed the positive impact of this through increased queries to local facilitators when this change followed, as well as voice recorded anecdotes on the platform.

An extension to this approach is the project-based methodology that we have deployed currently, where we introduce our learners to innovative thinking processes like world building, tinkering, observation, listening, feeling, questioning, problem solving, rapid prototyping, etc., where our fictional characters (as well as interesting case studies from grassroots communities) introduce the learners to different ideas and perspectives that help one to become aware of their inspiration and challenges, think innovatively and translate thought to action. This ongoing 6-month, 30-episode module in the Bihar and Jharkhand regions in India, is designed with an objective to ‘groom’ rural communities into the processes of innovation, which can help them to solve local problems for themselves and adapt to uncertainty in the real world. We are excited to see what the learners create.

6. Another emerging trend, not necessarily unique to EdTech, is the use of data analytics to assess and analyse behavioural activity to measure progress and develop more personalized experiences. How has data helped the platform and facilitators understand the needs of learners better?

By leveraging the key press feature on IVR to send out surveys and understand who was listening in, we analyse this data and identify interesting data patterns or anomalies, which has helped us considerably to stay agile and re-iterate our program, based on learner needs. These data points have not only informed our content and operations, but our monitoring and evolution processes have also evolved significantly over time to guide us on when and why to seek data, as well as how to use the data.

At present, we start our region specific programs with baseline surveys which are quantitative in nature. But given that life skills education is highly qualitative, we also  conduct qualitative surveys with a focus group from the larger audience. These focus group qualitative surveys help us understand the prevalent mindset, reasons behind the quantitative responses, and the relevant challenges that we can pick up and address in our episodes.

Once we wrap up this baseline, we take 3 kinds of inputs for each episode; an MCQ (multiple-choice question) based quiz to test recall; voice input to capture anecdotal responses; and an observation worksheet to capture field insights. These inputs help us understand engagement in terms of interest, remembrance and habits formed in our learners, apart from visible actions seen in project output.

There are other interesting tech behaviour metrics as well, which give us quantitative feedback on our work. For instance, analysing where the learners are dropping off, retention cohorts help to highlight tech and content related gaps, while observing day-to-day spikes and drops in users tuning in or voice recording submissions give an insight on field operations.

We also periodically assess anomalies in data and clarify assumptions by calling on a subset of phone numbers from our database to conduct one-on-one interviews. Voice recordings and written inputs from field teams are also tagged, clustered, and analysed to draw relevant insights – like relevance of content to learners, learning style or preference, any language or comprehension issues, resource availability, etc., which helps us improve our content and processes over time, get in touch with relevant experts for compiling informed FAQs.

7. Personally, what has been the most inspiring engagement you’ve had with learners to date?

Every time we go out on the field, or hold a training or listening-in session, or hear anecdotes that validate the need and scope of the platform – these experiences inspire us to keep going.

One such experience I would personally like to highlight, is when we were working with Radhaji from Waqt ki Awaaz Community Radio, which operates in a village near Kanpur, India. She actively focuses on women and girl-related topics in her content and requested us to tailor our content accordingly for a 4-episode pilot.

Our intent with the pilot was to try and understand the power of relatable storytelling to build conversations, and this proved to be an excellent platform to explore that. Our recurrent themes of non-cognitive skills were contextualized; our first episode focusing on what it means to dare to dream big for girls and women; the second one spoke about equal work opportunities for men and women; the third was on domestic abuse (which had become a common issue in villages during COVID); and the fourth on the value of being a sensitive man in today’s society.

We had designed the whole program as a conversation, interspaced with small stories. During the discussions that Radhaji and her team held with the audience, or even the ones our team held on IVR over the ‘Talk to us’ feature, very beautiful and heart-wrenching stories would come forward. Some in particular were of a father supporting her daughter to study against the societal constraints of the social group they belonged to; or of the frontline health-awareness worker and how she was treated in the village during COVID, and how she showed resilience to continue working despite the circumstances; another was about a female writer and what it means to be one in her village, and many more.

I had never imagined that this month-long pilot could be so enriching in its very nature. Post the pilot, Radhaji told us that women from the village were asking her when new episodes would come by, but sadly, due to bandwidth issues and also because the theme was a little different from the entrepreneurship and innovation space which we wanted to focus on at this point, we had to discontinue. We hope to resume working with Radhaji and organizations like hers, on more targeted themes and spaces, once we have established a more stable operational model for ourselves.

To date, Call-a-Kahaani has been able to engage over 13000 learners who’ve made over 59000 calls to the platform. From December 2021 to March 2022 alone, they engaged with a targeted group of more than 4700 learners in a region-specific pilot based out of Odisha, India, from where they received over 37000 calls, to engage with character-led interactive stories and activities.

In mid-May 2022, they have started working on a 6-month project-based learning journey, targeting a focus group of 5000 learners based in the states of Jharkhand and Bihar in India. The pilot is being conducted in a blended mode, with 30-to-60- minute weekly engagements with local facilitators, through which they hope to deepen the engagement and learning impact.

“We aim to take things gradually and create a community of 25,000+ active learners, by the end of 2024,” says Agrawal.

To enter sign up to our newsletter below.

View competition terms and conditions here  


"*" indicates required fields

This field is for validation purposes and should be left unchanged.