How we increased LTV by 20% and conversions by 30% in just a month by building an MVP for children’s educational platform

If you still think that NoCode development is just a tool for creating landing pages on Tilda, you definitely need to read this case.

In just a month, we built a technical MVP for a children’s educational service without any code at all and increased the conversion rate by 30%, and LTV by 20%.

Hello! My name is Rinat, I am the co-founder of the LowCode Development studio 021. We really built a full-fledged technical MVP and even achieved amazing metric growth in just a month. In this article, I will talk about the process, hypotheses and why NoCode is not about websites created by designers, but about a full-fledged and flexible tool that can be integrated into an existing business and automate processes in it.

Before that, the client reached out to the developers to perform the task by usual code. As a result, their cost was three times higher than ours, and the terms were twice as long. So NoCode is not only easier, but faster and cheaper.

About the client

Dizz.tv is a platform with educational broadcasts for children. They conduct live streams on topics that are interesting to kids, but with educational overtones. For example, a child can learn languages, maths and natural sciences by watching videos about Minecraft. That is kind of a useful YouTube that works by subscription.

Since the current MVP version of the product was assembled in a hurry, it used to work on a Discord messenger with a bunch of settings, as a platform. At some point, it felt too small and uncomfortable, as it cut off all conversions at the onboarding stage, and limited the creators in testing hypotheses. It’s time to move, and at the same time to test a couple of existing hypotheses.

Tasks set

We dug into the product, formulated a bunch of hypotheses that were tested by NoCode development, but in this article, we will analyze two specific ones.

Hypothesis 1: increase lifetime value if you show the child’s results obtained on the platform to the parent

Since the decision to purchase goods for children is made by parents, the product must also meet their needs. To do this, we decided to show parents an individual plan for their child and the progress over the past period. Thanks to the demonstration of the results, we planned to increase the conversion of payments in the next months, that is, to increase the LTV.

So our task will be to make sure that the platform can collect progress data from various sources (Discord, quizzes, streaming platforms), convert it into an understandable form, display it in the personal account, and also send it to the parent by mail and Whatsapp.

Hypothesis 2: increase the conversion rate if you call only those who watched the trial lesson at the beginning of the funnel

The platform’s sales model was as follows:

Step 1: The parent leaves a request

Step 2: The parent gets a free trial lesson on the record

Step 3: The parent receives push messages and mailings with a call to watch the lesson

Step 4: The manager calls the parent and sells

As you have noticed, this sales model involves a real person — a sales manager. You will need to pay him a salary, which is why the CAC cost is growing. At the same time, it is obvious that only those who watched the trial lesson buy the product.

So the second task was to make sure that we could know who watched the trial lesson and who didn’t. After that, we wanted to transfer this data to CRM so that the manager would not waste time on those who did not watch the trial lesson and could focus on warm customers.

Tools used

• Bubble — to create a personal account where personal information will be displayed. The service tracks the user’s time on the site and the child’s interaction with the system. The client already had a personal account on Tilda, but it did not work, because Tilda cannot display individual information for each user.
• Integromat — for connecting all services and data. NoCode is the server part of the system.
• AmoCRM — for operator-client and curator-client interaction + sales bot for automating the setting of tasks and transferring cards.
• Pact — for sending to messengers
• SendPulse — for emailing
• Google Analytics, Google tag manager, Metabase — an analytical layer to collect all the necessary data and transfer it to Airtable.
• Airtable — a single database. A place where data from CRM, analytics and all other services are concentrated.

What we did

To begin with, we needed to collect all the information about the child’s activity:

1. Watch time of training videos for a day/week/month
2. Completed interactive tests after the videos and homework
3. The child’s activity in a chat room during the educational broadcasts.

We decided to collect all this data and send the following message to parents once a week:

“This week your child watched educational videos for two hours and completed 5 tasks instead of YouTube and computer games.

He/she learned about:

• why fire can’t burn in water
• countries in which tigers and lions are found.

You can solidify your knowledge and have an interesting chat with your child tonight!”

We used Google Analytics, GTM and Metabase to collect activity information:

This is what the Metabase cabinet looks like. It collects all the analytics in its raw form. Here we write SQL queries to get data about the views of the child, his activity, visits to send it to the parent

And these are the raw data that had to be processed manually. And, of course, the startup does not have enough resources for this, so one of our tasks was to automate data processing. To do this, we made several dashboards and gave them to each user in the personal account so that the parent could view all the information on the child’s education directly in it.

But not all parents log in to their personal account so often, and those who visit, and so update their subscription anyway. Therefore, it was decided to send them statistics by mail and Whatsapp, for this we used Integromat.

At first, we had difficulties connecting to amoCRM, but since we are an official partner of Integromat, we had priority support. We were appointed a manager and we quickly solved the problem.

As a result, this solution allowed to increase the conversion rate from the first month to the second by 20%. Complaints from parents “What am I paying money for?” were all gone, thus relieving the support department and the repeat sales department.

Next, we needed to understand who watched the trial lesson and who didn’t watch at all, so that the sales department could focus only on warm customers.

In general, this is one of the top priorities for business: to identify early triggers that will help you understand whether a customer will buy a product or not in order to optimize the work of sales managers.

We also solved this problem with the help of No Code tools. We needed to:

• Understand which of the potential customers had already watched the trial lesson
• Monitor the watch time of the lesson in order to understand whether we should work out objections (30 % watch) or close for sale (70% watch).
• Transfer this data to the transaction in amoCRM
• Automatically assign tasks to the manager in CRM for warm clients

As a result:

1. The client left a request for a trial lesson
2. He received a letter and a message in the messenger with a unique lesson link (Integromat helped to link all components and generate links)
3. The client followed this link and started watching the lesson
4. As soon as the client had watched 70% of the lesson, the manager immediately called him to close the purchase.

Integromat linked all the components of this chain together and updated the status of the transaction when it changed. Salesbot (part of amoCRM, which allows you to automate actions) sent messages and assigned tasks to managers.

GTM kept track of how long the video was watched by a potential client. When a client views 70% of the video, a trigger is triggered, we switch the customer data sheet status in amoCRM to a special status. A newsletter is sent and the operator calls him to ask about the viewing experience and enrol the child in the course.

This is how the customer data sheet looks in Tag Manager. Here you can see that the trigger was triggered: the client watched 70% of the trial lesson, the manager can call him.

And this is how the user’s path scenario looks after viewing the trial lesson in full, all programs and services that help switch transaction status data are shown here.

After implementing this feature, we were able to increase sales conversion by 30% with the same number of sales managers. We optimized the department’s work, as they stopped making dummy calls to customers who were not ready to buy.

In short

As a result, during the month of work of our LowCode 021 team, we were able to improve the customer return conversion from the first month by 20%, and the sales conversion by 30%.

We have built a chain that automates the work of the sales department, set up a personal account with individual information for each client. All this only with the help of No Code and without using development at all, which is usually the most expensive part of the budget and time on the project.

So this case is another confirmation that NoCode is about flexible tools that can be integrated into an existing business, saving time and money, as well as about cool results.

In 021, a NoCode-development studio, we use dozens of proven solutions and help small and large businesses launch products in 2–3 weeks.

You can sign up for a free consultation on our website. In half an hour we’ll answer all questions, help select necessary tools for your idea and estimate the costs and terms.