Auto-translation used

Education + technology: our experience in EdTech

A little bit about us

We are a Troubleshooting Technology company, which has been engaged in the design and development of IT solutions for business, education and government agencies for more than 6 years. During this time, we have implemented dozens of projects, from mobile applications and web services to integration platforms and analytics and automation systems.

What are we going to talk about

One of our key areas is EdTech, digital solutions for education. We have launched and are supporting projects of various scales: distance learning platforms, certification and online testing systems, proctoring, e-schools and diaries, competence assessment and career guidance services, online training simulators.

Our solutions operate at the national level and cover more than 300,000 users, including in the regions. And it is the experience of launching in the regions that is the topic of this post.

To the case itself

We have developed and are supporting a large-scale online olympiad for schoolchildren from different regions. The platform includes the automatic generation of tests from the task bank, the assignment of tests, the passing module itself, and a proctoring system for integrity control.

Before the large-scale launch, testing was carried out in several schools. Everything was under control: the equipment was in place, test classes, work chats, a call center, and managers were in touch. The system was previously repeatedly tested.

The feedback came immediately. Someone doesn't open, someone doesn't start, someone can't log in to the test, and proctoring doesn't allow them to pass. As a result, about 10% of the participants successfully completed the test.

We began to disassemble each case. Reasons: outdated operating systems and browsers, weak Internet (sometimes so much so that the user pressed the button faster than the request could load). But the most unexpected and massive problem is that users simply don't read instructions and system notifications.

After testing, we worked on the errors, made the necessary improvements and successfully held the Olympiad, we will not touch on this.

But what's more important is what we've learned along the way.

The problem wasn't with the system itself. It was in our perception of the user. We looked at everything through the eyes of those who are used to stable Internet, modern devices and interfaces.

But in the regions everything is different.

  • People don't read the instructions because they're afraid of the interface itself.
  • Any notification on the screen is perceived as a system crash.
  • The devices are old, the Internet is unstable, and the browser may not have been updated since last year.
  • The guys we consider "digital" actually get lost when the process doesn't look like the usual TikTok or messenger.

All the requirements are different, the pace is different, the user is different.

We draw conclusions

What you emphasized for yourself:

  • For some, an "extra button" is already a potential problem.
  • Direct logic and large elements. Even if it doesn't look so good somewhere.
  • You don't read the instructions, but you can't always do without them. That's why we use videos or graphic sketches instead of text. And we don't let them skip

What would you like to recommend for your projects/startups/products:

  • The user is not always "digital", not always technically confident, and does not always read what is written.
  • It is useful to test a product on “difficult” users, especially if there are many of them expected. One test with an untrained person will give more than 10 casting tests.
  • Figure out in advance how we will explain. If the system requires instructions, they should be ready immediately. And it is better if they are embedded in the interface.
  • Put support at the start. Especially on the day of the launch. Even if everything is working, the user may have an alarm from scratch. And then it won't turn into a negative.

Comments 0

Login to leave a comment