V-0.1 test closed, what we’ve learned!

We recently shipped V-0.1 of ASTROSAFE (Prev. URSOR), a safe browser for curious kids ages 4 to 12, designed to make the knowledge of the Internet accessible to them without the inherent risks and problems associated with putting kids online.

6 months of R&D, talking to hundreds of parents and teachers, and prototyping ideas and demos, have made 3 things abundantly clear so far: 

1. Families absolutely need what we’re building, and are prepared to pay a monthly fee for it.
2. The positive societal impacts resulting from enhanced digital education through ASTROSAFE will be huge.
3. The positive impact on children’s emotional wellbeing will also be enormous if we build the right product.
   
   

V-0.1 was tested with a closed group of Alpha families, and we’ve learned a lot about what needs to be built for a successful public launch.

Here’s a recap of what we’ve built, what we’ve learned, and where we’re going next...‍

What we built‍

For speed, V-0.1 was built as a desktop app for Mac, Windows and Linux. We designed ASTROSAFE using FIGMA and built it all in Electron. More on this later though.

‍This is how V-0.1 works for kids:‍

• It’s a browser, where children can only navigate to links added by their parents. Each link inside the kid’s browser view is presented as a card, which leads them to the approved website when clicked on.
  ‍

• If a parent uploads a YouTube video to the kid’s browser, the video itself appears inside the ASTROSAFE platform, not on the YouTube website itself, and blocks their ability to click through to other unapproved suggestions. 

• YouTube’s recommender system is completely disabled, preventing what we call “Rabbit-holing”, the thing that happens when kids follow YouTube recommendations that lead to addictive unwrapping videos 100% of the time.

• If children do find themselves navigating to unapproved links by following hyperlinks on approved pages, they reach a landing page where they can ask their parents for permission to view.
• At that point, a request is sent to the parent’s dashboard, which is hidden behind a chunky lock switch. The parents can access the dashboard by clicking the switch and inputting a security pin.

This is how V-0.1 works for parents:

• The parent dashboard for V-0.1 is equally simple. Inside of the dashboard, parents can see 3 categories of links as a stack of cards:
• Approved links which have been added by the parents
• Daily links suggested by ASTROSAFE requiring approval
• New links Requested to be added by their kids
  

• Parents can open individual links inside of the parent’s dashboard, which essentially works like a standard browser without restrictions, and manage the stack of links like so:
• Block them if deemed inappropriate (Feeding the ASTROSAFE blacklist)

• Approve and add them to the child’s browser (Feeding the ASTROSAFE whitelist)
• Finally, because the parent’s dashboard works as a standard browser, they can navigate to a link they want to add and use the big green button to add them to the kid’s dashboard on the fly.

What we discovered

Building this product was challenging, but a lot of fun! Nothing beats testing a real product with real customers, and we learned invaluable lessons for future iterations, as well as for working better and faster as a team.

‍What we learned about the product

‍We succeeded in building a system that allows ASTROSAFE as a business, and as a community, to systematically create a database of safe content, whitelisted by parents, that will become more comprehensive and detailed over time. We also developed a system that allowed children to browse the Internet inside a walled garden in which parents have full control over the content they encounter.

‍

HOWEVER, while our assumptions for V-0.1 were ideologically correct, we discovered that in practice, the experience is not what parents were after for the following reasons:

• Parents are time-poor. Our testers didn’t like the laborious process of having to approve every link. The sentiment across the board was very much “If I’m paying for it, I want you to do the work”. 
• Children would always end up having to ask for access to an unapproved link at the end of every browsing loop, resulting in parents having to constantly log-in to check and approve new links. Annoying for parents.
• The experience allowed children to browse content added by parents, but didn’t really allow them to explore new interests and topics, leading them to quickly become bored of ASTROSAFE.

What we learned about our process‍

Learning to ship a product of this ilk as a team was interesting. We developed an efficient process for designing, building, and validating our assumptions. We also learned a lot about the technologies we’re building with and sharpened our tools for future cycles, but once we shipped we realised it was still all a bit to “waterfall”. We learned that:

• We could have validated our assumptions with a clickthrough prototype built on FIGMA earlier. We would have reached the same conclusions without the hassle of actually having to build a fully functioning product.
• If we did want to build a fully functioning prototype, we could have probably repurposed a Trello board as a parent dashboard, and we could have developed a Chrome extension for the kid’s homepage, glued together with Zapier.
  

What are we building for V-0.2

‍We have changed the value proposition of the product based on what our users revealed, and are moving much faster towards something we feel is ready for public consumption. Here’s what we think is needed, and we would love to know if you agree:

• A browser and search engine for children that allows them to search the entirety of the internet, filtering results gathered from all around the web, into a safety-checked kid-friendly format.
• The results are given to children in a visually engaging format that allows them to consume text, facts, videos and images inside of ASTROSAFE, without letting them go onto the actual Internet.
• We will keep the “Rabbitholing” blocking feature for videos, and by allowing children to consume content inside of ASTROSAFE only, we will fully preserve their anonymity as they learn and discover new content.
• A more comprehensive parent’s dashboard that allows them to automatically block other apps on the device they’re using, and set usage timers for ASTROSAFE as a handy screen management tool.
• A stats board that keeps parents informed of what children are searching, with a live feed that shows them what their children are looking at, block what they don’t think is appropriate, and learn what they’re interested in.
  

Some other improvements

• We evolved our UI with a new look and feel that makes the App look fun for kids while avoiding being childish. We know from experience that kids don’t like to use baby-ish things. They want to be like the grown-ups!
• We’re doubling down on proprietary Machine Learning tools to make sure search results are as safe as possible, cutting out hate-speech keywords, inappropriate images, and more.
• For our public Beta, which will be a very official V-1.0 we’re going to develop straight for iOS, which covers over 80% of the devices in our official user-base. More platforms will come later.
  

Here’s a sneak preview of how it’s all coming together, and we will be happy to share a proper walkthrough with you if you join our mailing list!

‍