Collective Intelligence platform: development progress, new rewards and the second public sprint
It’s time to share our progress on Collective Intelligence platform development.
Our sprints are now public, as well as our roadmap. The first public sprint took place at the end of November and, as a result of the vote, we brought five tasks into development. Here’s an outline of the progress on each track:
1. Improving incentives scheme for forecasters (first iteration)
The biggest single update to the system of incentives in the history of the Collective Intelligence platform was implemented with the November rating, granting a monthly reward to all analysts with a positive points balance. We saw an almost threefold increase in the number of active analysts and a 2.2 times increase in the number of answers per question the following month. In the December rating, twice as many analysts received prize money compared with November (8674 vs 4279). We’re looking forward to seeing the January results.
We’re happy to see that the direction we took is working, and have no plans to stop there. With the January ratings we’ll introduce another major update: the top three echelons of analysts in both crypto and traditional ratings will now be granted access to the Cindicator Bot. The top 10 analysts will receive an Expert plan for a month, second echelon analysts will be granted access to a Trader plan, and the third echelon of analysts will be able to try out the Beginner plan. We’ll prepare a detailed FAQ with instructions next week and contact eligible analysts directly once the January ratings have been delivered.
2. A reminder function for new questions — currently analysts must always have their phone on them to ensure they don’t miss new questions
One of the first features to be delivered was a new question reminder functionality. Now analysts can turn notifications for new questions in traditional markets, crypto markets, or in both categories on and off. Analysts who turned them on made 34% more forecasts than those who didn’t — since practice makes perfect, we’re counting on them to improve their analytical skills faster.
3. Improving filters and search functionality for future/past questions feed (first iteration)
Quantity does not always equal quality. After taking a look at our stats we noticed that the vast majority of our users only use filters to show them crypto market questions. Questions in this area did indeed become the most popular part of the question feed. Our previous UI, however, didn’t reflect this, offering fine-tuning for filters on particular futures or fiat currencies instead.
For our first iteration we decided to focus on cleaning up the UI and removing unnecessary features: buttons are now larger and the UI’s behaviour is more intuitive. We removed particular tickers that could be covered by the search function. As our user base grows in the future, we’ll begin to add further functionality. As of 24 January, new filters have gone live in the iOS and web applications, with an Android update due next week.
Bad organisation of past section. It would be better to group all forecasts already made by the user, separating them from past events not forecast.
The option of showing all your previously answered questions came with the new filters. Sort them by date of results to show only the ones that have their results announced and points allocated!
4. When technical work is being carried out on the platform, don’t just sign users out and kick them off the app — create a message explaining that there are technical works ongoing to improve the app, and that they will end in XX minutes.
Point taken — we now display a notification during maintenance. The message is custom: if maintenance is going to take longer than a few minutes we will provide an estimate.
5. Improving statistics screen (first iteration)
This is the task we’re working on now: there are tons of possibilities and we want to make sure we get it right. One of the recurring suggestions from our community is to add achievements and gamification elements and we decided to take our time to think it through: done well, it could be a killer feature for the app, but if it isn’t done properly a lot of development power will have been wasted.
Remember that we collect and thoroughly examine all of your suggestions on the platform sprint Trello board. The next public vote will be partly dedicated to prioritising possible statistics improvements.
Speaking of the next public vote, it will take place on 1–7 February. The process will be slightly different: we’ll group tasks into several key areas we’ll be focusing on in Q1 and we’ll ask our community to prioritise tasks within those groups. We know that there are many different ideas worth implementing on that board, but it might be helpful to focus our community on paths we are exploring at the moment, given that a general consensus has now been reached.
Interested in joining our community? Please fill out the form at https://cindicator.typeform.com/to/OCtoNW