It’s been a rough quarter for altcoins and ICOs. The euphoria of 2017 feels like a distant memory.
So how’s CoinFi doing?
The good news is that we’re focused on the long-term, so a crypto winter (if indeed winter is here) doesn’t affect our plans.
However, a bear market like this is as good a time as any to re-examine some fundamental assumptions.
Let’s take a look at our roadmap for 2018Q3, 2018Q4, and 2019Q1
- Custom Newsfeed with news type classification
- Improved news and content aggregation overlaid on price charts per cryptocurrency
- Improved custom watchlist
- Advanced ICO Metrics (token retention rate, token distribution graph)
- Improved data pipeline
- Market data warehouse
- Accept staked tokens for advanced functionality
- Coin Screeners
- CoinFi Events Calendar
- Initial beta versions of trading signals
- Prototype of machine learning algorithms for content classification, quality assurance, and fraud detection
- Token-incentivized crowdsourcing
- Social functionality including upvote/downvote and sharing
- Revamped homepage
- Research & Analysis
2019Q1 – Trading Signals
- Abnormal volatility detection
- Abnormal price movement detection
- Abnormal volume detection
- Increased social mention detection
- Best price exchange identification
- Best currency pair identification
- Source code release detection
- Launch premium subscription for trading signals
- Accept community contributed trading signals
That’s a lot of stuff!
Of course, we were the ones who came up with this long list of features in the first place. But why does this list exist?
In Q2, we addressed the mistake of trying to build too much, with too small a team.
In Q3, we’re addressing the mistake of following a roadmap created in late 2017 – an entirely different era of the crypto markets.
Throwing out the roadmap… kinda
Going forward, we’re not going to make anymore roadmap pronouncements.
The items on our current roadmap should be seen as a collection of ideas we’re validating and testing (among other ideas that we haven’t announced yet), rather than a concrete list of features we’re building.
For many startups, the first idea they raise funding on isn’t what sets them on the path to hockey stick growth:
- AirBnB started by literally putting air beds in people’s homes and serving them breakfast. The founders researched the number of air beds sold every year, because that was the market size they projected for their website, AirBedandBreakfast.com.
- Instagram started as a location-based social app called Burbn. They only relaunched as Instagram, the massively popular photo-sharing app we’ve all come to know, when they realized photo sharing was the only function in their app that anyone was using.
- Twitch, the live streaming gaming site acquired by Amazon, started as Justin.tv, which started as an online reality TV show about one of the cofounder’s lives.
Yet ICO projects are held accountable to a 1+ year long roadmap based on a whitepaper that’s been frozen in time. Although we’re still excited about what the ICO model has done to democratize investing, we think this part of the model is broken for many projects.
An academic, first principles approach may work for projects working towards groundbreaking technological breakthroughs, but CoinFi is a software startup aiming to serve the crypto investing market. We need to adapt to current market demand, not the market from a year ago when we wrote the whitepaper.
Early stage startups don’t succeed because they stick to the same idea they pitched to raise funding. The team stays lean and works towards product market fit, even if it means taking a detour here or a pivot there.
Then, only when they’re fortunate enough to find product market fit, do they scale.
While we’re still focused on the same mission of bringing Wall Street calibre market intelligence to the crypto markets, the way that happens may not look like the roadmap we planned out a year ago. It may not even look like the roadmap we planned 3 months ago.
We’ve also learned a thing or two in the past year…
In addition to the ideas on the roadmap, we’re testing new ideas as well. Ideas that come up as we’ve been travelling down this current path. The market is constantly changing and as we put prototypes in front of users and look at what’s possible in these markets, as a team, we get a better sense of what is and isn’t working.
Nothing has fundamentally changed. Our mission is the same – to be the leading market intelligence platform for the cryptocurrency markets. We’re focused on one thing at the moment, building a product that will serve crypto investors well.
In this market, the only projects that will thrive are those that build something that serve a real user need, in the current market.
So what were we up to?
We spent time in Q3 exploring, researching and testing new ideas – while also working on roadmap features. We took time to talk to users and adjust what we’re building to a new market that’s a full year removed from when we published our whitepaper.
We took the new data we’ve been ingesting and applied our data resources to identifying new trading signals that would serve crypto investors in late 2018 and beyond, and we’ve seen promising developments in this area.
We make no predictions beyond this quarter on what we’re building, and we will likely make adjustments as we flesh out requirements and put prototypes in front of real users.
Here is the current status of what we’re working on and some of what you can expect us to work on in Q4:
- CoinFi News public release
- Our coin pages still need a number of improvements before they’re useful to you. It’s something we should have addressed months ago. We’re working to fix this.
- We’re continuing to address data integrity issues.
- We’re working on a trading signals beta. The reason we haven’t been publishing research & analysis pieces lately is because our data scientists and traders are focused on uncovering new trading signals. More details will be coming soon.
- We’ll release the first use case for staking and we’re exploring additional utility for the token model
- We’ve been developing and testing trading algo infrastructure. We can’t release any details on this yet, but it is something we’re working on.
We realize without a long, concrete roadmap to follow, you may have some concerns.
But we don’t think it’s useful to you as a platform user or a token holder if we forego new opportunities that will benefit token holders in order to release features that aren’t viable in the current market.
However, we’ll continue to share what we’re up to and offer as much transparency as we can (without overly tipping our hand) in these monthly updates.
You’ll know when we make progress, and you’ll know when we make adjustments. We’ll take into account your feedback through Telegram and 1 on 1 user interviews, but do note that there are a hundred different voices telling us a hundred different things – our job is to act as curators and incorporate your ideas into our own model.
We do appreciate your constructive feedback and we’ll continue to seek it out and incorporate into what we’re building. But we can’t accommodate every opinion.
With that in mind, let’s take a look at what was on the Q3 roadmap: what we did, and what got deprioritized as we tested prototypes, gathered user data, and explored new, promising ideas.
Custom Newsfeed with news type classification
These features are being used in the CoinFi News beta. We had over 10 categories that were being tagged through a combination of moderators and machine learning.
After looking at what beta testers were actually using, we pared back the number of categories, removing a number of underused categories and introducing some new ones based on our user testing.
Improved custom watchlist
The watchlist has been significantly improved during beta testing of CoinFi News, and we’re working on integrating the watchlist with other market data.
Advanced ICO Metrics (token retention rate, token distribution graph)
We have a proof of concept live on ERC-20 token pages. See BNB as an example.
The infrastructure for Advanced Token Metrics is in place. One of our data engineers, Evgeny Medvedev, is a leading expert in blockchain ETL ingestion. He even published a piece on the Google Cloud blog showcasing the open source Ethereum blockchain dataset he helped create.
We’re taking our blockchain data and building useful metrics that can help you generate actionable trading ideas and signals that can give you an edge. Our data scientists and trading team are currently focused on this and will continue to prioritize this for Q4 via our trading signals beta.
Accept staked tokens for advanced functionality
We have a beta planned that will incorporate staking functionality, which you’ll be hearing more about shortly!
We’re currently doing user testing on CoinFi screeners to make your crypto research workflow easier.
Improved data pipeline
As mentioned in our last monthly update, we rolled out a new Coin database. However, the way this data is displayed still needs a fair amount of work before it’s useful to users. We’re working on improving the Coin pages to better serve crypto investors like yourselves.
Based on our user testing and shifting priorities, the following features have been deprioritized:
Improved news and content aggregation overlaid on price charts per cryptocurrency
While many users have indicated that they found this feature useful, the implementation itself still needed improvement. Users indicated that this was a nice to have feature, not a must-have. It’s still something we’d like to release but its not a high priority, based on your feedback.
If you feel strongly that this is a feature that you must-have in your workflow, feel free to reply to this email with an explanation of why this is essential to your workflow and we’ll take it into account.
Market data warehouse
Our data engineering work has shifted towards uncovering actionable trading signals and metrics from blockchain data, so this function has been deprioritized for the moment.
CoinFi Events Calendar
Based on user interviews, it seems like many of you feel that in a bear market, events no longer drive demand and trading opportunities the way they used to. Based on our data, we’ve found the same thing.
We did build an events calendar pre-populated with existing data feeds, with the plan to incorporate our crowdsourced curator functionality into the calendar. But because there were simply fewer market moving events in the bear market, this feature has been deprioritized.
Although we will still incorporate some of the events data into other parts of the platform, at the moment we’re not working on a standalone, crowdsourced events calendar.
The end of a quarter is a good time for reflection.
While we hit some of the quarterly targets on our roadmap, aiming to fulfill a roadmap planned a year in advance isn’t helping us achieve the goal of building the leading market intelligence platform for cryptocurrency.
In 2017 it made sense to release a retail arbitrage bot. In 2018, that’s a game for institutional players.
We’re approaching this project with a fresh slate. In this market, we believe it’s the teams that keep grinding, learn from their mistakes, and make adjustments that will survive and thrive.
Whether you agree with this decision or not (and we’re sure you’ll make your disagreement heard on Telegram), you can rest assured that we’re making decisions that we feel are in the best long term interests of the project and token holders.
There’s 1 more quarter in 2018, and we’ll aim to make it a good one.