Thoughts on Augur, the decentralized, blockchain prediction market
By Adam Siegel on August 13, 2018
I've been watching for the last few years the announcement, long delay, then recent launch of Augur, the decentralized, blockchain-based prediction market.
First off, despite it taking a long time to launch, Augur finally did, and kudos to them for doing so. Building a prediction market from scratch is not easy, building a prediction market on top of immature technology is harder, and building a prediction market to be completely decentralized with smart contracts is downright scary. This must have been a herculean, complex effort, and my hat is off to the Augur team who made it to launch.
I've played around with Augur and this was my initial reaction, as
recently posted in a discussion thread about the platform:
I was excited to use Augur when it launched. This was to be my first real DApp experience. My first step was to fund a wallet. That took 6 days for Coinbase to do whatever it was doing via MetaMask - and I have a Coinbase account already. The application took several minutes to load, at least the few times I went in. The initial user experience was not easy to comprehend for someone not steeped in crypto-speak which of course is independent of needing to have at least a basic understanding of trading. The mechanism being used is a double auction which requires a lot of liquidity to be interesting. Everyone thought Intrade was so great, but it was basically great every 2 or 4 years during elections, otherwise liquidity was pretty low. When I went in I saw tons of duplicate questions or poorly structured questions that clearly wouldn't have an easy time being resolved for a reliable outcome.
There's a reason prediction markets haven't really taken off in the public domain in a hugely scalable way, and it's not because people before Augur were incompetent, there weren't use cases, or because regulation/centralization. It's because the concept is non-trivial to master, questions need to be well structured AND interesting, and there often is not an immediate payout/gratification which makes it "boring" or forgettable. DApp or not, these problems are still going to have to be reconciled and they're only made worse by another layer of complexity that is crypto...
Let's remember the Augur experience available to the public is an MVP, and let's also leave any scary regulatory action aside (at least here in the U.S.) and imagine the Augur application experience a year from now. Performance will be improved, the UX will be better, and the questions will be higher quality as protocols, processes, and people's experience playing various roles matures.
Unfortunately, none of that will matter for driving adoption.
Instead, I believe the application - in the form of a prediction market as destination - has a low ceiling and may never get above a few hundred to low thousands of users. I was 0% surprised to read usage has been dropping since launch, not going up. For the last 12 years, first as Inkling Markets, and now Cultivate, In the prediction markets we've run, whether they're internal or external, the exact same usage pattern plays out, every single time. After initial excitement, usage decreases until you're left with a core group of users you can hopefully foster and grow.
That small group in internal corporate applications can still be incredibly powerful, but obviously that's not what Augur is going for. It wants scale. And to scale, Augur will only be successful if it can foster partnerships to build it in to existing experiences - like using a likert scale to agree or disagree with the prediction made in a news editorial (and being asked to update that prediction every once in awhile via text message as new information emerges,) or checking a box when you purchase movie tickets in your Fandango app to say "I think this movie will achieve > xx% score on metacritic" and a trade is made for you in the background.
People should never know they're using a prediction market, never know they're using a DApp, and never be exposed to the complex machinery behind the scenes. All someone should need to understand is that they're responding to a question they're being asked and getting paid in some relevant currency like the loyalty points they already collect and know in whatever app they're in.
Having REP/Augur, or any of the other blockchain prediction markets for that matter, attain some level of brand awareness by trying to create a prediction market destination site is too big a lift and will likely fail. Instead, Augur should strive to be the platform no one has ever heard of, but everyone interacts with.