Data and moats

My foray into UGD & data moats here - hope it's of value.

We've been working on an open access psychometrics web service for self-development and recruitment purposes - that's based user generated data in psychology - for over 2.5 years now. If you're interested in this: ping me!

With the rise of open source LLMs and ChatGPT we're now believing that the secret sauce to dominating most markets is user-generated data (a data moat) and user opt-in (engagement/retention). If you collect unique data that your users create and provide killer user experiences, you'll do well.

User Generated Data

User-generated data (UGD) is a goldmine of info your users generate while interacting with your app or service. This data is a treasure trove of insights into user behavior, preferences, and needs. This data can help you can uncover trends, gain insights, and deliver personalized experiences that'll have users hooked.

User Opt-in and Activation

Opt-in mechanisms, provided you collect quality data via good UX, bring in unique data to you directly. This means giving users the power to say "yes" or "no". User activation, on the other hand, is all about getting users to generate that precious data. Communications/direct engagement, gamification, personalized content, or whatever gets users to spill the beans about their experiences and feedback will work.

Examples of Value Creation and Capture

Some companies have unique user-generated data that creates value and builds an unbreakable moat.

LinkedIn uses UGD from professional profiles, job postings, and networking activity to create a one-stop-shop for professional social networking, sales and recruitment. LinkedIn is the go-to resource for pros and employers, though legacy and archaic in many ways (and not aligned with your recruitment interests as they make money on postings and SaaS).

Airbnb relies on user-generated data: property listings, reviews, booking history to create a trusted marketplace for rentals! They focus like crazy on a fantastic UX and fostering a sense of community - they've built a brand that's bounds ahead of traditional lodging options.

Use-case: Psychometrics and Occupational Psychology data

When it comes to recruiting, user-generated data is a game-changer. Psychometric tests and occupational psychology tools, like personality assessments and aptitude tests, create data that can help recruiters single out top candidates, and at the least screen out weak fit candidates. Such processed data into a meaningful form (eg. Candidate X is very Adaptable, has high cognitive ability, and is a very high cultural fit to your team), recruiters can pinpoint candidates who are not only skilled but also a great cultural fit (if your team also provides such data), leading to teams that can last for the long-run. This is what we're working on at Gyfted - would love to chat with devs who are into psychology (ping me if so)

Value for LLMs and Companies like OpenAI and Google

In this context, UGD is like rocket fuel for LLMs and companies like OpenAI and Google. Massive amounts of UGD powers their algorithms to generate more accurate, relevant, and useful outputs. This leads to more value to users - further solidifies their competitive advantage in AI. Although my bet is open source LLMs might upend this -- the power of distributed thousands of top devs and data scientists and entrepreneurs working on their niche and unique LLMs/models will beat these corporations.

I'd say user-generated data is where it's at for devs looking to create value - ideally it's unique in-house data that gives you a moat. As tech keeps evolving, those who own not only the data but also the user experience will win.