In a nutshell, Quora isn’t a complex application (purely speaking of its feature set). Building a pet Quora “clone” at home is possible in the sense of creating an app supporting:
- User sign-ups and profiles (including messages)
- Blog (an incredibly simple one)
- Activities with questions and users – following, blocking, etc.
- Keeping stats for views, edit log.
- Content and profile feeds
You can expand on that and add some of the recently introduced features (like Links), additional interactions between users and questions/answers (like “Answer later”, “Request an answer”, “Bookmark”) – and it’s still technically feasible.
What Makes Quora Challenging from a Technical Standpoint?
Handling massive volumes of data and traffic. Quora serves about 200,000,000 monthly users browsing across hundreds of thousands of topics and tons of questions and answers (probably in the billions).
Being able to search/update and render the corresponding resources, within a few seconds, for tens of thousands of concurrent visitors at a time requires not only an expensive and rock-solid server infrastructure, but a collaborative effort in scalability from an engineering standpoint.
Relevant content. Most people hang out on Quora because of its home feed, the email digest, and the top answers across each and every topic. Questions containing dozens or 100+ answers are also structured in a sensible manner thanks to an algorithm sorting the answers based on different criteria (number of views, upvotes, previous activity from the author in this topic and many more).
Machine learning is the main engine that runs that powerhouse. Merely listing “the latest answers/questions” will be a hit-and-miss game. The majority of the users will abandon the platform if it gets extremely hard to find valuable content or answers relevant to their favorite topics.
How About From A Business Standpoint?
From a business standpoint, the traffic and user base is also relevant to Adam D’Angelo’s background (a former CTO of Facebook), the prominent presence of the core team in the Bay Area, and the lack of a relevant and serious Q&A platform (compared to Yahoo Answers and other alternatives).
The media team is immensely helpful from a promotional standpoint, feeding popular outlets with high quality content and additionally boosting the platform’s presence on an ongoing basis.
If you want to build the core feature set of Quora, you can probably do it yourself. But that won’t scale past a certain volume of traffic or data and won’t provide the same feed quality (which is constantly calibrated thanks to users downvoting and upvoting questions and answers).
The same thing goes for the most popular platforms out there – Facebook, Uber, Airbnb, Instagram – a handful of them haven’t anything “groundbreaking” other than AI. Furthermore, scaling massive platforms certainly requires a large team of highly experienced engineers.