To be or not to be an app…the tradeoffs of building on-Facebook vs off-Facebook
This blog post comes out of a Q&A thread at Quora, a cool new startup that’s all about sharing knowledge. Awesomely, hanging out on Quora is spurring a renewed interest in writing, mostly because Quora is super social and I like riffing on ideas with other people.
The question was: What are some of the cost/benefits associated with building an app in Facebook vs building a standalone site?
So at ohai! we are supporting both on-Facebook and off-Facebook experiences. We’re seeing some interesting data so far with our first MMO, City of Eternals. When people play embedded through Facebook, their usage pattern looks something like this: 5-6 minute sessions about 8-10 times a day. So they are round tripping in and out of the game world many times a day. When people play directly at the City of Eternals site, they’ll play for 20+ minute sessions 2-3 times a day.
Here’s a screenshot of the embedded, in-canvas Facebook experience:
(click through to see the larger image)
Here are screenshots of the standalone experience at CityofEternals.com:
We’re doing a few things differently in the on-boarding between the embedded app and the CityofEternals.com experience, but I don’t believe they are sufficient to explain the entirety of the difference. There are a few different hypotheses that might explain these differences more thoroughly:
- The standalone site has a theoretically better playing experience, since it’s full screen; The full screen experience leads to different types of behavioral patterns in game.
(For example: Players might be more likely to partake in longer missions, Players might be more likely to join groups and do group activities, Players might be more likely to participate in exploratory vs discrete goal oriented behaviors, etc.)
- The full screen experience has fewer distractions, leading to a different type of relationship between user and product.
(For example: Players might be more likely to become citizens[1] of the world we’ve built, rather than just tourists, who come in and out to complete short, discrete tasks.)
- However, you can’t discount the ‘form’ factor of being very proximate to the social context where people already naturally live, which provides benefits around higher conversion rates / lower barriers to entry
(For example: Facebook has done an amazing job of building a site that people log into religiously – daily, multiple times per day – it’s easier to attach a new service to that existing behavior than to ask users to form a new daily relationship with a new behavior – such as going to CityofEternals.com and logging in.
Whenever you ask someone to learn a new daily behavior, the initial pitch better be compelling – to entice people to even try it, and then the value they receive in the first 60 seconds better be outstanding, to convince them that they want to try again. Every single second in your service is a new potential conversion point. Whereas on Facebook, the value proposition for logging in every day and returning has already been well established.)
- People are used to playing other social games in quick bursts multiple times per day, so that’s the mindset with which they approach embedded games to begin with.
Again, back to the principle that it’s easier to draft off of existing, established behaviors than it is to forge new ones. (Have you tried training your cat lately?)
We’re still in our early beta days, so we’ll be collecting a lot more data to test these hypotheses over time.
Then there are the cost/benefits around business related issues:
- The standalone site has a better chance of building a strong brand relationship with customers
- With a standalone site you have more control over the user experience
- With a standalone site you have to deal with a conversion rate hit of the first ‘try’
- With a standalone site you have less optimal access to user communication flows that are already prevalent in the on-Facebook experience.
- With a standalone site, you may be subjected to less volatility in your base platform (since the Facebook APIs have some external volatility that you can’t control)
- With a standalone site, you’re going to have to work harder at facilitating daily usage patterns
[1] Citizen (noun): I use the word citizen a lot when talking about the social web. A citizen is someone who is deeply engaged; someone who has a strong emotional connection with the community and the social software they’re participating in. Someone who is invested in the health, stability and future of the service they’re using. There are different levels of citizenship – some may be passive participants, some may become active policy makers. But the hallmark is some level of emotional identification with the community and environment. Our goal is to turn our users into citizens.
