Ask HN: Has Facebook Connect helped your user growth?

[yeti]

We're thinking about adding it to our web app to reduce registration barrier, but it is going to be a bit of work to integrate into our account system, which will mean delaying other features.

For those who have implemented FC, has it helped convert a lot more of your site visitors, and has it helped growth a lot through the mini-feed messages of your users?

Any feedback or stats would be great..

[mattmaroon]

The general impression so far seems to be that FC doesn't so much net you new users as it does increase engagement among current ones (if integrated well). That isn't surprising given what I've seen running my Facebook app (which is now at about 50k MAU after 2 months) and what I've seen of sites using FC.

Facebook Platform (and presumably Connect) is great for virality, but haphazardly slapping it onto something that wasn't designed for its strengths and weaknesses from the beginning nets you little. To use it properly, you have to design your app or site from the ground up around the platform. Otherwise you're wasting your time.

I haven't looked into connect much yet, but I'm guessing that if it allows for solid virality (invites, notifications, profile boxes, mini feed entries, etc.) people just haven't quite built the app that takes advantage of it yet. They're just adding it onto their blogs and social news sites and hoping, and unsurprisingly they're not getting much in return. Look through the portfolios of contractors who build Facebook apps and you'll see a lot of that. Some newspaper thinks they can just stick sports scores in a Facebook app and it will go viral. Those apps always have 12 active users.

So I guess my advice would be don't do it half-baked, or you won't get much from it. If you're going to integrate it, rework your entire product to use the viral hooks you get from it. It doesn't seem to be that much easier for a user than a streamlined signup process anyway.

[yeti → mattmaroon]

good suggestion, agree initially we'd be looking at it more for getting friends of our already active users aware of it, and then if they join, they can see on facebook the updates off site which will drive return visits. so it's incremental off existing users rater than driving huge viral growth of random people to our site at the start at least

[travism]

I just finished an FC implementation on a site I'm working on. The site still has local accounts, but users can now connect their local accounts to their facebook account Facebook or just login through Facebook (which actually creates a local account).

Beware: integrating with FC turned out to be a surprisingly large undertaking, although lots of that effort was spent a) trying to assemble an understanding from the FC docs about how the thing actually works and b) figuring out how to handle all of the weird cases (and the common cases, for that matter).

The FC developer docs are not all that coherent. The wiki seems to cover most of what you'd need to know, but lots of the information is on pages only linked to from obscure other pages, and there's not a good big-picture technical explanation. Even worse, much of the wiki is locked down so you can't fix it.

And there are a bunch of special cases to handle.. what if a user is logged into a facebook-linked local account on your site but is logged in as the wrong facebook user? Or what if they're logged into a facebook-connected local account and not logged into facebook? If you make a mistake, Facebook doesn't usually provide a helpful error message, and their javascript is all compressed, so it's a pain to try to use firebug.

If anybody's interested, my site's at http://mushpot.net. If anybody's got questions about the implementation, feel free to post them on mushpot and I'll try to respond.

[yeti → travism]

thank you, that's helpful. will ask more..

[dpeq]

Very theoretical responses - I suppose Yeti had those thoughts before he posted. I think he is looking for something like "Yesss, our signup rate went up 50%" or "Nope, not really a difference".

[yeti → dpeq]

yes, exactly :)

It should give an improvement in registration rate as probably half our prospective/users are on facebook also (the other half are in other countries where facebook isn't so popular) but what I'm trying to get a handle on is how much it will drive new visitors based on friends of users seeing mini-feed updates back at facebook.

Any real world figures to share?

[markessien]

You should know this by using mathematics. Do the following:

1. What age range is your primary consumer?

2. What country is your primary consumer?

3. What is the intersection between facebook users and the two questions above?

4. Based on whatever statistics you can find - how many users get turned away because of sign-up problems?

5. How many consumers will your initial marketing push reach - i.e, the consumers you hope to gain in the period when you don't have hard feature choices to make?

Intersect those 5 criteria and estimate what the benefit of facebook connect will be. You may want to contrast it with the benefits of the other features. Understand your numbers before spending time on stuff.

[yeti → markessien]

1. 13-20yo, 80% girls 2. USA, Hong Kong, Netherlands, UK, Belgium are top markets now 3. Roughly 50% 4. Roughly 60% 5. this is the tough part.

We have features planned that could improve stickiness and make it more likely visitors will enjoy and return and in time tell their friends. But doing those features won't help get more new registered users in the short term. So I'm still debating whether to add FB Connect first, get those users then add features, or make sure the user experience is great before adding FB Connect and driving traffic (if it really will help a lot)..

any practical suggestions would be great..

[pclark]

just make users want to register and then make registration really really easy - allow them to trial your service before registering, allow them to register without having to verify their email.

[yeti → pclark]

always good suggestion... well our current front page is pretty good to grab interest, but our registration flow has a lot of room to improve..

especially coz we require a plugin to use the site. we're working on that.

still thinking about whether to also drop the email/activation requirement, and make it optional to further reduce the barrier

[pclark → yeti]

whats the link to your site?

[xtimesninety]

One of my projects use it (http://eatmycharts.com).. I'd say it removed the registration barrier, but that's about it. People will come back for the service.

I can't track which users publish to their mini feed (is there a way to tell? you ask the user's permission and there's no javascript callback to know if the user approved on publishing). Most of my users that are in my facebook network prefer not to publish to their mini feed.

I myself don't like to use my Facebook account on other websites (unless I really want to use my identity on purpose). That's why on my website, I automatically assign a nick to the user (which they can change), so their full name won't appear on public.

Overall, I don't think FC helped a lot in my case.

[yeti → xtimesninety]

Good suggestion, yea I agree.. I would also make it so real name is hidden, they pick their own nick or if they are an existing user just link to that

[vaksel]

most people I know don't use it. Top 3 reasons seem to be:

a) Fear of giving away the login info to third party sites b) Privacy concerns, with every account on the internet being linked to them c) Too used to quickly register with BS info(name: AA), so they don't feel like they should bother with FC

Actually if I think about it...I don't think I've ever heard anyone praise Facebook Connect.

[chris11 → vaksel]

One thing I really like about the idea of Facebook Connect is that I don't have to manage multiple accounts. I usually only register an account on a website if I can see myself using it semi-regularly. That said, like you mentioned, I don't like having it tied to my real name, and Facebook account, so I've never used it. So if I want access to the member's area I usually just check bugmenot.com to see if they have account info.

So I guess what I'm saying is that if you want an open id system on your site, FC would be good because it has such a large userbase. On the other hand I don't know many people that use any open id system, so integrating with FC might be a waste of time.

[yeti → vaksel]

thanks for that, agree with u actually there is a class of user who would be put off by FC on a site. So we would only offer it as an option, in addition to our current account system, not the only choice..

[ryuio]

check out meetupdiary.com on alexa - the spike is due to connect.