Hacker News new | past | comments | ask | show | jobs | submit login
Ninite: How we got 18,000 beta users in 4 weeks (runitback.tumblr.com)
140 points by swies on Oct 26, 2009 | hide | past | favorite | 23 comments

I'm currently working on this problem right now with my company imVOX (http://imvox.com/trial.php). We're trying to finish up our client and slowly pull out of the real 'testing' phase of beta and more into an RC status and have load tests, etc.

Yet getting good beta testers is hard. Many people sign up, use it once, hit a bug and then never return. Understandable, but even emailing people (personally) hasn't been all that effective in getting people back consistently. I probably have 10 people that I can keep going to and know that they will use it again and again and provide feedback.

I've yet to go on a strong blog push yet, but thats probably my next approach.

Other methods to find bloggers: YouTube and Twitter Search. I've also been looking through friend's shared stuff on Google Reader occasionally to mark off ones to contact.

One twist that I have as problematic is that I don't have and 'special codes' to hand out to people currently. We had a system setup like that a while ago, but I think it was slowing downloads a bit so now its just a download link and you register for a (free) account in-client.

In addition for ideas of how to get more testers via blogging (commenting is important I know too), anyone trying it out would be really appreciated. I'm going to have them push the newest build to our download and auto-update servers later today.

The target market seems to be tech-savvy. Getting on blogs (with a relevant audience) is much more challenging for b2b and most consumer apps.

Getting them to read the blogs - yes. But getting on the blogs is easier, if anything, because many b2b-writer need content. Consumer stuff is different, but doable. What about giving a limited amount of accounts for free, to the bloggers. Not the high-profile ones that get tons of stuff, but the "middle class" so to speak.

Can you give more details on how many blogs you've contacted? How many covered you? How did you phrase your email to make it interesting to the bloggers?

Edit: Also, how you computed readership numbers for a blog?

I'm not the author, but my own advice is: Personal, relevant, five sentences max. The same as any email you'd send to anyone else you've never met.

As for computing the traffic numbers, first look to see if they have a media kit on their page or an RSS subscriber badge. Then look at the number of inbound links on Technorati and Google. Then check sites like Alexa and Compete. Also check the number of comments on the site, if there are any. Pay attention to the quality of the posts themselves. Are they writing original ideas, like Fred Wilson, or are they mostly regurgitating the results of PEW research studies? Have they won any awards? And lastly, take into account what platform they're on. A Typepad blog is going to have different traffic patterns than a Tumblr blog, simply because of the ecology of the platform and the userbase.

Using this you should be able to get a pretty decent estimate of the readership, especially once you have 10 or 20 blogs where you have exact data to use as a benchmark.

We contacted about 30, resulting in around 7 hits. Some of our biggest hits were secondary though, larger blogs would see us on smaller ones and contact us for a demo code.

Our pitch email subject line was a brief description of the product then "- Free demo codes for your readers". The body had a similar format: what we do, here's your code, contact us with any questions, thanks.

Lots of sites have feedburner badges to give subscriber counts. We found one blog that was clearly faking that though. Their traffic to us was way lower than an 80k subscriber blog would send.

The number of comments on posts is a good indicator too. Anything greater than 0 means they've got a fairly big audience.

"... We contacted about 30, resulting in around 7 hits. Some of our biggest hits were secondary though, larger blogs would see us on smaller ones and contact us for a demo code. ..."

Nice hack and something I appreciate because it's such a simple thing to do.

"... Lots of sites have feedburner badges to give subscriber counts. ..."

Did you try to obtain the stats from each site?

Pardon the expression, but people are complete suckers for sort of thing.

Given two similar services, one with a limit on registration and one without, you can bet that the one with the limit will garner some sort of popularity just by virtue of being "exclusive."

This doesn't mean it's a bad marketing tactic. On the contrary, it's obviously excellent for exposure... but if it ever becomes a popular service, saying "I was there first" is a huge deal for people. For reference: Lines outside movie theaters days in advance for the Star Wars movies, video game midnight release parties, and 8 hour lines outside Apple stores for a phone.

I'd guess that using signup codes is not about merely creating the illusion of exclusivity. I think the real reason behind using signup codes is so you can track your marketing efforts: where did I spend time/money and how effective was it? Without it you are flying blind.

I think it's important to note that immediate media coverage and an immediate surge in signups is not always desirable for something just released and it's of course not always a prerequisite for success in the long run.

It's your trending over time that really counts.

I'd just like to say that you guys couldn't have launched in a more perfect time. I browsed your site when I saw it linked to on HN and didn't think much of it. My mom came home the following day with a new laptop and asked me to install a bunch of software. I quickly remembered Ninite and had most of her stuff installed in no time. It was really great.

I just have to say I would use your service...

Definitely going to use this strategy in the future, but hope the reach of this post doesn't flood blogger's inboxes with similar schemes.

The approach makes sense, but something about spamming tons of blog owners turns me off.

I've used a similar approach, but only with the blogs that I had either made personal contact with previously or that I frequented.

On another note, your blogroll/Reader subscriptions are a great place to start if you follow other bloggers in your industry. Being an active member of the blogosphere also helps give you some clout.

As the owner of a blog that's #1 in its niche in the range of those described (10k-50k subscribers), I'm not against getting unsolicited messages at all if they're relevant and my readers would find it interesting. Indeed, I don't receive enough mail like that.. :)

fair enough. You've just given me the confidence to contact as many bloggers as I can (as long as they specifically pertain to our market).


Be careful though. It's not as simple as just contacting them all. You really, really, really, really need a story that's of the sort they actually publish.

I get relevant press releases from time to time that are in my industry/field but are so "corporatey" that there's little my readers would really care about. Twist whatever you have to say into a story that fits the mold of what the blog already posts - it might take reading their entire front page to get a feel for this ;-)

Would you mind pasting the link of one of the blogs? I am interested to see how they are framing the promotion.

Here's one: http://www.simplehelp.net/2009/10/08/how-to-install-the-most...

The general pattern was mention the private code as a teaser, describe/review the service, then give the details for signing up with the code.

Have you given up on the Baseshield sandbox concept? I don't see it referenced from Ninite (I'm guessing Baseshield isn't complete?)

good to see someone moving on this idea. I talked with a few people about this exact service a year ago and the general consensus was that if it got big you'd get slammed for violating EULA's. best of luck.

Getting featured on Lifehacker.com didn't hurt, I suppose..

That's true, but Lifehacker didn't cover us until after we launched publicly. We pitched them with a demo code towards the end of our beta and never heard anything back. Judging from the traffic they've sent recently we probably could have gotten 18k beta signups from them alone.

The biggest hits for our beta were sites we never heard of that found us on smaller blogs and reached out to us. Sites like http://www.dinside.no http://www.genbeta.com and http://www.walla.co.il.

I should write about our experience with non-english blogs later.

Congrats! That's great. Thanks for sharing your success!

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact