57 YC startups' Twitter follower growth over time

by shazow • 15 years ago

96 42 15 years ago

shazow11• 15 years ago

Any corrections/requests welcome. If I'm missing anyone, I'll be adding new accounts tomorrow, just toss them in this thread.

swies• 15 years ago

Could you add @ninite (YC W08). We just got the name a couple weeks ago.

DevX101• 15 years ago

How does the process of getting a taken name unfold?

biot• 15 years ago

How about over the last 12 months? Many of those companies are old-timers and their last 30 days growth is almost a flatline.

shazow11• 15 years ago

Check back in 11 months. :) I'm going to keep this showcase running and try to improve the interface to visualize long term change better.

modsearch2• 15 years ago

we'd love to see how @listia does over time (S09)

tel2• 15 years ago

Pet peeve on sparklines: they're clean because they get rid of chartjunk, but they're also pretty uninformative without context. Right now, they add almost nothing to the presentation ("@dropbox hasn't changed <much> since <some time ago>, but this other guy's popularity is <swinging wildly/pretty small and noisy>").

Tufte originally intended them to be inserted into text and contextualized by it. As an extension, he suggested adding small indicators of scale and position as colored dots for example.

Either that or just lose them and graph the relative results of each of the contenders. It all depends on what kind of story you want to tell.

Edit: just noticed that you mention that this information is tracked over the last 30 days, but since that's not localized to the graphs, it's pretty easy to not notice it.

shazow11• 15 years ago

Fair point. I couldn't come up with a better way to display 57 charts on one page which give some kind of interesting information without it being a total mess.

The information is actually tracked indefinitely from when I started monitoring it, and I have daily changes for more than 30 days for each of these accounts at this point. I also have an interface for browsing each one of them in a more granular fashion so that the chartjunk is visible (I can give you access to this if you're interested). Here's Posterous for example: http://goo.gl/QSGp4

Do you have any ideas for how to better present such a comparison between so many data sources? Keeping in mind that scale varies hugely between the smallest and largest.

Edit: To me, the most interesting thing to look at in this showcase is the "shape" of the sparkline (which indicates stability) combined with the percent delta change next to it (click on it to get absolute numbers).

simonw• 15 years ago

Could you add @lanyrd ?

snprbob86• 15 years ago

Cool! Would love the ability to click on the headers to sort by followers, following, and tweets.

ig1• 15 years ago

I've got most of the YC twitter accounts on (upto S10) on a list here:

http://twitter.com/imranghory/ycombinator/members

shazow11• 15 years ago

Ah thanks. I've been trying to weed out the ones with inactive accounts, so that's probably the main reason why some are missing.

eddylu• 15 years ago

@grubwithus please!

Alex3917• 15 years ago

@LaunchHear

@Swagapalooza

(Swagapalooza is our event series, LaunchHear is the actual name of the YC w2010 startup.)

ceelee• 15 years ago

Can you also add @tellfi (YC W11)

camdykeman• 15 years ago

Thanks for posting this. I've been trying to get an idea of just how useful/helpful YC is in creating a successful company but evidence in of this impact in terms of social media has been hard to come by. any other sources you'd suggest?

shazow11• 15 years ago

I guess it depends on how you define a successful company. In the capitalist sense, something like http://yclist.com/ is probably most poignant.

I wouldn't say that YC has a huge impact in terms of social media; they definitely help with getting press mentions, but getting engaged subscribers is really up to the founders.

Twitter is not relevant to all companies, but it's obvious that many are leveraging it very effectively. I wonder if investors would be interested in a dashboard like this for all their portfolio companies. The correlation to "real growth" is probably small, but still better than nothing.

tomdeal• 15 years ago

Interesting data, but unfortunately nothing you can see a trend in (other than the most well known companies have the most followers). One good advice you can take from this data: If you want to have many followers, you have to tweet more than once a day.

immad3• 15 years ago

Heyzap is number 8, woohoo.

shazow11• 15 years ago

I'm thinking I need to add big fat position numbers to each row and start tracking change in that. :P

immad3• 15 years ago

Being able to count from the top is part of the challenge.

sahillavingia• 15 years ago

9, actually. maybe there should be position numbers :)

matclayton• 15 years ago

Congrats. Mixcloud would be coming in @4 although not eligible :D

jellicle• 15 years ago

I don't understand those charts at all. Some charts show growth (right side of line is higher than left side of line) but the growth figure is negative. Some charts show decline, but the growth figure is highly positive. I have no idea what those charts are supposed to be illustrating - seems to be random noise.

joelhaasnoot• 15 years ago

And Dropbox is pretty much flat. I'm assuming that means data wasn't tracked between 0 and the 200,000 members they now have.

shazow11• 15 years ago

The data is only being displayed for the past 30 days. Keep in mind, just because it's flat doesn't mean it's 0. It could be constant at a higher number of growth. My apologies for not showing that effectively.

shazow11• 15 years ago

The charts are showing growth over time, as in delta change on any given day from the previous day.

Imagine you had a normal chart where each point is the absolute number of followers, but you take the derivative of that function (or use the deltas between days instead).

dekz• 15 years ago

I believe it's all constant growth, spike aren't indicating the growth, just a spike in growth. If that makes any sense. So if it's flat for example it's 100/day, if it's going up it's probably a spike over an average day, say 200/day for that day. I think the give away is that people don't seem to unfollow on twitter.

khangtoh• 15 years ago

So what's the value for these charts? Does it co-relate to the fact that the more followers the more successful the YC company is?

I really don't get the point of this post, and I'm commenting because I care about knowing the point of this post.

shazow11• 15 years ago

There's no specific point or statement, it's just "interesting data" that you can draw your own conclusions from.

For one, it's a nice overview of which YC companies are active on Twitter and which are not so much. Which are recently growing a lot, and which are stagnating.

It's also interesting to compare the number of followers to the number of following/tweets. We can see that Dropbox has very disproportional numbers, turns out this is because following @Dropbox on Twitter gives you a reward on their service.

ivankirigin3• 15 years ago

The one for Dropbox is wrong, if only that it misses the larger history and trends.

We doubled followers in a weekend by testing but not launching http://dropbox.com/free

Then we did Dropquest, and got another bump.

If we add twitter following to the getting-started quest http://dropbox.com/gs we could blow those changes out of the water.

A graph with context: https://www.dropbox.com/s/mwmyb8zmt4kufis/at_dropbox.png

shazow11• 15 years ago

Emphasis on "past 30 days". Your graph with context shows most activity before January, whereas mine only begins in February.

I wish I had the foresight to start monitoring earlier.

ivankirigin3• 15 years ago

You might be able to find other data sources, scraping http://twittercounter.com/ for example

Also, API endpoints for friendship might have a follower_since date, and you can approximate the history. In so far as unfollowing is rare compared to following, it would be accurate.

http://dev.twitter.com/doc/get/friendships/show maybe?

immad3• 15 years ago

That graph says Download at Dropbox. Secret info? :P

ivankirigin3• 15 years ago

I should have called it all_our_stats_when_you_work_here.png

http://www.dropbox.com/jobs#ABC

jdp23• 15 years ago

Interesting how asymmetric the ratios are for almost all YC companies: they're happy to have people follow them, but not in general interested in following back.

afinlayson• 15 years ago

Awesome Andrey keep up the good work!

57 YC startups' Twitter follower growth over time

Related Stories