TE
TechEcho
Home24h TopNewestBestAskShowJobs
GitHubTwitter
Home

TechEcho

A tech news platform built with Next.js, providing global tech news and discussions.

GitHubTwitter

Home

HomeNewestBestAskShowJobs

Resources

HackerNews APIOriginal HackerNewsNext.js

© 2025 TechEcho. All rights reserved.

Are we in another bubble? Empirical data from CrunchBase

61 pointsby kevenabout 14 years ago

8 comments

daniel_levineabout 14 years ago
I worked at TechCrunch for a year focusing mainly on CrunchBase data analysis.<p>There are a number of problems with your analysis:<p>I suspect you are using founding year. Unfortunately there is a tremendous lag between when a company is founded and when it is entered into CrunchBase. The same is true of funding data in particular where we saw only around 20% of fundings within a quarter of happening and only about 70% a year out. That is due to CrunchBase's continued growth (it's much better known now) as well as a natural reporting lag.<p>Second, CrunchBase is a very new product and as it turns out data is only reliable as far back as 2007 and even that took a lot of work. Some time has been spent pushing to get more accurate data further back but it is scattershot at best.<p>(Possibly, can't tell) CrunchBase investments are stored in a number of currencies, did you make sure to recalculate them? Yen can really cause problems :)<p>Lastly, your NASDAQ chart is from 94 - 2005 which never overlaps with reliable CrunchBase data even by your own admission. I suspect that graph will be a bit more telling and worrisome potentially: <a href="http://www.google.com//finance?chdnp=1&#38;chdd=1&#38;chds=1&#38;chdv=1&#38;chvs=maximized&#38;chdeh=0&#38;chfdeh=0&#38;chdet=1299226777603&#38;chddm=1699990&#38;chls=IntervalBasedLine&#38;q=INDEXNASDAQ:.IXIC&#38;ntsp=0" rel="nofollow">http://www.google.com//finance?chdnp=1&#38;chdd=1&#38;chds=1...</a>.<p>I do not necessarily think we are in a bubble and I am happy to see people diving in on data I just wanted to point these things out as it would be irresponsible not to.
评论 #2287633 未加载
andrazabout 14 years ago
Take this with a grain of salt.<p>CrunchBase does not by any means offer stable and comprehensive picture over the years. It has been maintained with different levels of commitment, especially for entering in old data (since Crunchbase did not exist in 2000).<p>CrunchBase is a great resource, but doing those kinds of statistics without appropriate research of how consistent is the coverage is at least sloppy if not willful negligence.
评论 #2287346 未加载
tuhinabout 14 years ago
A 38% decline in number of startups and a 31% increase in investments!<p>Also the underlying assumption seems to be exponential increase is the only reason for a bubble.
评论 #2287936 未加载
doolsabout 14 years ago
There's no discussion here about revenues. Wouldn't the only reasonable metric indicating a bubble be over-valuation of companies?
评论 #2287355 未加载
评论 #2287931 未加载
gojomoabout 14 years ago
I wouldn't trust CrunchBase for reliable trend data because its coverage over time is unlikely to be consistent. But, a similar analysis based on legally-required (and thus comprehensive) SEC Form Ds might give stronger insight.<p>At least two Form-D watching services have been mentioned previously on HN:<p><a href="http://stealthmodewatch.com" rel="nofollow">http://stealthmodewatch.com</a><p><a href="http://www.formds.com" rel="nofollow">http://www.formds.com</a><p>Though, I think they're relying on easier electronic access that may not go back more than a few years.
hessenwolfabout 14 years ago
"In 2010, there was actually a 38% drop in number of new startups."<p>Pffttt.... where did this come from? Is it not that after two years of a lousy economy everybody's saving were tapped out?
funthreeabout 14 years ago
Based on the last graph we aren't in a tech bubble _yet_<p>We are just entering the last quarter of 1998.
评论 #2287932 未加载
warrenwilkinsonabout 14 years ago
I doubt the failed startups of 2000 - 2006.. etc fully represented in CrunchBases database. This alone could account for the curve.