A friend asks: “what percentage of U.S. startups that raise a Series A do not go through an incubator or accelerator?” That’s a great question that I haven’t thought about before. So, I dug into the data to find out.
The chart here shows the annual number of Series A financings in the U.S. (bars) between 2010 and 2018, broken down by whether the company previously participated in an accelerator (dark bars) or not (light bars). The green line on the right axis indicates the accelerated-company share of Series A financings.
We can observe a steady, linear increase in the share of Series A startups that previously completed an accelerator program—from around just 2 percent in 2010 to 28 percent today. The direction of the line does not surprise me, but I the magnitude does—around one out of every four Series A companies today participates in an accelerator.
To break down the averages over the last few years:
2015-2018: 21%
2016-2018: 23%
2017-2018: 25%
2018: 28%
There is reason to believe these figures represent an upper-bound of true accelerator activity. In PitchBook’s data they refer to these deals as “accelerator/incubator”. As I have written in the past, these two models differ substantially and many programs parading as accelerators are not in fact accelerators in the true sense of the word. That said, there is growing convergence towards the accelerator or at least accelerator-like model today.
This leads to the obvious question: how far up? Nobody really knows. However, the accelerated-company share of Series A financings shows no sign of slowing right now—and in fact, it may itself be accelerating (see what I did there).
That aside, there are some clues suggesting we may be nearing peak-accelerator in the U.S. The charts below show the number of new and existing active accelerators (using PitchBook’s “accelerator/incubator” definition) along with the number of accelerated companies in recent years (actuals and as a share of total venture activity). In the case of accelerated companies I include actual figures as well as an “adjusted” figure that takes into account deal reporting lags at the early stage (this is just an estimate).
In both cases, my best guess is that things are beginning to flatten out at the accelerator stage in the U.S. And my best guess is that in a flattening environment there will be lots of consolidation. The best programs like YC and Techstars will expand their dominance and the marginal ones will whimper into the night. I also expect to see consolidation around the "network model." Techstars is the most visible example of this, but others include DreamIt and MassChallenge. Finally, where there will be growth, I suspect, it will come from corporates. More corporates will either run their own programs (eg, AT&T, John Lewis) or will partner with others in the industry to launch speciality accelerators (eg, Techstars’ many corporate partnerships).
Regardless of what the future holds, startup accelerators have seen an impressive rise in a relatively short amount of time. As the data suggests, accelerated companies are being validated by Series A investors as they garner a larger share of this activity than ever, with no signs of slowing on that measure at least so far—even as the overall size of the market segment (accelerators, accelerated companies) appears to be stabilizing.
