Fully vs Partly Remote Performance

US Tech Series A deal pace rose slightly from January to February, by about 6%. When adjusting for 10-year historical seasonality, it rose by about 7%.

Fully vs Partly Remote Performance

Since the Pandemic forced a work from home reality on the world, we’ve seen that startups are more “remote” than they were before, blurring the lines on where a company geographically resides. In the past, I’ve done several performance analyses based on startup geography – this is a subject I follow closely. It’s hard to find uniformly tagged data sets to work with, but thankfully Y Combinator has tagged their alumni companies with the mutually exclusive “Partly Remote” and “Fully Remote” tags. I wondered how those two populations of companies did in raising Series A rounds, as a simple binary success metric. The table below shows that the Partly Remote companies outperformed the Fully Remote ones.

Observations

* Statistics aren’t destiny. Great companies can be of any geographic configuration, but this data shows the odds are worse for Fully Remote companies.

* For statistics wonks, the difference in means in the two population is 3.9 standard errors. For regular people, there is essentially no chance that the main conclusion is due to statistical noise.

* The underlying causality is an area for rich debate. I have opinions on this, but those would take many pages.

Caveats

* Of course, what we’d really like to measure is some quantification of cash returned or value created. But this takes about a decade to happen, and then tells us about the startup world as it existed a decade prior. Raising a Series A round is almost a necessarily condition to ultimate success, happens much sooner, and is an easily defined binary variable.

* For most analyses I do, I focus on US Tech companies, but in this case, I did not filter by geography or sector.