Benchmarks for Startup Spend and Cash Burn(brex.com)
brex.com
Benchmarks for Startup Spend and Cash Burn
https://brex.com/blog/state-of-spend
35 comments
Obviously belongs on every real HN thread
For any given discussion on HN there is usually a song of his that applies.
It's definitely an interesting report of some of Brex's aggregate data, but certain statements are inaccurate and misleading. Not sure if this was intentional or not.
For example:
> CRM: 61% of startups use Copper (formerly) ProsperWorks for their CRM
Then below:
> CRM: Salesforce 13.0%, Pipedrive 26.1%, Copper 60.9%
That adds up to 100%. So startups don't use any CRMs other than those three? Bullshit. Therefore that "61% of startups use Copper" statement is also false.
Also weird that they think Intercom is customer support software, cause it's more of a CRM for us and everyone we know.
For example:
> CRM: 61% of startups use Copper (formerly) ProsperWorks for their CRM
Then below:
> CRM: Salesforce 13.0%, Pipedrive 26.1%, Copper 60.9%
That adds up to 100%. So startups don't use any CRMs other than those three? Bullshit. Therefore that "61% of startups use Copper" statement is also false.
Also weird that they think Intercom is customer support software, cause it's more of a CRM for us and everyone we know.
Hey, thanks for reading. Yes we should have been more clear that generalizing our data implies that "61% of startups use Copper", but as we don't have universal coverage across every company (yet) we'll have some missing information.
As for Intercom, that's interesting we use it mostly for support chat.
As for Intercom, that's interesting we use it mostly for support chat.
Hey HN -
Henrique here co-founder of Brex (YC W’17). I wanted to introduce the HN community to some new data we gathered at Brex as part of our new blog.
We specifically looked to highlight data that is not available to other startups.
+ We focused on startup cash burn rates by funding stage (pre-seed, seed, Series A, etc.) because for many startups there is always this question about how much you should be burning.
+ Separately, we looked at some of the most popular vendors for startups on Brex in categories like cloud services, CRM, rideshare etc.
+ We analyzed the size of the "stealth" startup market. For Brex, it was hard being in stealth for over a year and it's helpful to know just how many of you are out there
+ We looked at regional trends in startup cash burn to help startups adjust their benchmarking for their locations
Note, we were careful to aggregate the data and only share things we thought could actually help startup founders and early employees
Let me know your feedback!
Henrique
Henrique here co-founder of Brex (YC W’17). I wanted to introduce the HN community to some new data we gathered at Brex as part of our new blog.
We specifically looked to highlight data that is not available to other startups.
+ We focused on startup cash burn rates by funding stage (pre-seed, seed, Series A, etc.) because for many startups there is always this question about how much you should be burning.
+ Separately, we looked at some of the most popular vendors for startups on Brex in categories like cloud services, CRM, rideshare etc.
+ We analyzed the size of the "stealth" startup market. For Brex, it was hard being in stealth for over a year and it's helpful to know just how many of you are out there
+ We looked at regional trends in startup cash burn to help startups adjust their benchmarking for their locations
Note, we were careful to aggregate the data and only share things we thought could actually help startup founders and early employees
Let me know your feedback!
Henrique
Doing a COL adjustment/average employee salary to burn rate could be interesting to normalize the data. In other words is the salary of engineers driving the disparity or are there other factors behind the burn rates.
Exactly. Our plan going forward is to start breaking down burn by major factors like salary, rent, vendor spend.
Another interesting one we're looking at is the breakdown of burn by industry and stage. We can also then compare the major factors across these breakdowns to give really good comparisons for startups.
Another interesting one we're looking at is the breakdown of burn by industry and stage. We can also then compare the major factors across these breakdowns to give really good comparisons for startups.
Other interesting normalization metrics to consider is # of FT employees, contractors & revenue.
Yes, all worthwhile normalizations!
Burn is the numerator, $ raised or cash-in-the-bank is the denominator.
Would be really good to get a 2x2 grid of monthly burn x money raised for Series A companies. Or if that's too hard, a "Months the Series A will last at at current burn rate" bar chart.
Would be really good to get a 2x2 grid of monthly burn x money raised for Series A companies. Or if that's too hard, a "Months the Series A will last at at current burn rate" bar chart.
We can definitely add a runway / months to exhaustion to our next analysis.
That would be good. Runway is one metric -- the one I'm most interested is "how long are you _planning_ the A to last for?"
If you raise eg $10M and you ramp burn to $400k monthly, the answer is 24 months. Regardless of where you are in those 24 months.
If you raise eg $10M and you ramp burn to $400k monthly, the answer is 24 months. Regardless of where you are in those 24 months.
Sure, it might be interesting to compare rounds this way, on how long each round is planned to last.
"82% of startup failure is due to cash flow problems."
That seems unlikely. That may just mean that 82% of the time, the founders don't pull the plug until they totally run out of money.
"Cash flow problem", as a real term rather than an euphemism, has a very specific meaning. It's when the business is profitable but there's a lag between paying out cash to buy needed labor and materials and receiving payment for goods and services. Manufacturers and farmers have that kind of problem.
That seems unlikely. That may just mean that 82% of the time, the founders don't pull the plug until they totally run out of money.
"Cash flow problem", as a real term rather than an euphemism, has a very specific meaning. It's when the business is profitable but there's a lag between paying out cash to buy needed labor and materials and receiving payment for goods and services. Manufacturers and farmers have that kind of problem.
Ahh, the dreaded Cash Conversion Cycle (or "Succeeding yourself to death") I'm a wearable device founder, and it's AMAZING watching how often this happens to my hardware friends. Especially once they begin raising serious amounts of capital.
I had a friend who got into Walmart -- he was ECSTATIC. His product was a smart electronics product, and quickly shot up to being a heavy seller.
It sold so well---that he was bankrupt within six months.
Why? Because he paid his supplier upfront. Walmart didn't pay him until 60 or 90 days after sell-through. After the first run, Walmart wanted more, his P&L (accrual) showed profit, so he took a loan. Then another. Then suddenly he couldn't get any more loans. And he couldn't get a hold of product. And because of his fixed costs, there was no way to future payables could save him.
So common, and no one seems to know about it.
I had a friend who got into Walmart -- he was ECSTATIC. His product was a smart electronics product, and quickly shot up to being a heavy seller.
It sold so well---that he was bankrupt within six months.
Why? Because he paid his supplier upfront. Walmart didn't pay him until 60 or 90 days after sell-through. After the first run, Walmart wanted more, his P&L (accrual) showed profit, so he took a loan. Then another. Then suddenly he couldn't get any more loans. And he couldn't get a hold of product. And because of his fixed costs, there was no way to future payables could save him.
So common, and no one seems to know about it.
What should he have done? Negotiated a 60-90 day payment cycle with his supplier?
seeing "got into Walmart" used like "got into Yale" is so weird and strange to me that it reminds me how increasingly clumpy and centralized everything is becoming in the world.
Instead of a mesh network of equal nodes, we have a tree of exponentially larger nodes (from bottom to top)
Instead of a mesh network of equal nodes, we have a tree of exponentially larger nodes (from bottom to top)
Sorry, why wouldn't you use invoice factoring to solve this problem? Unless margins are so low that the factoring fees would put him into a loss. You'd get very competitive rates against someone like Walmart.
I just read by the Nike founder and he describes how for many years they had massive problems keeping the business afloat despite growing like crazy.
Right, we would define what you're pointing to as a subset of cash flow - you are specifically highlighting "liquidity" when you talk about the lag there, but I agree we could have been more clear. 82% is referring to failure for financial reasons vs. management, regulatory, etc.
> Top startups who “graduate” (a success metric) from raising another round of funding from pre-seed to series a burn more than the benchmarked average.
Startups that spend more money need to raise more money, I don't know what I'm supposed to take away from this fact.
Startups that spend more money need to raise more money, I don't know what I'm supposed to take away from this fact.
But startups don't just raise money because they need it, they also have to be given it. That said, there is a question of what parts of this are causal that we plan to tease apart in future analyses.
> they also have to be given it
How many startups needed to but were unable to raise capital in successive rounds? If the failure rate of successive raise attempts is low then I'd suspect that the driving factor would be the desire to raise capital.
Given that we've been in a record bull market for nearly a decade and there are signs of slowing, it may be hard to extrapolate these findings to the future. It would be interesting to try to model what things would have looked like on this data set at lower levels of investment. This type of information would be hugely valuable if there is an investment regime change occurring.
How many startups needed to but were unable to raise capital in successive rounds? If the failure rate of successive raise attempts is low then I'd suspect that the driving factor would be the desire to raise capital.
Given that we've been in a record bull market for nearly a decade and there are signs of slowing, it may be hard to extrapolate these findings to the future. It would be interesting to try to model what things would have looked like on this data set at lower levels of investment. This type of information would be hugely valuable if there is an investment regime change occurring.
Matriculation/graduation rate from round to round seems pretty low (that is, new funding failure rate is high), based on this chart cited by pg:
https://twitter.com/paulg/status/781743196963540992
Tracked down the original post here:
https://mattermark.com/startup-graduation-rate-surprisingly-...
https://twitter.com/paulg/status/781743196963540992
Tracked down the original post here:
https://mattermark.com/startup-graduation-rate-surprisingly-...
Great data here, thanks! Yes, part of the difficulty of this analysis is the low number of graduates makes it hard to see what's causal.
Agreed that from a data perspective, it would be fascinating to see how an investment regime shift would change these benchmark indicators, which should also be useful to startups.
https://youtu.be/JjjUj1ypVmI