Depends on kind of business, profit level and size.
A small (sub $1m in annual revenue) services company will typically sell on some multiple of "seller discretionary earnings" - basically; how much cash can the owner operator take out in a year (including their own salary). A "typical" multiple would be 2-4x SDE.
A larger services company will typically sell on some multiple of EBITDA (earnings before interest, taxes, depreciation and amortization). The main way this differs from the above is that the owners/CEO salary is not included. Traditionally the multiples here has been 4-6x EBITDA (which is what a "value oriented" private equity fund will seek to pay), but recently there is so much dry powder in the private equity world that those multiple has been pushed up, particularly for technology companies. 6-10x is not unheard of, more is possible.
A product or software (especially SaaS) company might sell on a multiple of gross revenue instead of earnings. Depending on how profitable the company is, this may or may not be a higher value than the above EBITDA multiple. Above a certain size, I would expect a SaaS company to sell for at least 3-4x revenue, more if growth is strong, even more for larger businesses.
In general you'll see multiples go up for a) strategic fits for larger acquirers (eg they can sell your product to their existing customers) b) growth and c) larger revenue companies.
This latter point might be confusing - why would a larger company not only get a higher price because the earnings/revenue is higher, but also get a higher multiple of that revenue? This "multiple expansion" occurs because there are (lots) more available capital to buy larger companies. Smaller companies are riskier and also the transaction and other costs (operating, optimizing) are similar for a $100m vs a $10m deal. There is also more leverage available at better terms for buyers of large companies.
The company in the article appears to be more of a software company & less of a service company. From the conversations I've had, I would expect to see a valuation in the range of 2-5x the company's earnings.
(The term of art is SDE/Seller's Discretionary Earnings, and is basically the sum of all the money the owners are able to take out of the business, including paychecks and bonuses.)
In the few conversations I've had, it's always been revenue. This provides an incentive to spend lots of money on user acquisition, even if doing so means losing money (or not making as much).
I wish there were a standard assumption that if a business is profitable, then we use profit instead of revenue, and (crucially) use a much larger multiplier. It feels like I'm always having to remind folks that revenue ≠ profit, and that normal guidelines (we invest in companies with $X00,000 revenues) might need to be adjusted for smaller, breakeven or profitable businesses.
Could be either one. The methodology you use for valuation reflects your priorities. Warren Buffet calculates valuations based on free cash flow[1].
For what it's worth, the company I work for currently just got sold off for $4.4bn USD by our parent company. Based on quarterly filings, that equated to about 2x annual revenue and 9-10x EBITDA for last year (not sure what the multiple was for net).
I'd pay more for a business doing $500k in revenue with $250k profit (50% margins) than a business doing $1M in revenue with $100k in profit (10% margins).
Usually it's dependent on the person selling (or their advisor). If a company is very profitable (e.g 50% margins) then an earnings multiple makes sense and the advisor will essentially present the company in such a way that it's pretty clear to the potential buyer that they're expected to think of valuation in terms of profit multiple. Incidentally, lots more buyers out there are comfortable doing that then revenue multiple. However, a revenue multiple has become the norm for companies with very highly recurring revenue (e.g. SaaS) and where the company has been run for growth instead of profit. There are many fewer buyers who will do that though, and getting leverage for deals like that is harder (a big driver of returns for some funds)
No sensible buyer (i.e. not strategic) is going to pay high revenue multiples for a private illiquid company. Exceptions to this might be when they have some advantage (existing customers etc) to sell to. But nearly all PE shops overpay on large deals.
By strategic, I don’t mean “sensible” or “smart”, I mean it’s a buyer where the asset is considered strategic, hence it would always fit under your “when they have some advantage” umbrella.
It assumes a lot of costs are fixed, and do not scale with income. Software is weird like that.
Double the 500k R / 250k P company without more expenses, and you now have 1 M R / 750k P. Double the 1M R / 100k P and you now have 2 M R / 1.1 M P.
Especially if a firm can come in and do a round of layoffs (Replace support with outsourced, fire marketing, replace devs with outsourced).. it would be pretty easy to get the margin really really high for a few years, which is all they may be looking for.
Revenue would be nicer I guess, but if it's software that simply works then maybe they are not much different? Not how revenue and profit compare in other types of industries that is.
IIRC I think it was Instagram that when they talked about being bought by Facebook that they sat down and agreed that if Facebook bought Instagram that Instagram would be X% of the value of the combined companies and thus they came up with that number.
It seemed like a non traditional route as it wasn't the usual X earnings multiplied by a time period.
Comments
In the text the author discusses also selling the company outright - question: how does one calculate a price for this usually? A few years' profit?
Depends on kind of business, profit level and size.
A small (sub $1m in annual revenue) services company will typically sell on some multiple of "seller discretionary earnings" - basically; how much cash can the owner operator take out in a year (including their own salary). A "typical" multiple would be 2-4x SDE.
A larger services company will typically sell on some multiple of EBITDA (earnings before interest, taxes, depreciation and amortization). The main way this differs from the above is that the owners/CEO salary is not included. Traditionally the multiples here has been 4-6x EBITDA (which is what a "value oriented" private equity fund will seek to pay), but recently there is so much dry powder in the private equity world that those multiple has been pushed up, particularly for technology companies. 6-10x is not unheard of, more is possible.
A product or software (especially SaaS) company might sell on a multiple of gross revenue instead of earnings. Depending on how profitable the company is, this may or may not be a higher value than the above EBITDA multiple. Above a certain size, I would expect a SaaS company to sell for at least 3-4x revenue, more if growth is strong, even more for larger businesses.
In general you'll see multiples go up for a) strategic fits for larger acquirers (eg they can sell your product to their existing customers) b) growth and c) larger revenue companies.
This latter point might be confusing - why would a larger company not only get a higher price because the earnings/revenue is higher, but also get a higher multiple of that revenue? This "multiple expansion" occurs because there are (lots) more available capital to buy larger companies. Smaller companies are riskier and also the transaction and other costs (operating, optimizing) are similar for a $100m vs a $10m deal. There is also more leverage available at better terms for buyers of large companies.
Wow this is a fantastic explanation of multiples expansion. Exactly answered the question I had while I was reading. Thanks
Yes, service companies usually sell for 2-3.5x annual recurring revenue (ARR), which is measured by averaging your last 3-5 years of revenue.
And you'll need to be easily replaceable.
The company in the article appears to be more of a software company & less of a service company. From the conversations I've had, I would expect to see a valuation in the range of 2-5x the company's earnings.
(The term of art is SDE/Seller's Discretionary Earnings, and is basically the sum of all the money the owners are able to take out of the business, including paychecks and bonuses.)
FEI and other brokers write a bunch about valuing small businesses: https://feinternational.com/blog/how-do-you-value-an-online-...
Recurring revenue or profit?
In the few conversations I've had, it's always been revenue. This provides an incentive to spend lots of money on user acquisition, even if doing so means losing money (or not making as much).
I wish there were a standard assumption that if a business is profitable, then we use profit instead of revenue, and (crucially) use a much larger multiplier. It feels like I'm always having to remind folks that revenue ≠ profit, and that normal guidelines (we invest in companies with $X00,000 revenues) might need to be adjusted for smaller, breakeven or profitable businesses.
Typically an earnings multiple will be higher than a revenue multiple (e.g 8x vs 4x - which can be equivalent if the profit margin is large enough)
Could be either one. The methodology you use for valuation reflects your priorities. Warren Buffet calculates valuations based on free cash flow[1].
For what it's worth, the company I work for currently just got sold off for $4.4bn USD by our parent company. Based on quarterly filings, that equated to about 2x annual revenue and 9-10x EBITDA for last year (not sure what the multiple was for net).
[1] https://www.entrepreneur.com/article/66442
It's almost always revenue, but I'm not sure why.
I'd pay more for a business doing $500k in revenue with $250k profit (50% margins) than a business doing $1M in revenue with $100k in profit (10% margins).
Usually it's dependent on the person selling (or their advisor). If a company is very profitable (e.g 50% margins) then an earnings multiple makes sense and the advisor will essentially present the company in such a way that it's pretty clear to the potential buyer that they're expected to think of valuation in terms of profit multiple. Incidentally, lots more buyers out there are comfortable doing that then revenue multiple. However, a revenue multiple has become the norm for companies with very highly recurring revenue (e.g. SaaS) and where the company has been run for growth instead of profit. There are many fewer buyers who will do that though, and getting leverage for deals like that is harder (a big driver of returns for some funds)
No sensible buyer (i.e. not strategic) is going to pay high revenue multiples for a private illiquid company. Exceptions to this might be when they have some advantage (existing customers etc) to sell to. But nearly all PE shops overpay on large deals.
By strategic, I don’t mean “sensible” or “smart”, I mean it’s a buyer where the asset is considered strategic, hence it would always fit under your “when they have some advantage” umbrella.
It assumes a lot of costs are fixed, and do not scale with income. Software is weird like that.
Double the 500k R / 250k P company without more expenses, and you now have 1 M R / 750k P. Double the 1M R / 100k P and you now have 2 M R / 1.1 M P.
Especially if a firm can come in and do a round of layoffs (Replace support with outsourced, fire marketing, replace devs with outsourced).. it would be pretty easy to get the margin really really high for a few years, which is all they may be looking for.
Also maybe, some forms of financing are based on revenue, and just care that there is enough profit to cover the interest e.g. https://en.m.wikipedia.org/wiki/Revenue-based_financing
Revenue would be nicer I guess, but if it's software that simply works then maybe they are not much different? Not how revenue and profit compare in other types of industries that is.
IIRC I think it was Instagram that when they talked about being bought by Facebook that they sat down and agreed that if Facebook bought Instagram that Instagram would be X% of the value of the combined companies and thus they came up with that number.
It seemed like a non traditional route as it wasn't the usual X earnings multiplied by a time period.