How to calculate the Discount Rate to use in a Discounted Cash Flow (DCF) Analysis

We look at how to compute the right discount rate to use in a Discounted Cash Flow (DCF) analysis. This post is a supplement to a blog post titled “What’s your TRUE customer lifetime value (LTV)?  – DCF provides the answer“.

My thanks to my partner Stan Reiss, who co-authored this piece with me, providing all the expert math help.

In that blog post, we discuss why it is valuable to apply discounts to future cash flows when calculating the lifetime value of a customer (LTV). This discounted cash flow (DCF) analysis requires that the reader supply a discount rate. In the blog post, we suggest using discount values of around 10% for public SaaS companies, and around 15-20% for earlier stage startups, leaning towards a higher value, the more risk there is to the startup being able to execute on it’s plan going forward.

The Discount Rate should be the company’s WACC

All financial theory is consistent here: every time managers spend money they use capital, so they should be thinking about what that capital costs the company. There can be many sources of capital, and the weighted average of those sources is called WACC (Weighted Average Cost of Capital). For most companies it’s just a weighted average of debt and equity, but some could have weird preferred structures etc so it could be more than just two components.

To calculate WACC, one multiples the cost of equity by the % of equity in the company’s capital structure, and adds to it the cost of debt multiplied by the % of debt on the company’s structure. Because interest in debt is a pre-tax expense, the cost of debt is reduced by the tax rate (it’s effectively tax deductible).

The formula is


Ke =  the cost of equity.  This comes from the Capital Asset Pricing Model (CAPM), described below.

Kd = cost of debt. This is the average interest rate on the company’s debt. To be completely correct, it’s the coupon divided by the market value of debt, since the value of company bonds fluctuates, but generally this is too complicated for the exercise at hand and, unless the company is in distress, just looking at the book value is close enough.

T = corporate tax rate.  The right number to use is the marginal tax rate since you’re trying to make a marginal decision, and that’s typically 35% in the US.

Ve = value of equity.  Company market cap less cash plus debt. For a private company, best estimate – probably based on last round price.

Vd = value of debt.  As described before, the proxy is book value.

Simplifying this for Startups

For most startups, equity is the primary method of financing, so it may be helpful to simplify things and state that WACC equals Ke (the cost of equity), which effectively also means that the Discount Rate should be equal to Ke.

Computing the Cost of Equity – The Capital Asset Pricing Model (CAPM)

The cost of equity, Ke, comes from the CAPM.  What investors expect to earn on their investment in the stock. If they conclude they won’t get this return they’ll sell the stock and the price will go down, if they conclude they’ll get more than this return additional investors will buy the stock and the price will go up, eventually driving the return to Ke in equilibrium.

The basic CAPM formula for Ke is


  • Rf = Risk free rate of return.  A good proxy is a US government bond of a duration that’s commensurate with the time frame an investor would think of when owning the stock.  The 5 year T-bill is a good proxy. Today the 5 year T-bill yields 1.7%, the 10 year 2.2%, so a 2% risk free rate is a good proxy.
  • B (Beta) = Sensitivity of the expected stock return to the market return. Have to use history to estimate.  Mathematically it’s the covariance of the historical return of this particular stock and the market divided by the variance of the market. So B = Cov (Rs, Rm)/Var(Rm). The best way of getting at this is to look at the beta of similar public stocks. For public SaaS companies, the beta today seems to be about 1.3.
  • Rm = Market rate of return – what the investors expect the market to return. The public markets have returned around 8% per year over the last decade, and one would think that that’s a reasonable rate expected by investors.  There could be different opinions (for example the 5 year rate of return is a lot higher).  If a company is private, one would expect a much higher rate of return.

Plugging all this in for a SaaS company, one would get

Ke = 2% + 1.3 (8% – 2%) = 9.8%      ~ 10% for a public SaaS company.

For a private, or higher risk company, Ke will depend on the assumption on Rm (the market rate of return). Reality is this is highly volatile and situation specific – sometimes one can raise cheap money and sometimes one can not.  While a lot of situational judgment should be applied, Cambridge Associates, which tracks the stronger venture firms, claims a 30 year venture return of 17.7%, and that’s probably the best proxy.

So for a private SaaS company one could assume

Ke = 2% + 1.3 (17.7% – 2%) = 22.4%      ~ 20% would be a good estimate to use

For reference our Beta calculation came from averaging Google Finance Betas for a selection of public SaaS companies:

  • – 1.33
  • Workday – 1.53
  • ServiceNow –  1.11
  • Netsuite – 1.5
  • LogMeIn – .96
  • Liveperson – 1.35
  • Demand ware – 1.31.

The newer SaaS public cos (ZEN, HUBS, MKTO) haven’t been public long enough to calculate a good Beta.


For SaaS companies using DCF to calculate a more accurate customer lifetime value (LTV), we suggest using the following discount rates:

  • 10% for public companies
  • 15% for private companies that are scaling predictably (say above $10m in ARR, and growing greater than 40% year on year)
  • 20% for private companies that have not yet reached scale and predictable growth

Is there an argument to be made that startup SaaS companies shouldn’t be using a different discount rate to public SaaS companies, as their goal is to show that they have the needed unit economics to become a public company? Yes, there probably is. We are charting new territory with this analysis, so it will be interesting to hear readers’ feedback on this question.

About the Author

David Skok

Stay updated

Related Posts

Customers hate being sold to. Customers hate being sold to. They don’t mind getting expert...

In this article I interview Aaron Ross, co-author of a new book, Predictable Revenue. Aaron...

An Interview with Andrew Quinn of HubSpot Having a high performing team means: Hiring the...

  • U2GT

    Great summary and reminder for everyone. The one issue with this approach is it is based on historical data. Beta calcs are no always statistically significant and can vary signficiantly over different periods of time, etc. There are other data points to consider. Another source for Ke is the returns implied by VC/PE models which look forward. I see many VC/PE models for tech deals that aim for a 3x return within approximately 4 years which is a 31.6% CAGR. Based on these models, 20% to 30% is a more reasonable Ke range for a private company DCF (the VC/PEs are discounting at a higher rate than the historical studies above). The WACC can also be determined from LBOs and comparable transactions without relying on historical statistical data. The leverage portion of the WACC equation comes from the debt used in a particular deal, the debt rate is a market rate for the industry and and leverage on the deal, and Ke is the forward return that the VC/PE is requiring (what they modeled based on exit assumptions). For exampe, if a large application software company goes private, all the metrics in the model are a comp for the WACC equation that can be used for other application software companies. Of course this data is not as readily available to the public like the data discussed above, but it is more accurate because it measures the expected return (forward looking) for the particular transaction and industry. If a PE firm takes a company private and assumes a 25% return on deal with 70% leverage, then the WAAC is easily calculated without any regression analyses, and can be used as a practical, real-world data point for other deals. In other words, the best “market” WACC data is embedded in all VC and PE models.

  • Thanks for your comment – this is very helpful. Do you have a good guess as to what the WACC would be on that PE deal that you mentioned looking for a 25% rate of return, with 70% leverage?

  • Michael O’Hare

    Hi David. Your SaaS blog is superb! 25% is actually too aggressive and not a good example. 30% to 40% in equity is more market for industrial companies and 40 to 60% for technology (because debt limits on EBITDA are about the same for both, but there is a big difference in valuation multiples, which reflects the higher growth in technology). Here is a practical software example: a company with $100M of revenue (growing 20%) with $20M of EBITDA will likely realize a value in the 12x to 15x EBITDA range ($240M to $300M). The first practical constraint is how much debt can be used in the buyout. Generally speaking 6x-7x is aggressive, but let’s assume 6x is possible in this example. 6 x $20M = $120M. So the debt market has dictated the leverage ratio of 44.4% for the WACC formula above (no need to review comps). Average rate on the debt will be about 6.5% and let’s assume the tax rate is 40%. If the company sells for $270M (13.5x EBITDA), then equity value = $150M and debt value = $120M. The VC/PE has certain exit values in their model that imply an equity return. In my experience the investor is targeting a 25% to 35% return over 3 to 5 years, but this is not an input to the model, but an output from their estimate of risk, the exit time and value, and the price they bid to win the deal. Almost all the models I see result in a 25% to 35% equity return target, but investors may
    underwrite to higher or lower rates depending on the situation. At closing, the implied WACC for this deal = [($150M/($150M+$120M)] x 30%] + [$120M/($150M+$120M) x 6.5% x (1 -40%)] = 16.7% + 1.7% = 18.4%. Note that if the
    leverage ratio changes, the WACC won’t change much because the investor’s
    expected equity rate of return will decline with less leverage (less financial
    risk) and it will increase with more leverage. Each closed VC/PE deal has a model with the ‘expected’ return. Again, I admit this information is not
    readily available, but it’s out there. Based on my experience expected Ke for a VC in a high growth technology investment (e.g., a Series B round) is 25 to 35%, but there is no leverage on these companies. For larger companies
    like in the example above, Ke is also about 25-35% but it’s a leveraged return
    (it would be lower without leverage). So the WACC for the unleveraged Series B round might be 30%, but the WACC for the larger PE transaction is 18.4%. This makes sense because the unleveraged Series B has more business risk, and less financial risk. Another way to describe both of these scenarios is that the investor is expecting a “cash on cash” return of 3x within just under 4 years.
    If it’s an early stage company and unleveraged deal, then they return 3x
    their money or a 30% unleveraged IRR, and if it’s a larger company with leverage, they still get 3x their money over the same period (30% leveraged IRR). I see aggressive investors underwriting as low as 2x “cash on cash” and disciplined investors targeting 4x “cash on cash.” I hope this adds a practical perspective to the excellent theoretical explanation above. Looking forward to your future posts.

  • Michael, many thanks for taking the time to write such a thorough explanation. This is really helpful, and provided a new insight on how to think about this. Much appreciated!