Your addressable market? Bottom-up!

Over the past few months, I’ve been privileged to meet many talented scientists and physicians who have created wonderful technologies for improving patient care, but who have little experience determining whether their technology could be the foundation for a successful new business.  One recurring challenge is quantifying the addressable market – not a skill that is typically taught in science, engineering or medical school.  Here’s my approach, using a hypothetical example.

Imagine you are a physician-scientist who has just invented a coronary-catheter-based treatment for vulnerable coronary plaques – a treatment designed to dramatically reduce the incidence of heart attacks and death.  Your colleagues urge you to talk to experienced VC’s, who will help you transform your brilliant invention into a standard-of-care product (and make you impressively wealthy at the same time).  Your colleagues also point you to a helpful blog post on presenting to VC’s, which instructs you to quantify the addressable market.  Okay, no big deal.  You pull out your “AHA Heart Disease and Stroke Statistics – 2009 Update,” as published on the AHA website, and put into your presentation:

    In 2006, the total costs associated with treating coronary artery disease in the US were $165.4B.

What’s wrong with this picture?

Besides the fact that $164.5B is about 30 times the market size for drug-eluting stents (one of the largest-ever medical device opportunities), this number provides no insight into the quantity and pricing of products that could ultimately be sold by your company.  Even the direct hospital-only costs of CHD ($54.6B) are still too high-level to understand the addressable market.

Mark Peter Davis provides a good definition of the addressable market on his blog.  Put simply, the societal cost of the medical problem is not the same as the scope of the business opportunity.  The societal cost includes the direct payments for drugs, physicians, and providers (including instrument costs), as well as indirect costs associated with the condition.  Despite the brilliance of your product, your new widget is unlikely to capture all of that value.  Even the hospital/provider cost includes the cost of running the facility, staff and overhead, and other instruments needed for your procedure.

So how do you go about sizing the addressable market?  In concept it’s simple – a bottom-up approach:

      Q. Number of patients who may realistically be treated with your product


      P. Average selling price per instrument


    M. Addressable market M = P x Q

Let’s explore this a little further, using our CHD treatment device as an example.  Here’s some data from the AHA:

    In 2009, an estimated 785 000 Americanswill have a new coronary attack, and about 470 000 willhave a recurrent attack. It is estimated that an additional195 000 silent first myocardial infarctions occur each year.  The estimated number of  PCI (stent/angioplasty) procedures performed in US in 2006 was 1.313M, and the estimated number of diagnostic-only angiography procedures was 1.115M.

How many patients might be a candidate for your new catheter-based CHD treatment?

Step one: Break down the population of all patients with the condition (CHD), by their current treatment:

      A. Patients receiving PCI: 1.313M


      B. Patients receiving angiography: 1.115M


      C. Patients with known CHD (usually treated with statins, e.g. Lipitor): some larger number


    D. Patients with CHD, but not yet diagnosed:  some number that is even larger than C.

The best sources for patient population data include medicare (CMS) data, peer-reviewed journal articles, medical specialty societies, market research reports, stock analyst reports and company announcements (SEC filings, investor conference presentations, and earnings calls).  Academic scientists and physicians often have access to market research reports and stock analyst reports via their institutional libraries.

Step two: For each group, estimate the percentage of each group that might be candidates for the new procedure.  In our example, people who are already in the cath-lab (groups A and B) are much more likely to be candidates for a catheter-based treatment then the much larger population of patients not yet in the cath lab (groups C and D).  Groups A and B are higher-risk patients who are more likely to benefit fro the new treatment, and much of the effort of the new procedure (vascular access, etc.) is already performed in the existing procedure.  In this case we will assign 0% to groups C and D, meaning that none of these patients are addressed by our product.  We might talk about this as “upside” in a discussion with VC’s meaning that the availability of a new treatment might pull more people into the cath lab.

For groups A and B, we need to estimate percentages of each group that might be candidates.   For example, there may be some subset of patients that are medically contra-indicated to our treatment.  Alternatively, there may be some  subset of patients that are in higher need of our new treatment (e.g. if they are already maxed out on other forms of treatment, but still have high risk factors such as high cholesterol).  The size of these subsets is sometimes identified in the medical literature or from the same sources as our patient population data.  More often, you’ll need to survey some potential customers (usually about 10 to 20) and create your own estimate.  (There are some very good consultants who can help you with this).  Be sure to capture the customer’s reasoning as well as their estimates.

Back to our hypothetical example, imagine that our hypothetical survey data shows that 75% of PCI patients will need the new treatment, and 25% of angiography patients will need the new treatment.  So the total number of addressable US patients will be 1.26M.  To make a worldwide estimate, I’ve often seen people apply a rule of thumb.  For interventional cardiology, we assume that the US is half the world market, so the total number of addressable patients for our product would be 2.52M worldwide.  The rule of thumb varies by medical specialty, so you’ll need to find someone in the specialty who can help.  Also, market research reports, analyst reports, and company reports might contain real data.  If so, use it!

Step three: Estimate P, the average selling price.  There are many ways of developing a price estimate.  It’s an art and a science.

Two methods I can describe briefly are the comparable method and economic value method.  The easiest is to use a comparable.  For our new treatment for coronary heart disease, we might price it like a drug-eluting stent, at about $2K per unit.   A variation on the comparable method is to estimate some premium to the comparable, if the new device is expected to perform somewhat better than the comparable.

The economic value method is more complicated, and is based on an economic analysis to determine the economic benefit achieved by the device.  An oversimplified example:  Assume the use of 100 of our CHD treatment devices will prevent 15 heart attacks in the first year.  Also assume that each heart attack would cost $20K to treat.  Then on average, each of our devices will save $3K, justifying a $3K price.  Of course, we will need to prove out our assumptions, to convince payors to pay this price.  For our CHD treatment device example, we’ll use a price estimate of $2K, which is more conservative.

Step four:  Calculate the opportunity.  An annual treatment population of 2.52M patients worldwide, at an estimated $2K price each, yields an addressable market of about $5B.

In this post, I have used an example of a single-use medical device for a single application.  Reusable devices and capital equipment need to also take patient-throughput, procedure capacity, and number of provider sites (e.g. hospitals, offices, OR’s, etc.)  into account.  Products that include capital and single-use components have all of the above challenges.  Products that will be used for additional applications over time (e.g. if our device could also be used in peripheral vessels) will need to have this calculation repeated for each application.

Calculated from the bottom-up, the addressable market size serves as a realistic, defensible guidepost for later estimating potential revenues.  Since those revenues ultimately drive the return on venture investments, getting the addressable market size right is critical for venture investors.


4 thoughts on “Your addressable market? Bottom-up!

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s