The market for X is a billion dollars! The demand for Z is going to grow at 18% annually over the next five years! Being quantitative about a potential business opportunity is essential. It’s a powerful way to justify your belief in a product or a technology or a commercial strategy. But we’ve encountered too many entrepreneurs who clearly don’t believe their own numbers, or who seem to believe that the markets they are attempting to serve are so new and esoteric that their magnitudes are unknowable. And these are the entrepreneurs who suffer the most, because they are forced to make critical decisions, relying exclusively on their “gut” or on the sentiments of people they trust, or on the sentiments of people recommended by people they trust. This is no way to run a business!
So when investing time and resources into building a market model, how do you make sure you end up with a useful decision making tool, one that you believe in, one that you can use as an advocacy tool? In our experience, the following six considerations are critical in making sure you get what you need out of your modeling efforts.
Consideration 1: What are you going to do with the model?
In the end, collecting data and building models isn’t the ultimate goal. But these activities can help you make informed decisions (based on facts and clear assumptions instead of pure conjecture) and help you better understand opportunities and risks. So the first two questions we ask when building any market model are:
- Who is going to be using the model output?
- What decisions will it inform?
Are you trying to convince an investor about the growth potential of a new and emerging segment? Or are you trying to compare the revenue potential of two different markets, to tune the focus of your commercial efforts? The better you understand the intent of a model, the easier it will be to design, build, interpret, communicate and ultimately maintain.
Consideration 2: How good is good enough?
Market modeling is expensive! And that expense scales directly with the amount of precision and breadth you try to achieve with your model. So it’s critical to take a step back and think about what kind of modeling exercise is going to meet your needs. At Chrysalis, we often break market sizing exercises into one of three classes. And each (or at least two of them) has their place.
THE SWAG:
Sometimes there’s no need for a scalpel when a pocket knife will do. For companies exploring markets for a new technology or considering expansion markets for existing technologies, Chrysalis often performs a Technology Market Fit, where we consider the needs of customer segments that span dozens of potentially diverse industries. In down-selecting opportunities, it can be useful and efficient to make broad, directional estimates of the size of the potential opportunity. Is this a $10M market or a $4B market? The data for the SWAG can come from industry analyst reports, from looking at the revenues of the major vendors in the space, or from quickly estimating the spend per global customer/patient and multiplying by a simple ASP.
THE TOP-DOWN MODEL:
Danger lies ahead! The top down market-sizing activity is what underpins the models contained in many of the purchasable, 3rd party market research reports. A typical top-down market sizing strategy is to start with a macroscopic estimate of the size of an entire industry, and then to break that industry into smaller market segments using publicly available data sources (e.g., NLP-based publication/citation frequency, revenue splits from earnings reports, population and demographic data). It’s not difficult to produce a report that contains dizzying levels of segmentation, all presented in tidy tables and charts that imply a level of confidence and precision that may or may not be justifiable given the input data. At best, the output of these reports represents a directional view of a market, its major sub-segments, and its possible growth trajectory.
While there is a place for using these reports, many business leaders falsely assume that they can apply the general market trends captured in these reports to the specific subsegment that their business can address. If these leaders can agree amongst their peers and their investors that a directional view of the scale (and potentially the trajectory) of a market is good enough, then this is often the right approach. But all too often, leaders invest in a TOP-DOWN model hoping that they will be able to use it to build a robust framework for a business plan. What they really need is something more rigorous.
THE QUANTITATIVE OPPORTUNITY ASSESSMENT:
If it’s critical to understand the dynamics of an opportunity, if you need to understand the upper and lower boundaries of potential revenue for your product over time, if you are trying to build a commercial plan that focuses on the right customers, or if you are trying to compare the risk profiles of different business plans, neither the SWAG nor the TOP-DOWN model is likely to serve you well. What you really need is something we call a Quantitative Opportunity Assessment, which usually contains the following:
- Data collected from interviewing and surveying primary sources (e.g., potential customers who might buy your product)
- Data from secondary sources (e.g., reputable analysis reports, medical incidence and prevalence data) that can be used to triangulate and reconcile the implications of the primary data
- Precisely constructed segments (see consideration 3)
- Multiple scenarios that take into account alternative starting assumptions to build “bull, bear, and base” cases.
- Quantitative comparisons with similar historical technologies/products and the rate of their commercial evolution.
- An honest assessment of competitive threats (especially if you are building a revenue model)
- Consumer willingness to switch from one technology or product to the other (especially in validated, clinical environments)
- Transparency: understanding where the model is precise and admitting where the model is based on one or more low-confidence assumptions
- A plan and a process to revise and evolve your model as the dynamics your industry changes or as new information becomes available
Consideration 3: Are the segments precisely defined?
Specificity matters. The difference in size between a more generalized market segment (say, genomic testing for infectious disease) and a carefully defined subsegment (say, follow-up testing for resistance markers for patients unresponsive to antibiotics) can be orders of magnitude. And all too often, people settle for the larger, less precise estimate because this is all that they have, or they believe that one is a reasonable substitute for the other.
Furthermore, when estimating future product revenue, it’s critical to be brutally honest about what subsegments your product serves, where it is truly differentiated and where the status quo might prevail. More times than not, the long-term clinical opportunity for a given technology is vastly larger than the immediate basic research opportunity. But if your product is fundamentally unsuitable for use in a clinical setting, you may have unrealistic expectations about future revenues. This is equally true in certain applied markets like agricultural testing and forensics.
The need to define precise market segments is amplified when you attempt to communicate your projections to others inside and outside your organization. People’s initial biases are often unshakeable, and they hear what they want to hear. When stakeholders aren’t 100% aligned on the definition of a market, chaos generally ensues. Teams that thought they were aligned on the risk and potential rewards of pursuing a business strategy find themselves in an unholy tower of Babel.
Consideration 4: Where is the Data Coming From?
The life science and clinical diagnostics markets are unique. Compared to high-profile consumer and B2B markets, where data about market size, vendor share, average selling price, unit volumes, and geographic distributions are constantly collected and refined (often in real-time), data about life science customers and vendors is incredibly sparse (and often lagging). Even in highly attractive life science markets, the number of paying customers can be orders of magnitude smaller than in consumer markets, and building a consensus about purchase trends and behaviors is a significant statistical challenge.
So when you read a syndicated market report that appears to break down the served market for some young, emerging technology in amazing detail, segmented by dozens of disease research areas, subdivided by geography, and cross-referenced by vendor, you should really ask: “Where is this data coming from?” You might be surprised how many reports selling for thousands of dollars aren’t really based on any systematic collection of primary data, and are often created by analysts unfamiliar with the topic. Conjecture is cheap.
And if you’re hiring a firm to collect data and build you a market model, investigate the quality of their work. Do the people creating the model have a background in the subject matter? Are they steeped in the applications space? How long have they worked in the field? Have you seen examples of how they think and have put together models? Have other people you trust recommended them? If you look back on past work they did, how accurate was it? Can they explain the variances between what they authored and what really happened? What sources of data did they use as the basis for their model? Was it referenceable and objective? Are they willing to work with others to expose and tighten assumptions? Budget and time aside, did they incorporate primary research where secondary data was lacking?
Consideration 5: Is the model output realistic?
Even with the best of intentions, utilizing carefully curated data and statistically intricate forecasting tools, sometimes the outcome of the model just doesn’t pass the smell test. The hypothesis just isn’t realistic.
It’s easy to get carried away by the promise of a new revenue stream. The fervent demand for a new product/technology, captured in interviews and quantified in a survey that reveals huge unmet needs even for inferior solutions, can get an entrepreneur pretty excited. But in the wake of this excitement, revenue forecasters often forget the time and commercial investment needed to change customer buying behavior. In the clinic, early adopters often need to rewrite medical SOPs, to generate clinical evidence, and to convince payers to code and fund a new approach. Visionary entrepreneurs often think their product represents a transformational technical improvement, when in reality, its benefit to the patient is incremental. And in the time required for these medical and commercial transformations to happen, competitors and other market forces can quickly erode an opportunity that seems to be immediate and obvious today.
Consideration 6: When should you get help?
Some tasks are just better handled using a specialized, objective, external expert in building models and sizing markets. There are a number of reasons to hire a modeling specialist:
- Gain expertise and specialization that goes beyond your company
- Tap into an unbiased, objective and referenceable perspective
- Gain access to unique datasets and information resources
- Save time & money, free up internal resources to run your business
- Extend and challenge existing in-house models
- Get instant access to advanced analytical techniques (databases, machine learning tools, statistical frameworks. etc.)
- Market modeling is scary: internal stakeholders are in positions of power and often merciless with criticism even in the absence of valid perspective themselves.
If you would like to discuss your market modeling ideas and challenges with the Chrysalis team, send us a note at info@chrysalisbiomed.com.