Overview
Businesses looking to launch new products or services, or to bring existing products and services into new markets, are justifiably focused on trying to fully understand the uncertainty associated with such an endeavor. The risks of venturing into the unknown are great, so efforts are diligently undertaken to gain as much insight as possible into likely outcomes, and to develop plans accordingly
.
A very recent article in my local newspaper describes how the three-year-old arrangement to move a professional, major league sports team from its long-time home arena to a nearby but logistically different location “…hasn’t panned out.”. According to an executive of the firm that owns both arenas: “…to date it has been met with very modest success and not at the numbers we have hoped.”. The 25-year agreement, however, had an opt-out option that could be exercised within the first 5 years, and is now being actively pursued. In other words, the allowance for risk associated with this new market venture was built into the transaction from the beginning…that’s good risk management!
Customer research – in its many forms – is the foundation for all other efforts associated with new business models, and serves as the basis for the primary Go/No-Go decision. In many cases, a deeper look into primary customer research is needed to uncover risks that can have a significant impact on proper interpretation of the results.
Well-designed customer or product research will provide good insight into important elements such as attribute selection, feature ranking, price sensitivity, bundling preferences, target segment size, and the like. Through the use of sophisticated marketing tools, extremely well-crafted research serves as the foundation for go-to-market plans that often yield returns very much in-line with predictions.
Many instances, however, do not yield such well-behaved results, particularly where the product/service is so new that predicted customer behavior cannot benefit from prior experience, or where a business in venturing into a customer segment with which it has little or no history of involvement. In other cases, there are simply unanticipated factors, such as speculative market swings, developing events, or other unknowns whose collective impact can derail an otherwise robust predictive process. This variability in returns is the defining element of risk.
Measuring Risk
Techniques in risk analysis can help the pricing function contribute to a more thorough understanding of the potential outcomes in new product/new market launches. Realization Risk is the risk associated with not achieving predicted results, and is a necessary element in providing a comprehensive picture of future behavior. Stress Testing offers the insight necessary to confidently prepare for scenarios where the outcome – while being heavily unlikely – may prove to be unexpectedly adverse.
To better understand Realization Risk, take the case of a consumer service firm exploring a new product offering intended to appeal to existing as well as new customers. In-depth research with existing customers – including carefully-crafted conjoint analysis – shows that preference for a high-priced/full-featured offering is rather substantial: 24% of respondents expressed a willingness to purchase a package of new features at $500 per month. Applying the survey results to the full customer population of around 10,000 customers, the total expected number of customers who would purchase the $500 per month package is around 2,400…that’s $1.2mil. per month in revenue from this package, if you go strictly by the empirical research results (see figure 1). Sounds promising, right, but what does actual behavior tell us about the risk of achieving these predicted results?
A current customer-spend report shows that only around 800 out of 10,000 customers have spent $500 or more per month on the company’s existing offerings in the past year. The ratio of current spend vs. indicative spend is 0.34, revealing a level of realization risk that cannot be ignored (a ratio of 0.75 or greater would indicate a much higher level of confidence in achieving survey-indicated results). While the survey respondents were reacting to a package of features that may prove to be truly aspirational, it is necessary to explore the likelihood of fulfilling these aspirations with existing resources, along with the associated impact on expenses. Another – but very different – possibility is that the disparity between historical spending habits and willingness to spend on a premium package is an indication that the package may prove to be as appealing as the research suggests, and therefore may actually be underpriced!
In either case, the level of realization risk reveals an uncertainty of achieving forecasted results that requires additional efforts to understand, which is where stress testing can help. Expanding upon the research into the premium package (Package A, above), a lower-priced package also offered in the conjoint analysis returns survey preference results far higher than the premium package: 77% vs. 24%. The lower-priced package (Package B) offers a good-better-best choice of features at $50, $100, or $150 per month, respectively, vs. the premium package’s single price of $500 per month. Performing the same extrapolation of survey results on to the full customer base indicates a projected monthly income for Package B of $670,000 – only a little more than half of projected revenue than from the Package A. Additionally, computation of a current-to-expressed spend ratio is only modestly higher than with the premium package: 0.40 vs. 0.34 (see Figure 2). At this point, the good-better-best package appears to fall short of the premium package in nominal return, despite the high indicated preference.
Reacting to Measured Risk
In choosing among the model offerings, it is necessary to look at the risk of realizing the results predicted by any survey-driven model. Realization risk is the risk of fully translating model-based estimates into actual results. Reliance upon a small number of high-price customers exposes the business to a higher risk of achieving projected revenue and profit if a shortfall in predicted realization levels is encountered than would a lower-priced offering that is chosen by a broader client population.
Risk of revenue/profit achievement is also driven by cost-to-serve levels: the more these costs are fixed, the greater the exposure of net income to realization risk. An expense-based exercise, of course, can become quite detailed as various methods of cost allocation can be applied to different situations.
Nevertheless, examination of a highly concentrated vs. a well-diversified package offering shows how net revenue and gross margin figures can be impacted under various stress scenarios when exposed to a variety of realization risk levels.
Stress testing these projected results involves looking at returns under a variety of achievement scenarios, while applying the knowledge and experience of the business managers to establish acceptable levels of comfort with the scenario outcomes. In the example presented here, the low current-to-expressed spend ratio values of both packages begin to have more of a meaningful financial impact under high cost rather than low adoption scenarios.
Figure 3 shows projected monthly net revenue under a variety of customer achievement and cost-to-serve scenarios for both the premium and good/better/best packages. Customer achievement is presented as the difference from 100% of research-indicated results, while cost-to-serve is presented simply as a range of expense figures which include typical and expected levels, as well as those likely to be encountered only under more extreme circumstances.
As can be seen – and expected, both packages approach questionable profitability levels as achievement risk rises (lower percentage of achieving results) and/or expenses rise. However, despite lower overall revenue and profit levels, behavior of the lower-priced but more diverse package (Package B) tracks quite close to that of the higher priced option (Package A) through all but the extreme levels of achievement risk and rising expenses (in terms of %-difference from base forecast).
On a risk-adjusted basis, however, Package B actually out-performs Package A in nearly all scenarios, when viewed in terms of fulfilling the base forecast. (See Figure 4). The risk-adjusted difference from base forecast is higher for the broader-based/lower priced package in 84% of risk-expense scenario calculations, indicating that returns for Package B have a higher likelihood of realizing projected results than do those for Package A.
Conclusion
The case presented shows the importance of undertaking a deeper-dive into initial research findings. As demonstrated here, what at first appears to be a strong case in favor of the Premium Package reveals a strong consideration for the alternative Good/Better/Best Package when viewed under a risk-based examination. What at first was considered as a clear choice of one offering has now turned into a choice of two alternatives: (1) aggressive – high return, high risk; and (2) conservative – low return, low risk.
With new product launch where uncertainty is high due to the new product or service representing an expansion into truly uncharted territory, a more conservative approach may be subject to far greater consideration than it otherwise would be under less uncertain circumstances. This is where management’s knowledge, experience, and instincts add particular value. The collective wisdom of managing through past economic, competitive, and company-specific cycles offers unmatched insight into thoughtful interpretation and transformation of research results into a product plan. Post-launch customer research and monitoring is vital, and provides the critical early warning system of customer behavior that validates initial assumptions, and influences future direction. This task includes establishment of progress markers, development and adaption of reaction plans, and creation of a rapid and close feedback mechanism with the product’s earliest adaptors.