Innovation Risk & Return: Horizons, Uncertainty and the Teddy Bear Principle

by Kevin McFarthing

I’ve had a couple of interesting conversations recently focused on the strategic view of innovation using the 3 Horizons framework originally proposed by McKinsey, and the assessment of risk and return from innovation.

The 3 Horizons approach can be used to manage different areas of future business concurrently, albeit with different managerial approaches given the different requirements and characteristics. It can give a view of how innovation and other business-building activity such as M&A could extend a company’s reach into new sectors (Figure 1).  It can provide a time horizon to business growth in new and existing areas. It can give a perspective on the potential for new technology platforms and new markets (Figure 2).

Figure 1 – Three Horizons (Source: – See more here via @Ralph_Ohr

Figure 2 – a 3 Horizons matrix defined by knowledge of market and technology (adapted from @TimKastelle).

Simplistically, all you need to do is to put different labels on the axes of the nine-box matrix to create different types of horizon for the purpose at hand.

Of course, the real value in using any management tool or thinking aid is the decisions that are taken as a result, for example, the split of resource allocated to projects across each of the horizons.  A highly technology-oriented company such as Google reportedly takes a 70/20/10 split for Horizon 1/2/3, as shown in Figure 2.  Companies in less breakthrough-oriented industries such as consumer goods (CPG or FMCG) are likely to have a much higher allocation to Horizon 1 and almost nothing in Horizon 3.

Which brings us to risk and return, where both are often joined by their sinister sibling, uncertainty.  The green arrow in Figure 2 represents a logical assumption of both return and risk – the further you delve into the future and the unknown, the higher the risk; but also the higher the return that is expected. Horizon 1 is where most managers feel comfortable.  It’s where the highest degree of apparent certainty exists and, intuitively, the lowest risk.  It has the highest proportion of incremental projects and rarely presents a “bet the business” option.  Horizon 2 has higher corporate uncertainty, and Horizon 3 presents high corporate and industry uncertainty.

But here’s something counter-intuitive – estimates of the number of new CPG products that fail within two years vary; it could be 70-80%, 85% or as high as 95%.  Other industries have lower rates of innovation failure with the overall number estimated at around 40%.  Given the very high proportion of Horizon 1 incremental products in CPG, and that the key use of the Horizons is to influence the strategic management of the portfolio; this doesn’t look like low risk to me.  Or if it is, it certainly doesn’t represent a good return.

There may be valid reasons for failure, but the numbers are so high in CPG it’s hard not to suspect a systemic issue where the assessment of risk and return isn’t working.  It may well follow the “Teddy Bear” principle; it doesn’t do anything, but it helps you to sleep at night.

Don’t misunderstand me; I’m not saying that these products shouldn’t be launched.  It’s rather a need for more openness about the real risk, and an acknowledgement that part of the business model is, to put it bluntly, a “let’s launch and see what sticks” approach.  It’s resonant of the famous remark by Lord Lever – “50% of my marketing isn’t working; I just don’t know which half.”  It looks like a higher percentage of projects will fail the risk/return assessment; we don’t know which ones until they actually succeed or fail.

It begs the question what the failure rate of Horizon 2 products really is, and whether they really do represent a higher risk.  Given the failure rate of Horizon 1 products in CPG, could there be a stronger role for Horizon 2 options.

It’s well accepted that an ambidextrous approach to the management of opportunities in different Horizons is needed.  It would also appear that there can’t be a uniform way to assess different types of risk posed by each horizon.  It’s therefore important to understand the real risks each company faces in each Horizon, when by definition there will be benchmark information for Horizon 1, less for Horizon 2 and almost none for Horizon 3.

So perhaps it’s time to reframe our approach to the uncertainty of risk and return from innovation across different horizons. We shouldn’t assume that low risk automatically equals a good bet.  And we should gently take the Teddy Bear away.

image credit:

Wait! Before you go…

Choose how you want the latest innovation content delivered to you:

Kevin McFarthingKevin McFarthing runs the Innovation Fixer consultancy, helping companies to improve the output and efficiency of their innovation, and to implement Open Innovation. He spent 17 years with Reckitt Benckiser in innovation leadership positions, and also has experience in life sciences.