Strategic Alliances Fieldbook newsletter #17 – blueprint addressable market

Addressable market is a frequent topic for alliances.? How big is the market for services related to a technology?? How big is the market for services for a specific proposition enabled by a technology?.? How much of that market could we expect to win given our capabilities and market recognition?? And based on that how much should we invest to capture that market?

A mature alliance will be asking these questions permanently to make the right decisions to launch a new offering, invest in scaling an existing one or retiring an offering that has not got market traction.? This excerpt from the book deals with the general principles of calculating addressable market and gives a real example of one analysis.

If you enjoy the newsletter please follow me, subscribe to the newsletter, like, comment and repost.? The book is here if you’d like to read the full text.? https://www.amazon.co.uk/Strategic-Alliances-Fieldbook-Art-Agile/dp/103212900X/ref=sr_1_1?keywords=strategic+alliances+fieldbook&qid=1691319592&sr=8-1

The addressable market analysis is intended to be a high-level estimate of the amount of mutual value of working together. Part will be the revenue from the technology products and part will be the professional services (PS) work that might be available associated with a technology company’s products. Several analysts offer market analysis services for a fee, and there are a few ‘rules of thumb’ in the industry that are helpful. For example, it is common to assign a ratio to the relationship between tech and PS income. It’s typical to assume every dollar of technology sales will translate into between 2 and 5 dollars of services spend. Paying an analyst firm to produce an analysis will be the most comprehensive, but will carry a cost and will also entail a well-thought through value proposition because they will need to know the details of the offering to estimate the value. The second option is a variation on simply multiplying the technology firms forecast revenue for a relevant product by an industry ratio. It is quick, but obviously less precise. The authors have developed a model that lies part way between these two approaches, which has been used to guide PS firms on the potential value of services work and indicate to technology firms how many certified delivery experts they need to have in the marketplace. The example below forecasts the growth in technology revenue in the light grey bar on the left and the associated services potential in the right-hand bar. The line forecasts the numbers of experienced PS people required, which is useful for the tech firm enabling them and the PS firm recruiting them. This is an example based on modelling done in 2020 that forecasts the tech firm’s target sales of $2.6 billion in 2023 will create a services market of around $3.6 billion that needs over 8,000 delivery staff (Figure 7.1).?

This approach makes several assumptions and applies them to extrapolate from the forecast technology revenue values to forecast the PS addressable market. It then assumes the blended day rate of the PS firm to derive the numbers of people. This model uses an assumed ratio for basic deployment and a higher ratio for more transformative services. The basic deployment will need solution architects and engineers for configuration work. The transformative services will require more business analysis and functional skills in addition to the technical skills, which is the reason it has assumed a higher ratio because it will pull through more professional services effort and therefore higher services:tech revenue ratio. To illustrate the difference between the two service types, imagine in the first the client has gone through the process of designing their business and technical architecture, chosen a product and has the in-house expertise for functional design and project management. Their request to the PS firm will go something along the lines of ‘please can you provide x number of engineers to configure the product in abc way’. This is a technical implementation request where the client is taking the risk of product selection and architecture and payback, and just needs qualified technical people on a resource augmentation basis for configuration and coding. In the second scenario, the client has recognised a business problem and needs the consultant to investigate, diagnose, design a solution and a business case and then work out how to implement successfully. This second situation is like the digital transformation example from Chapter 2; it will require a wider range of PS skills, more effort and higher risk for the PS firm, which is why the model distinguishes between these two scenarios and estimates higher for the second.

Next, the model assumes that the technology company itself will provide a proportion of the PS staff, and that the client can also satisfy some of the demand via permanent staff or individual contractors augmenting in-house staff. If the technology firm has a big PS team of its own and is active in the market, and the technology has been around a long time, meaning there is a large pool of qualified individuals, then the client can access a high proportion of the skills without contracting with a big PS firm. These two factors are material in estimating the addressable services market as it removes that part of the anticipated demand that are most likely to be fulfilled by others. On the topic of competitors, the proportion of the addressable market a new entrant should anticipate capturing should be ratcheted up if their competitors are less well known and less credible, and their offering well differentiated because of their increased appeal to customers. Following that, the model distributes the PS values across multiple years, Because there is a compounding effect of incremental years’ growth as the chart illustrates – the services market is accelerating slightly faster than the technology market. This is due to two factors. One is the parallel effects of Moore’s law competing away technology unit prices and the increasing effect of inflation on labour rates. This means, all other things being equal, the same product gets more expensive to deploy over time because labour costs rise. The second is that once a technology is deployed, it needs to be managed in life. If a company buys technology every year for five years, then at the start of the fifth year, there is one year’s worth of new tech to implement and four years of tech to manage and maintain. This effect is less relevant if the technology is cloud consumption or Software as a Service (SaaS) licenses, because if you stop paying for the technology, you can’t access it and therefore don’t need people to manage it. The effect is very marked with perpetual licenses and hardware as you pay for it once and could choose to manage it for many years. So now we have built upwards from the forecast technology sales using assumptions to extrapolate a PS addressable market for deployment and run services valued in dollars. Finally, we can assume a blended day rate to convert the dollar value into number of days. Given there is around 220 working days a year after holidays and average sickness leave, simply divide the number of days by that and you have a forecast number of people required to implement the technology revenue target we started with. This process of assumption stacking has an obvious effect on the potential accuracy; however, the point of this exercise is not to imagine it’s possible to find a forecasting crystal ball that can get you to the nearest decimal point. The exercise is simply to sketch out a rough order of magnitude. Are we looking at a five-million-dollar market or a ten-billion-dollar market? Does that equate to 50 people or a 1,000 people we need to recruit and train? Having a model with drivers that can be adjusted to run scenarios makes the initial communication about the potential size of the prize much easier, and therefore the conversation about investment and reward more tangible to the decision makers in the alliance.