Editing Market segmentation (section)

== Identifying the market to be segmented ==
{{see also|Serviceable available market}}

The market for any given product or service is known as the ''market potential'' or the ''[[total addressable market]] (TAM).'' Given that this is the market to be segmented, the market analyst should begin by identifying the size of the potential market. For existing products and services, estimating the size and value of the market potential is relatively straightforward. However, estimating the market potential can be very challenging when a product or service is new to the market and no historical data on which to base forecasts exists.

A basic approach is to first assess the size of the broad population, then estimate the percentage likely to use the product or service, and finally estimate the revenue potential.

[[File:Tv-penetration-us-households.jpg|thumb|300px|To estimate market size, a marketer might evaluate the adoption and growth rates of comparable technologies (historical analogy method).]]

Another approach is to use a historical analogy.<ref>Lesser, B. and Vagianos, L. ''Computer Communications and the Mass Market in Canada'', Institute for Research on Public Policy, 1985, p. 37.</ref> For example, the manufacturer of HDTV might assume that the number of consumers willing to adopt high-definition TV will be similar to the adoption rate for color TV. To support this type of analysis, data for household penetration of TV, Radio, PCs, and other communications technologies are readily available from government statistics departments. Finding useful analogies can be challenging because every market is unique. However, analogous product adoption and growth rates can provide the analyst with benchmark estimates and can be used to cross-validate other methods that might be used to forecast sales or market size.

A more robust technique for estimating the market potential is known as the [[Bass diffusion model]], the equation for which follows:<ref>Mauboussin, M.J. and Callahan, D., ''Total Addressable Market: Methods to Estimate a Company's Potential Sales,'' [Occasional Paper], Credit-Suisse – Global Financial Strategies, 1 September 2015</ref>

: N(''t'') – N(''t''−1) = [''p'' + ''q''N(''t''−1)/''m''] × [''m'' – N(''t''−1)]

Where:

* N(''t'')= the number of adopters in the current time period, (''t'')
* N(''t''−1)= the number of adopters in the previous time period, (''t''-1)
* ''p'' = the coefficient of innovation
* ''q'' = the coefficient of imitation (the social contagion influence)
* ''m'' = an estimate of the number of eventual adopters

The major challenge with the Bass model is estimating the parameters for ''p'' and ''q''. However, the Bass model has been so widely used in empirical studies that the values of ''p'' and ''q'' for more than 50 consumer and industrial categories have been determined and are widely published in tables.<ref>See for example, Lilien, G., Rangaswamy, A. and Van den Bulte, C., “Diffusion Models: Managerial Applications and Software,” ''ISBM Report 7,'' May 20, 1999.</ref> The average value for ''p'' is 0.037 and for ''q'' is 0.327.