Since the period of independence Indian economy is witnessing a transition from agriculture based economy to a service led economy bypassing the industrial phase. In the era of economic liberalization this service sector ushered a rapid growth and boosted the overall economic growth to a moderate level. But there is an ongoing debate about the reliability of the service sector output which questioned on the actual sectoral composition of GDP in India. It is argued that the service sector output is over estimated due to inadequate reliable data availability imbibed with faulty methodology. To understand the economic development and the sustainability of economic growth of a country, it is inevitable to know the actual picture of output generated in the economy in terms of goods and services produced. In India, GDP is estimated as the sum of total production of three sectors – public sector, private organised sector and unorganised sector. Due to the non-availability of reliable information on a regular basis, the estimation of GDP for the unorganized segment of manufacturing and service producing activities, which covers around one- third portion of the domestic product, is mostly vulnerable. The GDP for this unorganised non-agriculture part is estimated by applying the Labour Input Method (LIM) as the gross value added by the labour input involved in the production process. LIM considers that the value addition of each worker involved in an economic activity is the same. This implies that all the workers possess the equal level of productivity irrespective of the skill and occupational heterogeneity, which is unrealistic. It may lead to an over estimation or under estimation of the GDP for the unorganised non-agriculture segment of the economy. As a result, the estimated value of GDP as a whole for the economy may deviate from the actual value of GDP, which has serious impact on policy formulation and budget allocation. Thus, it is essential to apply an improved LIM which is able to differentiate the value addition of each worker according to their quality in terms of heterogeneity in skill level and occupational pattern. This study is an attempt to improve the existing version of LIM by incorporating the productivity differences among the workers, which is able to capture the value addition of different categories of workers. Besides this a comparative assessment has been made between the proposed method and the existing one. To examine this, an application is made on private Health Care Service in India, where quality difference among the various categories of workers is vividly observable. Empirical illustration is given by using the information of NSSO (National Sample Survey Organisation) Service Sector Enterprise Survey (ES 63rd round, 2006-07) and Employment Unemployment Survey (EUS 61st round, 2004-05). Besides these NSSO Employment Unemployment Survey (55th round, 1999-00) and projected population figures from Population Census are also used.