OBJECTIVES: Value of information (VOI) analyses are performed to assess opportunity cost of uncertainty, usually from a payer or societal perspective. The aim of this study was to simulate the Expected Value of Sample Information (EVSI) from a manufacturer’s perspective and study the factors affecting the EVSI.  METHODS: A model linking the probability of acceptance of a new treatment to the incremental cost effectiveness ratio (ICER) and uncertainty around it was assumed, based on acceptable ranges of ICER reported in the literature. Uncertainty was characterized as the probability of ICER being above the willingness-to-pay (WTP) threshold of £30,000/QALY. Different scenarios describing the sensitivity of the approval chance to the ICER and uncertainty were considered. The manufacturer was assumed to earn a revenue of £20 million over 10 years if the technology got accepted.  EVSI was calculated for several values of prior ICER and uncertainty, using 10,000 simulated trials with true values of incremental costs and effectiveness equal to the prior means. RESULTS: Results indicated that EVSI varies substantially according to the prior ICER, probability of cost-effectiveness and the approval function. In general, when the prior ICER is close to the WTP threshold and the chance of approval is greater than the probability of cost-effectiveness at the threshold, EVSI is negative.  When increasing the sensitivity of the approval function to the ICER (thus the policy maker being more critical on ICER value), EVSI is more likely to be positive.  CONCLUSIONS: Value of information analysis could be useful from the perspective of manufacturers, but results are very sensitive to the parameters characterising the probability of acceptance function. Further research is recommended to quantify the probability of acceptance of a new technology according to cost-effectiveness results. The framework described here should be extended to allow for decisions of coverage with evidence development (CED).