Biotechnology Cost Drivers

October 22, 2018 |

By Mark Warner, PE, Founder, Warner Advisors LLC
Special to the Digest

Accurate cost forecasting for first-of-a-kind biorefineries is both challenging and necessary to achieve commercial success.  At its genesis is model development, which enables the brilliant concepts produced in the lab, and pairs them with reality of building commercial scale technology. Insights from both the benchtop looking forward and commercial-scale looking backward follow:

Breathing your own exhaust is harmful to company health –Optimism is abundant within advanced biotechnology, but can generate headwinds when it becomes exuberance during fundraising.  This becomes a problem when the high level of optimism keeps the company from fairly evaluating its technical challenges (aka, breathing the company exhaust).   Remember, just because you assume risks away in a financial model, does not mean they actually go away.

Focus on what matters – Techno-economic modeling is a key offering of my consulting practice and I am an advocate for a streamlined approach.   A credible first pass for fermentation-based processes can be done with a calculator and whiteboard, as the key criteria are easily identifiable.  My approach is simple for a reason, once you get past the top 5 costs factors (product yield on carbon source, labor, depreciation, media costs and utilities for example), not much else matters in the early phases and focus elsewhere can waste scarce company resources with low return on investment.  The top 5 cost factors will typically represent over 90% of manufacturing costs and significant efforts at modeling minor variations are usually a waste of time.  Focus on the big items and sweat the small stuff later.

Challenge historical perspectives, but don’t ignore them –One underlying theme common within start-up ventures is to challenge the status quo, which I fully support.  Challenging the status quo can be a mistake when there is a lack of understanding the current practice and a broad, un-substantiated assumption it can be done differently.  If you have a plan to build a better and cheaper mouse-trap that’s great, but make sure there is an understanding of what is available today and how the proposed version will actually perform better.  Change for the sake of change, or just driven by the “everything must be invented here” mentality seldom ends at the desired target.  Resistance to utilize readily available materials, that are cost competitive, is a common example of this.

Benefits of building large facilities – One of the most common lessons learned from the last round of biotechnology commercialization was that companies built too large and too fast.   While I agree, there were reasons for going big that need to be understood. Building smaller facilities requires less capital investment, but brings with it higher per-unit production costs. The cost advantages of scale are real and labor is a prime example.  While a commercial facility often has production levels 20 times larger than its demonstration-scale equivalent, it would normally only require 5 times as many staff.  The end result of a larger facility is a much lower unit cost as the capacity of production increases.   The figure below shows typical cost breakdowns per kg of product for pilot, demonstration and commercial scale.

Don’t go chasing waterfalls –Detailed sensitivity analysis of early stage financial models are often not a good use of resources.  I have seen significant workups with waterfall charts and tornado diagrams for models that have an overall accuracy range of +50%/-25%, where the individual components being analyzed do not vary within the accuracy of the analysis.  It at times has reminded me of a cat chasing a laser pointer, tons of activity, but no tangible result.  Meanwhile, limited resources are spent on the technical issues like yield improvement, that drive the highest impact on cost.  This is a point where the 90/10 rule applies, the vast majority of the early resources should be placed on verifying and supporting the key technical assumptions that go into the economic model and limited effort towards doing deep sensitivity analysis of variations that are within the accuracy range of the overall analysis.

Make sure the levers you plan to pull actually work –Levers to be pulled in a financial model are key parameters such as alternate feedstocks, byproduct valuation and lower product specifications which allow for improvements to the economic model.  Looking back from the perspective of having built and operated commercial-scale facilities, many of the economic levers assumed in the early stages do not turn out to be an option at commercial scale.  Lower cost alternative feedstocks that have not been fully demonstrated to perform are a common example.  Early work to identify factors that will have a material impact on ultimate facility operations is effort well spent.

Following the basic principles outlined above will produce techno-economic models that more accurately represent commercial reality and outline a roadmap to targeting development efforts. This will significantly improve probability of commercial success.

Mark Warner is a registered professional engineer with 30 years of experience in process commercialization, focusing for the last 10 years on taking first-of-a-kind-technologies from bench-top to commercial operation.  He is the founder of Warner Advisors, providing consulting services and acting in interim engineering leadership roles for advanced bioeconomy clients.  He can be reached at [email protected]or visit

Print Friendly, PDF & Email

Category: Top Stories

Thank you for visting the Digest.