Killer experiments are tests done to ensure a product's feasibility throughout its lifecycle.
Here are some common terms that might be used when discussing killer projects, or projects in general:
🔬 Related: Product Development in Life Science
No amount of experimentation can ever prove me right, but a single experiment can prove me wrong.
Killer experiments are tests and questions that will allow a company to stop funding a product idea before it is too late. Killer experiments allow you to save money in the long run by preventing the creation of a product or service that will not sell.
Most startup's business strategy is part right, and part wrong. The trick is to identify the strategies that are beneficial for your business, in order to overcome the bell curve of success. To do this, entrepreneurs need to identify potential deal killers, which are variables that are likely to prove fatal to the venture. Ask yourself: what factors could signify a failed product?
The best way to do this is to start early. Starting early will allow you to prevent spending too much money before it is too late. To do this, you should design experiments (technical and commercial based) that remove risk from the company. These experiments should:
The longer you wait to perform these experiments, the more time, money, and energy at risk. Increasing the delay to start will also increase a company’s willingness to accept the true results if they are negative.
Killer experiments are not only used to halt a particular product’s development. They can also be used in a positive manner! If the results of these experiments are favorable, the data can be included in pitches and presentations to investors or new customers. This can lead to an increase in your company’s valuation as well.
Killer questions are questions meant to kill a project before significant resources are put into development and perfection. Killer questions are the questions used while designing killer experiments.
There can be more than one killer question/experiment, spread out over the development cycle of a product. Frequent killer experiments will decrease the likelihood of sinking too much money into a project just because one experiment had a positive result.
These experiments will intentionally pressure test critical features in as little time (and using as few resources) as possible. Killer experiments do not have to be expensive or complicated! Keep in mind that these studies, the burden of proof is required to demonstrate that your idea is feasible, effective, safe, legal, and does not infringe on any IP.
Killer experiments should be performed all throughout the product development process. The five main parts of the developmental process are:
Killer experiments would ideally be performed first in the ideation and the definition phase. From there, killer experiments are designed similar to check-ins to ensure continued feasibility.
Now that you understand what a killer question is, it may be useful to understand what a killer question is not. Killer questions are not:
One of the most important parts is finding a clear answer. Completing a killer experiment should not leave you wondering. It should have a result that signifies the feasibility of a certain product or service.
Creating a scientific and accurate experimental approach while formulating your killer experiments is very important. There are 3 steps to this process:
Formulate a hypothesis
Test higher “Deal Killers” first
Develop a risk register based on the results
Your hypothesis can be anything you want, as long as the result answers an important question about the health of your business. Some examples of topic areas to test are: pricing, technology, customer demand, competitive response, etc.
One of the most important (yet hardest) things to do is to remain neutral throughout the process. Receiving a negative result in one of these tests can be discouraging, but you must remember why you carried out the test in the first place! You want to find out if you need to pivot and prevent the sinking of funds later on.
Carefully selecting the controls in your killer experiment ensures that you remove doubt within your experiment. Controlling the experiment efficiently will allow you to be more confident in your final results. Reading literature on the market, intellectual property, and past studies to get inspired can often help you design a better experiment.
This example will help you understand the processes of killer experiments. In the early 2000s/2010s, Netflix founder Reed Hastings wanted to test out a hypothesis for his business. His hypothesis was: Mailing DVDs is an efficient way to distribute movies.
In order to test this hypothesis, Hastings sent himself a DVD in the mail. Turns out, it was able to arrive in under 48 hours! His hypothesis was proven correct and he concluded that mailing DVDs would be a viable business strategy.
While this experiment was not perfect, it was able to propel him forward and continue growing his company. But what were some possible negative results that could have told him to kill the experiments? Here are a few ideas of killer results:
As you can see, if Hastings had encountered any of these issues, he may have had to reconsider his business plan. But, finding those hard results is always better sooner rather than later!
Successful killer experiments will remove risk and leave no room for doubt that the proposed product/service is an improvement over the status quo. The model created for the killer experiment must be meaningful to the professionals in that area, or the experiment might be non-convincing. A killer experiment that results in a definitive “kill it” or “don’t kill it” is a success, because it answered the question being asked. The results from killer experiments can also be used in other situations, such as an additional data point when discussing funding.
Q: How do you ask questions to your potential customers?
A: When you approach potential customers to ask about your products, you want to be sure to remain objective. This allows you to obtain results that are not altered in any way. Try asking them about their current products and what works or doesn’t work, instead of explaining your product.
Q: How do you know when to kill your experiment?
A: You will know that you need to kill your experiment when you get a negative result in your experiment, or disprove your hypothesis. These results can be discouraging but it does not mean that you need to give up entirely; pivoting to another idea is always an option.
Q: Is there a resource to contact someone to help you complete a killer experiment?
A: The answer to this question will depend on your stage in the cycle of your business. If you are still a student you can utilize your university mentor. If you are a small business you can utilize services provided by the SBDC, who offers free consulting for small businesses. Large companies may want to look into hiring outside consultants.
This information in this article is from a webinar hosted by University Lab Partners in partnership with the Small Business Development Center (SBDC) @ UCI Beall Applied Innovation, presented by Richard Creager, an SBDC business consultant. View the entire video here.
Richard Creager has more than 30 years’ experience in In Vitro Diagnostics (IVD). He was the SVP Molecular Diagnostics and Chief Science Officer for Beckman Coulter. He also served as Group Vice President for Beckman Coulter’s Immunoassay Business. Richard is a partner in IOI Partners, a management consultant firm for IVD and biotechnology companies. Dr. Creager also held senior executive positions with Kallestad Diagnostics and Sanofi Diagnostics Pasteur. Richard is a life science and tech commercialization expert. He is interested in using his passion to influence organizations in board assignments and consulting arrangements.
Download The Ultimate Guide to Wet Lab Incubators in Southern California, a handbook to assist life science start-ups through the entire decision-making process to find wet lab space.Download Now