How do you know you’re conducting thorough enough market research, in order to reduce the amount of unknown unknowns?

It's important to take an iterative approach to customer research that spans the product development cycle. No single research study can anticipate and answer every potential customer challenge or angle, so it is important to build iteration into your plans. I follow 4 main stages of research: 1) finding opportunities, 2) evaluating concepts, 3) UX Research and Usability, and 4) market acceptance research. 

Throughout each step, it is important to constantly revisit the core question - which is are we solving a real customer problem? Second, it is important to understand what product decisions and customer behaviors would make or break your launch. Priortize getting high fidelity signals on those items.

1) Finding product opportunities should be a consistent stream of research. These can be composed of win/loss interviews, NPS surveys, customer service calls, and ongoing customer interviews. The goal here is to surface pain points and identify any emerging dimensions customers may be using to evaluate your company relative to competition
2) Once an opportunity is identified, I move to the concept validation. In this phase, the focus is on testing basic concept ideas and how customers value that feature. Surveys can provide quick signals on receptivity. After you feel you have a receptive audience, I move to focus groups to help prioritize features and understand high level requirements. These groups also can help inform how much of an irritant the defined customer problem is and what is their motivation to change behavior.
3) UXR is where you get even more in depth into the customer acceptance criteria for the new feature. You will get more clear signals about the actual likelihood that the concept you've developed can be usable by customers and ultimately solve a problem for them. 

        4) Lastly, it's time to put the boat in the water. Market acceptance research and testing involves having a small group of customers test drive your feature. It is important to observe critical KPIs on behaviors and downstream results, but also to incorporate qualitative interviews into this phase as well. 

Everyone in your business knows something about your market, and your customers. So its not only important to do the research but to make sure you share it with enough people in your business to validate that you dont have any blindspots or other elements that you’ve missed.

You dont need to have all the answers yourself but your work does need to interpret what the research means, so even though there are unknowns, focus on being thorough around the ones your team/business has the most control over and responsibility for. There will always be things you cant control (the pandemic is a good example).

Great question, and I love discussing topics like "unknown unknowns." My primary advice would be: Don't mistake "thorough enough" for "specific enough" when it comes to market research. The most thorough research in the world can still have significant blindspots if it's not specific enough. In other words, it all comes down to the questions used to preface your research. If they're overly broad and vague, the best research in the world won't capture the relevant data. 

Begin any research initiative knowing what question(s) you ultimately want to answer, and what you will be able to do with those answers. If you start with "I want to know as much as I can about Demographic X so I can better market to it" the outputs may tell you a lot... of the wrong, or least relevant, information. If, on the other hand, you start with: "I want to know how Demographic X evaluates the switching costs associated with enterprise software, so I can justify the necessary behavior change" you'll be able to thoroughly investigate the pivotal issue at hand.