It's impossible to completely eliminate your biases while building products. Focus on identifying them and then find ways to counteract them.
A key step for launching great products is identifying and counteracting the biases which you (the product team) and other parts of the company inherently have.
Since product managers are the layer which facilitate communication between different parts of the company (Engineering, Design, Marketing, Support, Pricing, Distribution) they need to learn how to identify their own biases first. Then they need to attempt the herculean task of figuring out the biases of other people involved in the product development process.
Why is it so hard? Because bias shows up in just about every decision we make.
Fortunately, the same neural structures that trigger and reinforce bias are closely tied to the ability to self-regulate. Biases show up in certain predictable patterns and you can train yourself to identify these patterns
While there are many biases you should guard against, I have focused on 2-3 of them that product managers encounter very frequently
Confirmation Bias + The Bias to build
I think both these biases are closely related and something product managers face on a regular basis ( every other sprint ! )
Confirmation Bias: The tendency to think new information proves our existing beliefs. Every one of us has our own worldview. This results in us only seeing the opportunities related to the status quo. We end up missing out on the more-valuable opportunities. As we go building stuff quickly, sprint by sprint we need to pause and see if at times Agile methodologies are actually reducing divergent thinking?
I am someone who has been a huge practitioner and proponent of A/B testing things and then launching the "winners". But I feel it still needs to be pondered whether A/B testing, or weekly / bi-weekly sprints really lead to the next path breaking product. Yes they help in iterating and improving things. But would they lead to a zero to one product? So it sometimes makes sense to move away from the routine of building products (features) at a breakneck speed. Because a lot of your tests will still end up "confirming" your existing beliefs.
Bias to build: The other related bias is a bias for action or a bias to build. As a product managers you are expected to "do" something each sprint or as a product leader / founder you expect your team to deliver something every week. So we never let go of the opportunity ( sprint) to build something new. We assume the tests or the usage data will help us figure out what is working and what is not working.
It makes sense to spend some time trying to understand the problem at a deeper level. Allocate yourself a "research"story in some sprints and let the engineers work on clearing out some tech debt.
It is not an easy thing to do as you are also being judged on the quantity of new stuff you roll out. The leadership team needs to have the maturity to understand that the "research" is also work. As a founder or as a product team leader I have been guilty of exhibiting these biases too. In fact now that I think of it there have been sprints where I felt constant progress was rolling out something each sprint. Who wants engineers to work on tech debt or sit "idle" :). But if product managers need to spend some time digging deeper maybe sometimes engineers could also be allocated a story to evaluate / understand a technology better rather than just build build build !
Authority Approval Bias
This means building / testing stuff that will agree with the world view of key decision maker - your VP / Founder or the team which gives you go / no go approval to launch. Of course it is the easier option. But also a very sub optimal way to build. As a product manager question yourself whether you get any objections from the powers that be when you are launching something new. If the answer is hardly ever, then either you are one of the best product managers in the world or you have just been building what they want ! Just be aware of this and be prepared to "fight". The best product ideas would not be obvious to everyone in the company, and if they are then there is some issue.
A weird tangential authority approval bias is launching ( copying) something which the biggest player in your industry has launched. Then somehow the whole company will agree that it is a good idea :)
-----
When you learn how to notice these biases then you may be able to make deliberate choices that counteracts them.