"Do we go and build X or do we go and build Y?"
You’ll notice each time this question is asked peoples’ eyes tend to light up like a Christmas tree — with whichever assumption their brains can come up with first. Now assumptions get a bad wrap, some thought leaders say they should be avoided like the ‘plague’ . I’m embarrassed to admit that I felt the same way for far too long, but very quickly not making assumptions leads to indecision, and a lack of progress — due to the fear of jumping to conclusions. Without getting on my soapbox here I want to highlight the grave opportunity available to us when we make assumptions - assumptions are only bad when they are left untested.
Lets unwrap this a little bit — where do assumptions start?
Usually an assumption originates from a limited set of evidence — ones mind, from growing up in a small town, by not diversifying ones experiences. If you’ve experienced any of these cases you’ll recognize how easy it is to start building ideas and patterns up in your mind — which we tend to want to repurpose as generalizable rules that we can reuse and repurpose elsewhere in our lives. This, is when the problems start — when we try to take one idea which although limited; has evidence to support it, and try to apply it on a large set of cases.
Lets give an example; The humane AI pin which raised $230 million to support their product sold behind the idea of “Take AI with you everywhere” — this idea disintegrated once famous tech review Marques Brownlee called it "The Worst Product I’ve Ever Reviewed”. Soon after the company sold to HP for nearly half of what it had raised at $116 million.
Now as product builders, how do we avoid another Humane AI product fiasco? Well that is where testing comes in, see at some point the people at Humane AI said to themselves ‘this product will create enough value for users to warrants its 700$ price tag” which frankly; never happened. Investors may have seen it that way, the humane product team may of seen it that way, but the market and its pundits definitely did not see it as worth the price whatsoever and unfortunately those are the only ones that count.
Learn from Humane’s experience so you can avoid the same mistake. Test your ideas before you build them, and do so rapidly. We need to make sure the market wants what we are building and pivot or adapt when ample evidence against them wanting it. But an unchallenged idea is definitely a dangerous one — and most ideas supported by anecdotal are just this. That is what we aim to avoid with product validation , challenge your assumptions early on to ensure that what were building will actually meet user's needs.