The usual mistake
Most MVPs are built by subtraction. Take the grand vision, strip features until the budget fits, ship the remainder and hope. What you get is a thinner copy of the final product that nobody is proud of, and when it meets the market it teaches you almost nothing. It answers nothing because it never asked anything.
Start with the question
Every product we ship starts as a sentence with a question mark. Will retailers run live shopping from the store they already have? Will fans pay for a closer relationship with an entertainer they love? Will a driver put a QR code on the windshield instead of a phone number? The first version exists to get that sentence answered by real people spending real money and real attention.
Build only what makes the answer credible
Once the question is sharp, scope stops being a negotiation. Whatever the answer depends on gets built properly, with the polish real users deserve. Whatever the answer does not depend on waits, no matter how much everyone wants it. Quality where it counts, nothing where it doesn't.
Let the answer decide
Then comes the part that takes discipline: believing the result. If the market says yes, the roadmap earns its next chapter. If it says no, you have spent weeks learning that instead of quarters, and the next question is already better than the first. That is what a first version is for.