[raveenb]

My personal experience is that traditional software estimation techniques don’t work when done for a MVP situation. Yes it’s frustrating like hell to not know how long before your money, patience and interest runs out. In a non MVP situation some estimation models work, some 40% of time. But managers make it look like they work all the time. Cutting scope to hit deadline doesn’t count.

[raveenb]

Also the Viable part of MVP is subjective. As you build the MVP, your idea of what is viable changes. This needs to be accounted for. The minimum product has to be viable for a paying customer. If the MVP is not a paid product the you can get away with less polish, not so with paid products.

[davnicwil]

Exactly what happened with Box CI :-)

The viable I had in mind was 'it works' - later realised that actually viable was 'I think there is a chance I can ask people to pay for this' - turned out that second part around doubled the already overrun build time!