[carlosjobim]

> There’s a lot of entitlement and craziness from paying users too, and those are harder to ignore.

Somebody paying for your product is very strong signal. You know that such a person represents real world use cases for your product, and that their issues and feature requests are based on real world problems. Otherwise the chances are low that they would be paying for the software.

So helping them with what they want could mean that you've just tipped the scale enough for hundreds or thousands of people to become new customers.

And of course you should give them their money back to get rid of them if they're any kind of headache. Or tell them that their requested feature will be in the next versions, which is a new purchase.

[latexr]

> So helping them with what they want could mean that you've just tipped the scale enough for hundreds or thousands of people to become new customers.

Sure, if the request is reasonable and sensical. I entertain those and even help them formulate the request better if needed. That’s true of both commercial and open-source.

But I’m more talking about the users who demand features. Those who say the tool needs to have whatever idea they just thought of 2 minutes ago, despite no one else ever having asked for it and it not really making sense. Those users who only think of themselves and suggest features which require fundamental changes which would modify the behaviour for everyone, or the feature is in itself contradictory and there’s no way it could work.

[austinthetaco]

No other customers asking for the feature is not a signal that its not wanted by many others. Other people suggesting the same thing is a terrible measurement, as suggestions from paying users is already an incredibly small amount of users and some users may just thing it would be nice but not enough to warrant sending in a request. For example: it would be nice if I had a one-click export of my favorites on HN to a markdown. I'm not going to write to them to suggest it because it's not a big deal, but that doesn't mean it wouldn't improve my experience using the the product.

[latexr]

In other words, you’re agreeing with me. Again:

> Those who say the tool needs to have whatever idea they just thought of 2 minutes ago, despite no one else ever having asked for it and it not really making sense.

If no one is bothered enough by the lack of the feature to mention it, then the software doesn’t need it. And you’re ignoring the “and it not really making sense” part. If a feature is well-reasoned and makes sense in scope, I always consider it.

[austinthetaco]

need is a weird thing to hinge off of as software itself doesnt have "needs" and most software thats made doesn't fulfill a true pure "need".

[carlosjobim]

Suggestions from paying users are the only suggestions which have any value, since they represent the type of people who are willing to pay for the software.

If you have users who aren't paying you, there is no reason at all to pay any heed to what they say.

[predkambrij]

My favourite is the advice I read from personal MBA. You send them to your competition :)