Ask HN: How do you talk someone out of a bad idea?

[jaxn]

I have some friends in a marketing company who have created their own content management system. The programmer actually works for one of their clients and that client trades marketing services for programming services (I assume they are not loaning out their best programmer).

So now they have a CMS that is constantly needing to be updated in order to meet the needs of their clients. It often takes about 7 seconds to load a page (there is no cacheing, probably a bunch of redundant and unoptimized queries). They have also branded it under a pretty generic sounding name and a Google search reveals a bunch of other products, but not their product.

I have suggested that marketing is their core competency and not software development. I have suggested using open source or white label CMS systems and even branding those with their process, etc. They tell me that they have invested too much to turn back now (sounds like chasing bad money with good to me).

I don't want to tell other people how to run their business, but this seems so black and white to me.

What would you do? Have you seen someone make this case really well somewhere?

[AmberShah]

This sounds like a case of: you can't save people from themselves". You've stated yourconcerns and they've decided to go all-in. There's too many bad ideas out there to vet personally invested in them

[coryl]

I would say you just have to express how bad of an idea it is in one shot. A serious sit down and chat, letting everything go. If after that they still want to do it, then really, its up to them to fail and learn on their own.

[justsayin]

"You know when one of your clients thinks that they know better than you about marketing matters? They don't appreciate the bigger picture, they don't understand as much as you do, or even that there is more to understand? They're suffering from known biases that are obvious to you, being outside the situation? If they'd only listen to you and take your advice they'd be so much better off, but they're too closed-minded to see that? They need to just trust someone who knows what they're talking about - you - and let it go.

Well, that's exactly what you're doing with your CMS. You are that client, you're doing exactly the same thing. Listen to someone who knows what they're talking about, me, and let it go."

Disguise the second part - get them really involved in joining in with the first part, get them to recount examples, etc... leave them nowhere to run when the second part hits 'em.

[jaxn → justsayin]

I assume this was a throw-away account for a reason. Do you have experience in this situation? Can you share more about how that played out?

Regardless, I think this is a worthwhile strategy.

[scorpioxy]

I don't. Leave them to do whatever they want.

Some people are just not interested in listening and so I gave up trying to talk to them and now only offer my opinion if asked, even when it comes to computing. All they're looking for is someone to agree with them, even if it isn't the right thing to do.

As for this particular situation, I haven't seen it succeed before. What usually happens is the project becomes a money sink-hole for the company and you'd get called in to try and save it.

[timinman]

The idea of having your own web-startup is so powerful that making money from can be an afterthought. Even though they might not admit it right out, they may want to do this just to get in the game, regardless of their likelihood to succeed.

[dnsworks]

This sounds a case of: mind your own business. Sometimes people will only learn the hard way.