That's not the point; you can't say their advice is ineffective, while also condemning them for being effective.

The top comment derisively states:

"The whole business is a scam built around sending in some 26 year-old with an Ivy League degree and $100 haircut to regurgitate snippets from articles in HBR"

In the context of fining the company $500MM for being too effective.

At least in the US, where the punishment is relative to the crime; the court clearly agrees about their efficacy.

I'm not sure there is that much dissonance. In the opioid case, they gave effective advice, but it wasn't necessarily of very high quality - the reason you wouldn't do what McKinsey suggests isn't because you didn't think of it but because you would find immoral and illegal.

The reason McKinsey was being sued isn't because they were giving effective advice, but because it was likely illegal.

Because of this I'd understand that both can be the case. That said, I'm not sure McKinsey is always as ineffective as the top comment suggests.

You don't need groundbreaking innovation in sales from your consultants to keep selling too much opioids, a product that mostly sells itself. Better consultants should probably try to stop you, however.

Their advice was highly effective in bringing the company to bankruptcy, if that's the metric you were aiming for. Aside from destruction of private and public equity.

To play devil's advocate, they could be both ineffective and harmful. Just because they succeeded in getting the company to do something, that not mean it was the right thing. The sale of the drugs could have produced a net loss as well as harmed the people who bought them.

Ineffective = “having no effect”

Harmful = “having a negative effect”

How can they be both?

You can be effective at selling opiates to people, but then end up in an investigation for selling them illegally and thus neutralising any of the revenue you had initially brought with fines for doing so illegally. Here's no effect.

Of course a lot of brutal strategies work well in environments which do not expect them. However trusted environments are often more efficient.

With a comma? Ineffective doesn't mean no-effect, it means 'not the intended effect'.

I.E. "McKinsey worked with Purdue to implement an ineffective business strategy that resulted in bankruptcy for the business, due the harmful nature of the product and the predatory nature of the strategy."

Sure, but I would think generating significant amounts of revenue would tilt the probabilities towards "consultants do actually do something".