I've had customers complain to my face that we tailor the products to our needs, and are surprised (and sometimes vocally disbelieve) that we don't use it, ourselves.
In certain industries, dogfooding is considered bad, and sometimes illegal.
Isn't that the worry? Say a bug in your timekeeping software underreported hours worked and that lead to workers not being paid their agreed upon rate. When it goes to court you now have to prove that you didn't maliciously introduce the bug to save on employment costs, whereas if it is a third-party providing the software it's their problem. "Conflict of interest" is ultimately just another way to say "I don't want to assume the liability when things go sour". Nobody cares about conflict of interest when things go right.
Again, "conflict of interest" is what people often say when they don't know if it will affect them, but don't want to take the risk. "Regulation requires it", as illustrated in the first comment, is what people say when caselaw actually demonstrates that they would be liable.
The fun thing about language is that it is fluid and can take many forms.
No, it's not.
It's "malice is not a reasonable explanation for things going wrong". It only changes liability when it follows malice, what is much rarer than you seem to think.
It is also a way to assure people that you won't do something bad.
It cannot possibly be the case that Microsoft runs on Open Office and MacBooks.
They only started eating their own dog food when people started making fun of them in public.
They use Zoom & Slack internally, not their own video and chat offerings.
Dogfooding is a good thing.
You’d be surprised how few companies actually do it.
I can tell you that devs that use their own product deliver better quality.
Perhaps you also need to use the product that your competitors use to understand what you need to beat.
EDIT: I see above that others have claimed the opposite, but my experience has been different. I simply cannot take a call from a cafe in Zoom without complaints, but Meets is regularly tolerable, according to those I meet with.
That gets repeated a lot but I have observed the opposite when you have a diversified company dogfooding an internal product that isn't strategic in their portfolio, or is targeted at a market segment that the company doesn't belong to. I have seen companies hamstring themselves by using a product that isn't the best offering for them or a poor technical fit, only because it is their own offering. Also, in the worst case, companies become develop tunnel vision in the market, because they don't regularly use the competition.
If you have a large, international software company targeting small to medium business customers in the US, dogfooding would be counterproductive. It would probably harm their strategic customer base by overcomplicating their product with features they don't need at the same time it slows the parent company down by using a product that's poor fit.
The idea behind dogfooding is that you get more/better feedback and more internal pressure to fix problems.
But the product team should already be getting feedback from a diverse set of customers. And if they're not, that's what needs to be fixed.
Dogfooding can also result in overprioritizing only your own company's product needs, and at worst the ultra specific issues that one particularly vocal VP is having, to the detriment of the overall customer base. (I've seen it happen way too many times.)
Companies should use the best tools for the job, not necessarily the ones they make themselves. If those are the same, then great. If not, then also great.
Not a company, but OpenBSD dogfoods exclusively. And they do deliver a good product.
It might be a goal to move everything internal (I believe Google is now using Google Apps internally, but they didn't for awhile, even after other businesses were).
For example, Stripe is a _very_ large company but their product is geared mostly towards small and medium size businesses.
It's very possible that in the process of making Stripe Billing work for Stripe, they'd make it work less well for their actual customers.
It's the same reason why Intuit probably doesn't use QuickBooks to do their accounting.
When dogfooding, you have to be very careful that you don't end up in a bubble and only see your own needs as the priority.
There's a propensity for developers to start using their own products then changing it to suit themselves thinking that the regular users use the same product the same way. You end up with this mismatch of expectations and it can even get so bad that the developers don't believe the users who tell them that things need to change.
However, I find it can get tiresome / has potential compilations.
Customer's can be terrible with incomplete, bad idea, and nonsensical requests. But sometimes internally you get the worst "Someone spilled some words into email that don't sense / no you don't actually want that / bro this is accounting software not Photoshop." situations that folks who are more familiar with the origination might make, but customer's might not make.
It takes an extra level of internal discipline to deal with dogfooding side effects at times IMO. Still, a good idea none the less.
+ you get to keep up with their advances so you’re not relying on your customers telling you about your competition
+ in the case of chat apps, if your infrastructure goes down you can still work as a team to coordinate the remediation
I've actually had an instance or two over my four years there where I discovered a bug in our PROD/App Store version myself; simply from using the software enough.
It's really unfortunate that this isn't the case for a lot of dev shops. Dogfooding is incredibly important to QA.