Very much so, in fact Close.io <http://close.io/> was built out of our own sales team's need for better sales software / we hated all the CRMs we tried. Then it turned into its own product. We would have never known how to make Close.io better than existing legacy CRMs except for seeing the pain our sales teams actually had while doing sales.
Now the entire focus of the company is on this internal-project-turned-product and now we get to use Close.io to sell Close.io, which is great. We get to solve our own problems (usually) and others benefit from it (and our product gets better).
Same with CircleCI (https://circleci.com). We were running CI and then CD for ourselves practically from day 1. Everyone who writes CircleCI uses it every day.
It's great to dogfood, but IMO its not enough. Many customers will have different ways of using the product from you and different workflows. I'd recommend doing heavy customer development very early on to understand those differences.
We needed to develop both prototypes and production apps faster, so we built tooling...which turned into Exponential.io (tooling as a service).
> seeing the pain our sales teams actually had while doing sales.
This is a great way to find a product. At a minimum, you know you're solving one company's problems (your own). As it's likely there are similar companies, then it's a good indicator that there is potential for the product.
Now the entire focus of the company is on this internal-project-turned-product and now we get to use Close.io to sell Close.io, which is great. We get to solve our own problems (usually) and others benefit from it (and our product gets better).
Wrote more about it here: http://blog.close.io/coworkers-as-your-first-customers-your-...