The problem is the title "cofounder". The job description is much too vague to mean anything, and if you put definition around it, you lose the appeal.
Let's look at headhunting for some inspiration. Headhunters hate it when companies don't know what the requirements are for the role because it is impossible to search for. Headhunters also dislike it when the role is very custom and requires multiple domains, because there is a very small candidate pool and low chance of matching. Headhunters love standardized roles because they can find a (large) pool of people that match those roles and skills.
Finding a great "cofounder" is the headhunter equivalent of finding a great "programmer". The first question any headhunter would ask is, what kind of programmer - frontend, backend, UX, systems...?
So let's begin defining "cofounder" into a role that can be matchable.
If you define a founding team as the minimum set of skills needed to make the startup succeed, sales (which includes talking to users and finding PMF) is definitely one of the key skills. Coding is another.
But.. b2b sales are different than b2c sales. And arriving at PMF is very different based on the type of product you are building... you have UX based products, business needs based products, media products, domain knowledge based products... some PMF ideas require years of industry specific domain knowledge to arrive at. So sales cofounder still isn't very matchable because most people can only sell a very narrow set of industries or ideas.
If you did somehow manage to find a person who could sell pretty much anything... you have to realize how valuable they are to other industries as well. Instead of going to high risk startups, they can probably sell hedge fund money level of deals.
Coding founders are often the more "ideal" solo founder because if you can code a full app well, you can essentially build a working startup. Also, there is very few industry alternative that are exponentially more profitable for developers. BUT coding founders often get tempted into the valley of solo complacency. Why build a billion dollar business with others when you can build a million dollar business by yourself?
Finally, the cofounder ecosystem is leaky. Cofounders do not go back into the market - successful founders either run their business or become an investor. Serial cofounders find their niche and do the same play with the same people they know. Why expose yourself to risk if you've found a great team already?
Summary:
- vague role, hard to search for
- custom role, small candidate pool
- lots of competition on sales side for more profitable, less risky alternatives
- lots of competition on coding side for less painful alternatives
- no "growing pool" of experienced candidates - if you're "good" (successful) you're often permanently off the market
The candidate pool that you're left with is a bunch of "unknown potentials" that have resolve. It's in essence a very low hit rate ecosystem.
In a sense, it's better to leave the job description vague too, because as soon as you begin vetting the candidates, you lose the only thing they have going for them - their willingness to face the unknown.