It's an interesting time to be in product marketing because I think there will be significant shifts in the next few years in how we think about go-to-market. There's a fair amount being written today about how go-to-marketing motions have evolved from inside sales to inbound marketing to product-led growth and are heading towards more community-led growth. Each phase is additive to the one before it (i.e. companies are not going to stop doing one and move to the next but find more success in combining strategies) but I think a lot of the same skills will persist.
First, PMMs will ALWAYS have to be exceptional communicators. Specifically, they have to be able to simplify the complex and not only write in their own voice, but typically in the voice of their company or sales team. They have to be able to understand a process or scenario that they're often not a part of and come up with ways of influencing it. And they have to be able to tell a story. Secondly, they have to be able to understand the dynamics of their market. This starts with who their customers are and how these people are changing or being challenged. The means by which a PMM influences or relates to their customers has changed and will continue to change but constantly listening to those customers and periodically picking your head up to evaluate whether the dynamics of the market have changed can often help you partner with experts to execute in the right way. As an example, my team has and will invest much more time with our customers telling their story, helping turn them into acvocates and build and develop their own communities. This is different from where we spent our time five years ago but involves many of the same skills.
More towards being data driven and on the strategic side. I love Product Marketing because it is a very "scientific" craft, you don't really have to guess at anything. You get out there, understand your target market, talk to buyers/users/champions/decision makers, understand their pains, and then map those pains to your products solutions - and get that message into the market. Where you get to "experiment" is which of the 15 needs you uncovered is the primary one, but you can leverage your demand generation engine to help test those. So a lot more towards the data side of the house, conducting tests, analyzing the outcomes, making shifts, and going again.