What research goes into informing Product what they should be delivering and what is the tangible output of that research?

You'll definitely want to chat with a research agency or your internal researchers to build a game plan, but I think it starts with a list of questions you want answered. Nothing too formal, just conversationally, what insights do you need to know that will help build product strategies? For me, that starts from talking with PMs and Sales people and understanding current context and then looking for where they may be a gap. 

One of our products is localized only for US-EN, but we have a large volume of global users. So research that would show user affinity for that product within other languages might help our PMs to view localization efforts as a way to increase macro feature usage. But that hypothesis can only be created once you have a good lay of the land for your product and your opportunities and --importantly-- if those opportunities map with your product direction. For example, if you identify a key lever that can increase user adoption, but your product team needs to make the product more stable becuase you dont have the uptime that you desire, then you may not be at the stage where you want user growth just yet, because you'll set up users for a bad experience.

As for tangible outputs from your research, I think a good way to think about it is what are features that are exciting to users and how believable is it to users that you could deliver those features. Those that are user delighters that are also things you could deliver, bring those back to your PM team with your own POV on why they matter and have a conversation about their potential impact. 

(For clarity, when you talk about feature believability, think of it this way. Say we're making TVs. Your data shows that users want the brightest panel because they want to watch tv in rooms with a lot of sunlight. That may cost a lot to build because top of the line panels will increase your cost of goods sold. High cost equals high price. So when you get into a premium TV market, youre competing against Samsung or LG. Does your brand have the reputation to win over those customers? Or do customers buy your TVs for other reasons? So its about mapping the wants from users to users faith in your brand. Unless youre trying to break into a new market, then you'll want to overlay brand building campaigns on top of new feature creation, and cant just rely on informing Product to win in this category) 

Research

• Competitor analysis to identify gaps to fill and opportunities to strengthen your unique value proposition 
• Customer research to understand needs and willingness to pay
• Market and analyst research to inform how to strengthen your position in your category
• Win/loss analysis with sales to identify common blockers and hurdle

With regard to the tangible output, see my answer to, "How do you influence the product roadmap if the roadmap has already been laid out by product?".

The first piece of research is typically WITH the product team - aligning on the key things that we don't know, but need to know to make better product or feature decisions.  

Upfront you should lay out that we will try to learn X in order for us to decide whether or not to do Y. 

If you do this upfront work, you will have such a better chance of making an impact with your research. 

From there, it is all about determining the techniques that will allow you to leran the requisite piece of information and/or test necessary hypothesis. I try to start "low cost" whenever possible. This may be competitive research / using a competitors platform to figure out how others solve a problem, doing research as to what industry analysts (eMarketer, etc.) or users are saying about your or competitive tools online (twitter and youtube can be great sources of feedback). Plus don't forget to talk to your sales / customer success team if you have one. If those approaches can't get you the information you need, there are always more time intensive approaches like setting up customer interviews, survey research, and diary studies. 

Tangibly though, I find that the output usually looks something like: 

• What we learned 
• Why it matters: 
• Implications for the product
• The opportunity size of those implications 
• How you got the information: 
• The process by which you learned the information for purposes of driving trust 

There is a vast breadth of research that can inform Product decisions. Being at a larger company affords me the opportunity to take advantage of our highly skilled Consumer Insights team and methodologies.

Working with a 10-year old brand like Words With Friends, immediately signaled to me a necessity for a refresh on consumer research.

We conducted a category segmentation study to understand new potential types of gamers that may be interested in word games as well as their motivations for wanting to play.

Longitudinal studies are valuable in understanding the qualitative types of behaviors and emotions over long periods of time. How does a certain type of boost or feature make our players feel or react over time?

Lifecycle studies allowed us to understand where in the game experience our players were engaging and converting and where the drop-offs were occurring. The tangible output being specific improvements that we could pinpoint in the meta experience.

For shorter, cost-effective research, PMMs can always conduct in-game surveys to glean awareness of a feature/campaign or sentiment around a feature.