Wrong but useful. Working @Twitter, tweeting about data science and people management.
"Wait, it's all her views?" 
"Always has been"
You can add location information to your Tweets, such as your city or precise location, from the web and via third-party applications. You always have the option to delete your Tweet location history. Learn more
When product people say it, it often means something like "the magic person who uses Big Data™ to find a simple change I can make to give me a 10x advantage." This is obviously bad; maybe it could happen, but it probably won't. Classic expectations mismatch.
But what's annoying is that data people say it all the time too. For them, it often means something like "I want to make recommendations and have them acted on immediately." There's frequently an implicit assumption that PMs/engineers should seeking out data advice as well.
If this is your expectation, you will not like working in a data role that isn't pure machine learning/optimization. Data people are usually not decision makers, and they are not necessary to product in the same way that developers are.
PMs will not seek out your perspective unless they think it's adding something they don't already have. And no, it's not using Big Data™ to find a simple change that brings a 10x advantage. Making a PM value your perspective is far subtler and happens over a longer time horizon.
It's partly about being on their side, even when you think some ideas are dumb. Answer some seemingly pointless questions, help run experiments you think won't work. Don't go overboard, but be supportive. It builds trust, and it's damn hard to influence people who don't trust you
And as you're building that trust, always ask WHY they care about the things they're asking. Curiosity helps build trust as well, and more importantly, it helps you get ahead of the curve.
I've also heard that being a thought partner means being involved in project ideation, but IMO, the ideation stage is already too late. Ideation is often about the what and the how of building product, but data is uniquely good at figuring out what should be next.
By asking WHY a PM cares about certain data requests, you get a sense over time of their underlying concerns and motives, and THAT is what you need to drill into. Deeply understand those issues, how they connect to the rest of the product, their past and present trajectory...
Then develop a narrative about those things, and let it be a little overly simplistic. If you want someone to act on your recommendations, make them easy to understand. Tell them how they're supposed to interpret your data, even if it doesn't come naturally to you.
And most critically, find a way to directly connect your point about what's next to what they're doing now. It demonstrates you aren't off in la-la-land, that you care about building things WITH them. It builds a bridge so it's easy for them to seek out your perspective.
True partnership (thought partnership or otherwise) takes time, effort, humility and curiosity. Time is the hardest to achieve in the tech world, where everything moves so fast, but the other three are possible on shorter time scales.
Of all of them, humility matters most. Yes, young data scientist, you are smart and people should listen to you. But the people you work with are also smart, and you should listen to them too! Maybe they're wrong, but with an open mind, you may just learn something.
And either way, it's ammo for later. You need to know where people are now if you have any hope of convincing them of where to go next.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.
