AMA: Apollo GraphQL Principal Product Marketing Manager, Kaitlyn Barnard on Market Research

Here is how I’d structure my 30-60-90 day plan as a PMM at a new company. Depending on the scope of your initial work, these timelines can be condensed.   30 Days Start by listening. Schedule 1:1s with as many folks as you can across different teams to see where they think the opportunities and gaps are. Sales, CS, product, executives, etc.  Identify the assumptions already baked into the current strategy to see if you can validate them (or not) within your research. You’re a fresh set of eyes a ...Read More

The most effective approach is framing research as a revenue opportunity, not a nice-to-have. Leadership moves faster when they can see a direct connection between what you learn and what it means for pipeline, deal size, or new market potential. The question isn't "can we afford to do this research." It's "what does it cost us not to know this." A few things that have worked for me: Get buy-in on the questions before you start. Write a short exec memo outlining what you're trying to learn, why ...Read More

There are a lot of great tools out there that are available for free (or with free trials). You just have to get a bit creative and be willing to experiment with which combo of tools work best for you.  Internal sources Customer interviews and prospect conversations Sales call recordings (ex. Gong) Support tickets and NPS results Win/loss reports External sources Review sites: G2, Gartner Peer Insights,  Capterra, TrustRadius. Reddit, Hacker News, Discord, and Slack communities where your person ...Read More

The fundamentals haven't changed. AI gets a lot of attention right now but the highest-signal research methods are still the ones that have always worked. Talk to people. Customer interviews, prospect conversations, and user research are still your best source of insight. AI can’t replace a real conversation. Listen where your audience already is. Reddit, Slack communities, Discord, LinkedIn groups, and forums are full of unfiltered opinions.  Mine review sites and internal data. G2, Gartner Pee ...Read More

There's 2 ways I like to validate if the information I've uncovered resonates: Talk to real people from your target audience. Whether that's a prospective user, a current customer, or an internal SME with the same role as your target audience, validate that information with them and see what resonates. One thing to keep in mind here though, just because one person says yes or no, doesn't mean it's going to be the same for everyone. Test a few assets to see if it scales. Use the information you'v ...Read More

Across my last few roles focused on API infrastructure products, one of the most impactful research findings has been around developer behavior and how much it has shifted. A lot of developer and engineer roles at organizations don't follow typical titles anymore. A software engineer can own purchasing decisions, influence the roadmap, run evaluations, and champion tools to their team. All without a title that signals any of that. And the buying committee within accounts looks a lot different th ...Read More

Not everything needs to make a huge splash. This is where having a tiered launch system helps manage expectations. Some features aren't going to get the same amount of eyeballs as others. And that's ok! If everything gets a full launch moment, nothing feels important. Here are a few ways I approach positioning non-remarkable features: Lead with impact, not the feature. Even for small updates, find the "so what". Lean into the reason the feature was built in the first place. Sometimes even just a ...Read More

If you want to go deeper on the tactical side of using AI for research, here is how I actually use it day to day: Choose the right tool for the job: I use Claude and ChatGPT regularly and they have different strengths. There is also a growing ecosystem of PMM-specific AI tools out there worth exploring. That said, I still prefer building my own workflows in Claude or ChatGPT. You get more flexibility and control over the output, and it's one less tool to add to your stack. Competitive intelligen ...Read More

Yes! But with a couple of important caveats. AI is great at quickly aggregating and drawing insights from a large amount of data. Here are some of my favorite use cases: Competitive intelligence: For example, crawling your competition's products and positioning across a variety of channels instead of spending hours manually doing that. Trend detection at scale: You can monitor how your category is being talked about across multiple channels and set it up to do so on an ongoing basis and synthesi ...Read More

I'm going to cover some basic ways I look at this because this is what most companies will have in place to be able to track: Views / traffic: Is it getting eyeballs? If something hits, you're going to see the results in the numbers pretty easily here. Conversions (signups, pipeline, etc.): Is that traffic turning into anything meaningful? You may start getting views, but are those users doing what you want them to do? Activations: Are those leads that are signing up or turning into opportunitie ...Read More

This is different at every company I've been at! At startups, it's typically up to you and a bit more scrappy. Trying to get as much information as you can with very minimal resources (i.e. Google is your best friend). At bigger companies, I tend to see they have a lot more resources, whether that is dedicated data or research teams, or just budget to invest in better research tools or outside research firms. I don't have one favorite framework. I tend to find that the best choice is: The one th ...Read More

In another question, I’ve included what my 30-60-90 day plan would look like if you want to check that out. Here I’m going to expand on my priorities to get myself ramped up to be effective.  The goal in the first 90 days is to build a foundation that makes everything downstream easier (messaging, launches, roadmap conversations). My top priorities early on: Understand the buyer journey: Who is actually buying, who is influencing the decision, and where do deals stall? This shapes everything els ...Read More

The first step is getting clear on what decision the research needs to inform. Unfocused research takes forever. Know exactly what you're trying to answer before you start collecting anything. A few ways to work efficiently:  Start with what you already have. Internal data, existing research, sales call recordings, review sites. You have a lot of existing insight right in front of you before even needing to go external.  Use AI to compress synthesis time. Summarizing transcripts, tagging themes, ...Read More

You can get a lot done with very little if you're strategic about where you spend your time. Here are a few of my go-tos: Start with what's already out there: Google, G2, Reddit, and industry forums are free and full of unfiltered customer language and competitive insights. Don't underestimate how much signal is sitting in public review sites alone. Talk to people: It costs nothing to schedule a 30-minute conversation with customers, internal SMEs, former colleagues in the persona you're targeti ...Read More

Not having customers doesn't mean not having access to your target persona. You just have to be more creative about where you find them: Tap your network: First and second degree LinkedIn connections who match your ICP are a great starting point. Oftentimes, people love to talk about what they’re working on and the challenges they run into. Find them where they already are: Reddit, Slack communities, Discord servers, and LinkedIn groups. Your target persona is already hanging out somewhere. Go t ...Read More

Being the first PMM is a huge opportunity, but it can feel overwhelming when there's no existing foundation. The most important thing is to not try to build everything at once. Pick one or two high-priority research questions tied to a real business decision and start there. I included a 30-60-90 day plan in another response, but it’s also important to build credibility up front to be successful as the only PMM.  A few things that help: Become the person who connects the dots. Sales knows why de ...Read More

This can be so tricky because everyone consumes information in different ways so it really differs depending on the company and the teams. My #1 piece of advice is to try to leverage the information sharing processes your organization already has in place before introducing something new. If you can piggyback on processes people are already used to, it'll expedite this and improve your chances that people are actually reading the research. Here are the ways I've seen work at different organizati ...Read More