Leaner Times Now: Why Product Ops Is More Important Than Ever

In today’s fast-paced, resource-constrained environment, product teams are under pressure to do more with less.

How do you maintain alignment, drive efficiency, and achieve strategic impact?

Product Operations is the answer—the strategic enabler that helps your product org thrive in leaner times by ensuring your team works smarter, not harder.

Denise Tilles, co-author of ‘Product Operations’ and seasoned Product Leader with 15+ years of experience, joined us to share why Product Operations is more important than ever and provide actionable strategies to help you demonstrate your team’s value to the executive team, especially when resources are limited.

Watch the session and explore how Product Ops can help your organization achieve more with less while staying aligned, efficient, and focused on high-impact outcomes.

Key Takeaways

Product Operations enables rigor, focus, and propels the innovation flywheel.

Core imperatives in any company…regardless of challenges.

– Denise Tilles

Transcript

The following transcript has been altered for readability.

Welcome and Introduction

Ana Andrade: Hello, everyone! Welcome to the Product Ops HQ Virtual Meetup! This is the April edition. I’m Ana, your Community Manager here at Product Ops HQ, tuning in from Portugal. It’s great to have you all today. We have an amazing session lined up for today, so I’m pretty excited.

Let’s give it a minute or two for more folks to join us. Meantime, say hi in the chat and let us know where you are tuning in from. Joining me today is Leah from Dragonboat. She’ll help with your questions and share some useful resources.

I’m seeing a lot of familiar faces here and some new names. For those new here, a big welcome. Product Ops HQ is your go-to place to connect, share experiences, and grow together. If you haven’t joined our Slack yet, please do. We just crossed 1.5K members in the Slack community. Thanks to everyone for making it such an amazing space. Stay tuned there because we have some news coming up.

Housekeeping Notes

Ana Andrade: Today’s topic is about why product operations is more important than ever. We will kick off with an insightful talk from our special guest, followed by a live Q&A. Please add questions in the Q&A box as we go along, or raise your hand during the live Q&A to speak directly with our guest speaker. This session is being recorded, and we’ll share the recording afterward.

A Word From Our Sponsor, Dragonboat

Ana Andrade: Quick shout-out to our sponsor Dragonboat, an AI-powered strategic portfolio management platform designed for everyone in a product operating model—from CPOs to product managers, product operations, and teams. Dragonboat helps set clear strategies, build portfolio roadmaps, and deliver products that delight customers and accelerate business outcomes. Dragonboat is more than a platform; it’s your partner in achieving product operating excellence and also supports our community. If you want to learn more about Dragonboat, check out dragonboat.io.

Guest Introduction

Ana Andrade: I’m excited to introduce our guest speaker, Denise Tilles. Denise co-authored the fantastic book “Product Operations” with Melissa Perri, author of “Escaping the Build Trap.” With over 15 years of product leadership experience, Denise supports companies like Insight Partners, Red Hat, Bloomberg, and Novo Nordisk by strengthening product operations, product management maturity, and product operating models. Denise, thank you for joining us today and supporting our growing community. The stage is all yours.

Denise Tilles: Leaner Times and Product Ops

Denise Tilles: Thank you, Ana. Hey, everybody! It’s exciting to see our community out in large numbers—this is exciting. I’m dialing in from Brooklyn, New York. We have a beautiful spring day, though it was winter two days ago, so I think the weather’s still deciding if we’re actually springing forward or not.

I’m super excited to talk to you about Leaner Times today. Last year, at the PLA Product Ops Summit, I discussed the challenges following Spotify’s massive layoffs and the role of those doing the “work around the work.” I noted that meant us—product ops professionals. This year, I’m coming to you guys with an update: things are looking better, yet different opportunities and challenges exist. Today, I’ll reinforce why product ops remains vital in any economic climate—positive, negative, challenged, or strong. I’ll share some case studies, proof points, and hopefully open this up for a lively conversation.

Denise Tilles: About Me

Denise Tilles: A bit about myself—I have 15 years of product leadership experience on both operational and consulting sides. Yes, I co-wrote “Product Operations” with Melissa Perri. I advise, consult, and coach CPOs and VPs of product around product operations and product management leadership. My work involves advising teams, enhancing product management maturity, skills development, team assessment, and establishing effective product operating models.

I think Graham said this last week on LinkedIn that there’s not one product operating model, and I totally agree. It really depends on the context and the company, and what value your particular product ops team and product management team are working towards.

I’ve been fortunate to work with Bloomberg, Nova Nordisk, Takeda, DaVita, and other companies, especially in life sciences, and I actively teach as well. Evangelizing the importance of product ops is a significant part of my role—something many of us do daily.

The book, I think I mentioned this at our February meetup. Melissa and I are finally recording the audiobook version of our book, “Product Operations.” Many have requested this, and we expect it to release mid-year, updated to reflect AI’s impact on product operations. Afterward, we plan to publish a second edition, incorporating more community case studies. I encourage you to contribute your experiences.

The Value of Product Operations

Denise Tilles: Today, what I want to discuss—and we’re all believers here—is how product ops enables focus, rigor, and the innovation flywheel. These core imperatives remain critical in any challenging environment and within any company. I’ll share two actionable strategies to help you effectively showcase your value.

A significant part of our role in product ops involves enabling faster and better-quality decision-making for product managers, ultimately leading to improved business and customer outcomes. However, we must constantly demonstrate our value since we’re often seen as a cost center. How are we providing value? I want to share with you a couple of ways to think about that.

So our new type of Lean.I want to share some interesting numbers. Yesterday, I gathered layoff figures, but this morning I checked trueup.io for the latest updates. These numbers represent tech employees impacted by layoffs, primarily US-based but indicative of global trends. In 2023, we saw a massive spike—429,000 tech layoffs across the board, not just in product ops.

Last year at the Product Operations Summit, it was March, and we wondered what layoff trends would look like moving forward. Thankfully, layoffs dropped significantly—by about 40%. For 2025, numbers appear nearly flat, but interestingly, projections jumped 42% from yesterday to today. The world’s geopolitical and economic conditions are rapidly shifting, and such volatility surprised me. So I’m like, “What happened? Was it Tesla?” I don’t know. I’ll be curious to see what happens tomorrow. 

Poll: Has the Shopify Memo Impacted Your Organization?

Denise Tilles: Leaner times are here, but there are other challenges and opportunities as well. What do I mean by that? Consider the Shopify memo, which replaced Spotify’s memo and set new industry expectations. Shopify’s CEO emphasized that any non-use of AI must now be justified. This new norm calls for incorporating AI into product development phases, such as prototyping, to accelerate learning and collaboration. Additionally, there’s now accountability tied directly to AI adoption.

Informally, I’m hearing from clients and colleagues that many companies embrace this as their new standard. This shift carries significant implications for product operations. I’d love to hear from you: Has the Shopify memo impacted your organization? Let’s do a quick poll.

Denise Tilles: Let’s give it a minute here. “Business As Usual” (BAU) responses are declining. Okay, let’s share the results. So interesting. We’ve got “execs are talking about it, but nothing tangible yet”. “Yes, we have to justify any new hires why AI cannot do the work.” I’m surprised it’s that low. Okay. And then “BAU (Business As Usual)”. Interesting. 

Would anyone like to share insights? Kaya had her hand raised first. Ana, can we take her comments?

Ana Andrade: Okay, we have a lot of people raising their hands, so I’m going to unmute you, and you can share your insights.

Tori Oellers: Hi there. My name is Tori. I’m VP of Product Ops at a company called Keyfactor. It’s in the cybersecurity space. And for us, it’s certainly something that we’re hearing as a direction of investments that we’re making, or work that we’re doing to leverage AI, but there’s still a dialogue of what we want that to look like. So teams are giving latitude to advocate and request tools and access.

One thing we have done internally is it created a roundtable where we meet regularly to share tools that we’re testing or trying out and how we’re adopting those and how we’re utilizing those. And that’s kind of being used as sort of the steering committee for the company’s AI practices. It’s that kind of collective team.

Denise Tilles: Interesting. Has your team explicitly referenced the Shopify memo?

Tori Oellers: I have not heard it referenced internally at our company. I think in general, the sentiment of the market and the teams is that that is certainly the direction. I saw you had worked with Insight Partners. They’re one of our investors as well, as well as Sixtry. And those two are also giving that direction to make more of a motion and a move towards adopting AI in the work, not just in product operations across the whole organization.

Denise Tilles: Right, that makes sense. Thanks for sharing, Tori. We’ll revisit this at the end if we have time. I anticipate the “Business As Usual” responses will decrease over the year, but thank you all for your valuable insights.

Denise Tilles: So as we think about—well, we’re at BAU now, and things will probably keep slowing. We’ll be thinking about what it looks like to use AI. Will we be measured on that? How do we guide change in a shifting scene? We have the backdrop of layoffs—maybe not as bad as late 2023, but still there—plus AI.

A lot of people here see AI as a sort of force multiplier. How do we use it, while leading with the story of human skill and AI support? How do we tell that story? We must differentiate ourselves. In any market—bull or bear—companies always need rigor and focus on understanding product adoption, listening to customers, tracking resources, and where to place bets in terms of R&D allocation. We need champions of change thinking about what new ideas look like, how to set up systems so people stop wondering how to do work and just do it. Product operations have always offered that key value proposition.

Below are two quick examples from the book, updated for today’s age of generative AI.

Case Study 1: Athena Health and Capitalization

Denise Tilles: One of the case studies that really resonated with readers was Athena Health. Athena Health wanted to speed up their capitalization process. How they thought about how they thought doing this more efficiently. Capitalization, how we think about it, is a fixed asset on a company’s financial statement. You want to be able to amortize those expenses over time to increase EBITDA. Athena Health really started thinking about framing the work that was being done in a more automated way. 

Initially, they had been doing very long spreadsheets with thousands of rows of people kind of guesstimating what was done as a sort of in-progress versus delivered. They started to leverage understanding the problem first from the product team, but then started to leverage how can we put this into Jira where we can pull these reports in a more organized way. So thinking about framing it from capitalization point of view, what’s a new feature or foundational tech versus an expense of something that had to be fixed right away, a bug to make sure that they were maintaining that trust reactively, and then the experimental development investments. 

So they wanted to make this more accurate and less painful. Teams were taking like days to get this done every quarter. And this is like a really bad Groundhog Day. How do we improve this? They ended up making this more automated, at least that first cut. And they saved hundreds of hours in terms of capitalization efforts. And this was pre-AI. AI was a gleam in everybody’s eyes, right? 

How would this look today in capitalization, supercharging with AI? Doesn’t replace it, can’t do it completely. You might leverage AI to streamline capitalization around those digital footprints, like we said, with Jira, GitHub, etc. The auto calculation of time spent on capitalization, and then tracking those project starts, task completions, and code deployments more accurately. 

But we still need human expertise. How are we making that very fuzzy distinguishing moment of the technologically feasible versus available for sale, those different phases, and how those can be capitalized or not? And then contextualizing all of the allocation decisions for auditors. If you’re a GenAI company, how do you think about those unique accounting complexities? And then you still need the human judgment for AI training costs, upgrading versus maintaining. 

Capitalization is still a core need, but how do we partner with AI to think about ways to make it more efficient, more cost-effective, and hopefully faster and less painful or less painful. So I’m seeing clients starting to use AI this way now. I imagine we’re going to keep seeing an evolution of that, but this is one example.

Case Study 2: Fidelity and User Research

Denise Tilles: Another one was a case study from the book with Fidelity and thinking about research, user research, and how do we make sure we’re really building the right thing. And Fidelity has a smallish 28, as a lot of people use a research team, but they wanted to be able to support the hundreds I think close to a thousand PMs in making sure that they could get qualitative inputs that were reusable for all of the teams, but they had this small team to be able to enable the scale. 

They realized they had to create playbooks to guide those common needs and help teach the product managers to fish, and create an internal research certification. Not a huge certification fan myself, but I thought this made sense, and they made it kind of a fun thing to attain within the company. And they would create something like a rank user needs document where product managers could sort of work with clients which were hard to get a regulated industry for Fidelity. And so when they had time with their clients to really be able to make the most of it and thinking about more structured and templated ways of doing research with their clients.

Product Ops was just helping operationalize the inputs faster and better quality decision making. The success metrics, how do they get those insights in? And then what are those insights to action? What’s that rate look like? And then how do we tie those actions towards business and customer outcomes? Still very important. 

How does that look now? And sort of thinking about the research democratization right, so we’re probably a lot of us doing this now that generating at least the first cut of interview questions, right, based on the research goals. The pattern analysis within the feedback data. Taking a first crack at the research templates and playbooks and then starting to hone and adjust. And guidance on the research methods as you sort of using that as a back-and-forth as a refining mechanism.

But we still need the human expertise, right? How are we thinking about, you know, is this research design valid? Claude may tell you that, but you need people with that expertise to understand those sorts of edge cases to really understand, “Is this the right way to be addressing and are we getting good information out of this?” The nuanced user behaviors and emotions, the nonverbal things, right?

The strategic decisions based on findings. Certainly, generative AI can give us a sense of that but still requires human judgment towards that. And then teaching the critical thinking skills to PMs. Wouldn’t that be amazing if Gen AI could do that, but it doesn’t. So that’s still on our teams to help with those more humanistic needs.

This is an example that you might want to share with your company and think about where are the areas we could delegate the draft work and then where can we make sure that we’re leveraging product ops in a more strategic way, to make sure that we’re really getting higher value efforts out of them? 

So that would be the strategic alignment, the cross-functional. And that’s another topic I want to talk about some other time. Is product ops morphing into Ops-Ops, R&D-Ops? I’m seeing a lot of buzz around this right now. Cross-functional ops, scaling strategy tools and systems. And how do we think about the lower value? Still important, we need to do it, the meetings and the admin, the recordings, the documentation, at least taking a first cut the initial analysis and research.

This has been an effective way to explain to teams and explain to executive teams that here’s where the sort of differentiation happens. One doesn’t replace the other. It’s really humans driving AI with the leverage.

What Can Product Ops Do In The Age of AI?

Denise Tilles: So what can we do? I know many of us are already doing this, but I want to call it out and frame it in a way you can share with your teams. Be intentional about highlighting what we do ourselves versus where we use Gen AI. Don’t blur the lines—show where human skill ended and AI support began. Start with the outcome: where did we apply AI, and where did people make the key calls? Tell the story of that partnership and how each part added value.

Sometimes I hear from people I coach: “I don’t want to say AI helped, so it won’t look like we’re replaceable.” But we need to open the curtains. Be very clear about using AI for lower-value tasks so we can focus on more strategic work. So be really intentional about telling that story. 

Sharing out how we unlock value. Sharing out again. Product ops is already hard to explain, so it’s even more important now. So in terms of efficiency, the customer focus, the innovation, cost management, right? And faster and better quality decision-making, but all enabled by the AI-Human Expertise Partnership.

And being the force multiplier, we talk about that in the book. So what’s essential, we’re all doing this, is keeping the pulse on product manager needs: sentiment, productivity, value and impact, and what the collaborative experience looks like. Help teams use AI in the right spots, not just for the sake of AI. Set simple rules: when to try it, when to skip it, and which tasks make sense.

Thinking about AI as augmentation, not replacing us. Again, from my first point, being really clear about how we’re leveraging it and how it’s helping power better outcomes with more time to focus on the strategic stuff. 

And leveraging this community. It was great that we had all of the folks in February sharing out how they’ve been able to leverage AI and tips, and use cases. I’ve talked to a few of you already, and I hope to hear from more of you, but this is an amazing community. Make sure you’re leveraging it to understand what’s working, where you run into challenges, and different tools, different applications. I think that’s the best thing we can do.

Denise Tilles: So hopefully I’ve left you with a key takeaway. We know this product ops enables all of these core essentials, the rigor, the focus, the innovation flywheel, and every company needs this, whether we’re laying off, whether we’re increasing headcount, whatever the challenges are, it’s key, it’s core. I will leave you guys with a question to hopefully start the conversation. How do you show value?

Thank you. Okay, so we can start with that, Ana, or we can start wherever you’d like.

Live Q&A

Ana Andrade: Thank you, Denise. This was super helpful! So now it’s your turn, everyone. Please drop your questions in the Q&A or raise your hands if you’d like to ask your question live to Denise. So I will kick off with a question. You talked about focusing Product Ops on strategic work. What would you say are the top three things that a product organization should do to drive strategic impact?

Denise Tilles: Yeah, I mean, I think that’s sort of an evergreen type of challenge and opportunity, right? So you want to make sure that the teams are really focused on developing an outcome-driven practice. It’s so easy to get into that sort of drip-drip feature-delivery-output. So really keeping it high level, making sure people are focusing on the outcomes. The visibility, right? Where are we going and why are we doing this and backing it up with that qualitative and quantitative data why we’re going there.

And then the success, right, the strategic impact shouldn’t just account for feedback, but the strategies from feedback to roadmap delivery to go to market making sure that we’re understanding where we’re going and then how we’re measuring that and are we going back (hopefully) to understand that progress. 

So I think the three things would be the outcome-driven practice, more visibility on what you’re working on, and then accounting for the feedback, understanding as you’re trying to sort of deliver on your strategy and the go-to-market, and getting that market feedback as well and measuring the impacts.

Ana Andrade: Okay. Thanks. That’s super helpful. I think we have a lot of people here. Maybe I will start with those who want to share their questions directly live, so I will allow Anamika to join us live.

Anamika Singh: Hi, first of all, great to meet both of you, Ana and Denise. Thank you for, you know, kind of organizing this and, you know, walking us through all of those insights. That was really helpful. Especially for me, because this is my first product ops conference meeting ever.

Denise Tilles: Oh, good. Welcome.

Anamika Singh: Yeah, so I just had a question to you. You obviously spoke about AI and how it’s really important to kind of integrate it in a lot of, I guess, the manual work that happens so that you can focus more on the strategic and the other decisions sort of that you have to make within product ops. So I’m just wondering to What extent should we you know sort of like rely or depend on the outcomes generated by AI, Especially when you are looking for something that requires a high level of accuracy, right?

Like let’s say you are trying to do some sort of data analysis for you know to help a product manager or to drive a data-driven decision. So obviously you need the analysis to be really accurate so if So that’s what I mean that okay you know of course I want to use AI to make the work like faster, compile data quicker, but how much do we rely or depend on it, in that sense.

Denise Tilles: Well, I think it’s kind of the common sense that so many folks are talking about, we discussed in February. And I’m seeing with a lot of product and product ops leaders that it is a great first step. It doesn’t replace that critical thinking. So it’s, I don’t want to say it’s a vomit draft, but it’s a really good first draft, but it really has, it’s so easy to like, “okay, good.”, and move it over to a deck or whatever and be done with it. You really have to analyze it and put it through that critical lens of your context and what makes sense at your company. So in my mind, it’s a really great first draft that needs to be sort of groomed and managed and brought into perspective by an actual human, if that answers your question.

Anamika Singh: Yeah, that is. Thank you so much.

Ana Andrade: Awesome. Thank you. Okay, Chris. Hi, Chris. Welcome.

Chris Compston: Oh, hey, Denise. Thank you so much for sharing the presentation today. You had a question there at the end, which was, “How do you showcase value?” So the way that I’ve been thinking about this over the past year or so is… A lot of the times when product ops Leaders get asked the question of who is your customer? Who is your user? Who are you doing the work for? The answer is often the product organization. Maybe it’s a product management leadership position. Maybe the product managers, the product teams. And I’ve started to, I guess, subscribe less to that concept and more to the concept that the end customer is also our customer. So if we’re trying to maximize impact and performance, increase quality of decision making in the teams, the purpose of that is the end customer finding value for them, therefore driving business impact. 

And so that’s where I come from. That’s the position I come from. So when I’m talking to potential clients who were looking to hire me for my services and we want to talk about how do we measure the impact of product operations. It always has to be linked to something that is impacting the business through the finding of customer value, solving customer problems and any way we can do to help solve that. 

I had a really great conversation with Hugo Fores, if anyone knows him, and Kiara Gardner about us as well. I do believe in some cases that we have to be quite pragmatic and sometimes it does have to get stuff done. There’s a bunch of things that just have to happen on the more tactical level. But when that initial conversation is happening about how should we showcase value, it should be on that endpoint, the more strategic angle of how are we helping the organization to drive customer value and business impact?

Denise Tilles: Right. Yeah. So I think what you’re saying is not in terms of more story points delivered or anything tactical, but trying to bring it to a higher strategic point of view. Yeah, it’s hard to do it, though, right?

Chris Compston: Yeah, I mean, even like Yeah. I mean, even like operational metrics, you know, we reduce lead time for X or can we, you know, reduce the amount of time that things go through QA? All of these have potentially negative outcomes anyway. But measuring those things is useful and it’s eye-opening. It’s good to hold up a mirror to the teams to understand themselves how they might. But the team, the product talks as a function is never going to be that successful if all they can do is reduce lead time or introduce a new process that makes something slightly better if it’s not connected to that end value.

Denise Tilles: So whatever direct or indirect thread that you can put through the needle towards that bigger picture, outcome, do that is what you’re saying?

Chris Compston: Yeah, absolutely. And then, you know, the example I have here is I was working with a client on a product intake process. We could have just introduced a process and taken a step back and gone, okay, things are better now. Yeah, process created. That box has been ticked. But actually, when you think about measures in terms of, yeah, someone said here, outcomes versus outputs, Danielle. We actually were able to identify that because that new process existed, where intakes may have taken two years to see the light of day, they were now taking like a quarter. Which is still quite a long time, but when you’re gathering things from the rest of across the business and with business clients in an enterprise setting, SaaS enterprise if you can identify, you’ve introduced a process and the teams are able to build something for those businesses to get to their customers. And now they’re increasing their revenue, then product operations is going to look quite impactful and therefore incredibly valuable.

Denise Tilles: Love that. Thank you.

Ana Andrade: Thank you, Chris. Okay, I’m going to pass the mic to Peter. Peter, welcome.

Peter Boersma: Thank you. And thank you, Denise, so far. My question is, how much of AI’s support of product ops do you think will happen inside existing tools like Dragonboat, Asana, Jira versus developed outside of those tools and potentially in-house?

Denise Tilles: That’s such a good question. And I think I see it sort of growing simultaneously because they’re going to be addressing different challenges and opportunities and use cases, right? You know, products like Dragonboat or Asana or any of those tools that help us as product ops folks, I think we’ll definitely be seeing that as a value prop. I think we’re seeing more actual integration and use of that as opposed to say a year ago when people were saying things were AI-enabled. But I see just as much of a need internally because you have those contextual challenges and or types of data that can’t necessarily be running through a public type of a public excuse me, tool or, you know, GPT. 

Couple of the case studies I’ve been hearing about so far for the update of the book is something that assesses over the transom requests from customer support or sales for new features and kind of does a quick assessment what that value is before product managers even sort of pick that out to sort of look at that. That I imagine will be using a lot of sensitive information in terms of the customers who requested their lifetime value or the amount of their annual spend. I see it in sort of equal measure. I guess the question to you is. Of the tools that you use Which do you feel like are the most, which one would you say if you had to call one out, is the most useful in terms of integrating AI in a measurable, not measurable, but a real impactful way, that you’ve seen or have you seen any yet?

Peter Boersma: Good question for me, the tool I use most at the moment is a ChatGPT detector. So it’s not supporting me. It’s me trying to burn down the work of others, but that’s me.

Denise Tilles: Hmm. Okay. But in terms of the tools that you mentioned, like Asana and Dragonboat and Jira, are you seeing any of those doing it particularly well?

Peter Boersma: As a former employee of Miro, I see Miro doing good stuff in terms of AI, analyzing work, analyzing documents, preparing that for the next step in the product development lifecycle.

Denise Tilles: Cool. All right. Thank you.

Ana Andrade: Thanks, Peter. Okay, we have one last raised hand, and then we are going to move to the Q&A. So Vered, Welcome!

Vered Yeret: Hi. Hi, Denise. We met last year at the conference. I’m one of the agile coaches who were let go from Spotify, doing work around the work. Yeah, nice to see you again. So I have a question whether you see an opportunity with allowing or maybe even leveraging product operations to support the transformation to use AI either in the product organization, or maybe even beyond the product organization. You know, to kind of take the product operations skilled in driving change and leverage it to help, okay, how do we use AI? What’s blocking us? What can we do to improve? Just another thought that I had lately.

Denise Tilles: So your question is, can product ops sort of be that model to how not just product teams, but the rest of the company sort of leverage AI?

Vered Yeret: I would say lead the transformation of either the product or the whole organization to use AI in a better way.

Denise Tilles: Yeah, that’s a great point. And I think product ops could be uniquely situated for that or working in concert with what we call in the book the super friends of other ops teams, because being able to showcase specific uses, I think will help inspire other teams. So if it’s part of an all-hands, a quarterly all-hands. That might be part of it of just like here’s you know five minutes of a case study of how product ops or product ops may find another person within the company of how AI was leveraged. And then that actionable takeaway. I loved being able to include that so what? So if it was something that was super specific to say engineering What’s the bigger picture that everybody could take away from that? So I think product ops could be really useful in terms of creating that framework of showcasing different ways of leveraging it, but also bringing it back to the “so what” and how can I and marketing do that or what would that mean to me in customer support so Yeah, I think that’s a great point.

Vered Yeret: And do you think there is an appetite for allowing product operations to lead those? That’s like always been the struggle to kind of, you know, sell that, “hey, we can lead this. We have the experience. We’ve done similar things let’s just use us or leverage us in a meaningful way.”

Denise Tilles: Yeah. Don’t ask permission, beg forgiveness. Step into the breach and do it. Totally think so. Yeah.

Vered Yeret: Yeah, you’re right. You’re right. Thank you.

Ana Andrade: Thanks. Okay, let’s move to these written questions. We have a lot of them. I will try to resume it. But the first one is regarding capitalization. Our engineers manually log their time against Jira epics using a Jira add-on. How can we use AI to support this, as the time spent data measured may not be accurate, using time spent in status, etc.

Denise Tilles: How can we use AI to support this as the time spent data measured may not be accurate. I guess I would open that up to the community. I don’t have a quick answer on that. Anybody? Okay. Okay, we have Kate. Kate, go for it.

Kate Mortenson: Hey, excuse me, this was actually related to a question that I posed in the chat. Because it wasn’t directly for you, Denise, but I kind of wanted to pose it to the community. My product operations team is pretty small. It’s really just myself and I’m trying to figure out how I can use an agentic AI to do that kind of fact-finding for me. And there’s a lot of different ways that you can, from what I’m researching, it sounds like there’s a lot of different tools that are available in a lot of different ways that can be constructed. But I would recommend that you look into something called an MCP or a “model context protocol”, which essentially is you turn your AI agents into data chauffeurs who will go to where the data lives, grab it, and then bring it back to you. And so I think when we talk about using AI and product operations, what we maybe should start considering is that we can use AI to outsource that sort of stuff like that the repeatable fact finding efforts that we have to do to reduce that time.

Ana Andrade: Thanks, Kate. Okay, let’s see what else. Uh… there are a lot of questions around tools. So, are there any tools or resources for AI that you find emerging or trending that are beneficial for teams to leverage? And also Jeff also asks about product development tools like Dragonboat that are using AI in leading-edge ways.

Denise Tilles: So the question, let’s see here, trying to find the question.

Ana Andrade: I kind of summarized one from Tori. Any tools or resources for AI that you find emerging and trending? And Jeff’s because they are both around AI tools.

Denise Tilles: Yeah. I tend to go to the community and find out what people are using because it really does differ based on what that need is. We just had a question around how to get more accurate reporting information in terms of capitalization with the engineering aspect and hearing about MCP. That was certainly helpful because I have heard that before. So I guess I would say in terms of the context would matter a lot. In terms of what you’re trying to solve. So I couldn’t say like, you know, go to Gemini. This will solve everything or MCP. I think it really is contextual. 

So as you come up with these questions, my best recommendation is to bring it to Product Ops HQ, and find out what people are doing today. Otherwise, we could be here for probably three hours, considering the different potentials.

Ana Andrade: Definitely. And thanks for the shoutout.

Denise Tilles: Do you want to address the Dragonboat, Ana, or any aspect of that?

Ana Andrade: Well, I can tell you guys to check out Dragonboat, they are built-in on AI. So yeah, definitely check it out. 

And I’m going to ask Jeff to unmute himself. Hey Jeff, welcome.

Jeff Duncan: Hey, thank you. Yeah, it’s great. Great to hear from you everyone, again. I think just a little bit more context on ours. So we’re a very small, very small product ops team. There’s basically two of us. And so we’re obviously looking for ways to force multiply this work across the team. And so again, one of those main things is tools and things like that. Yeah, one of the things that we’re running into is that there’s a huge push for AI just not because of any kind of Shopify memo, but really just we have a whole team dedicated to driving AI. And so one of the things that we’re constantly being asked to do is try to find new and innovative ways to leverage AI in our in our space on a regular basis. I think more of my question was around like, Are there any tools that are doing that? Because I think… Yeah, we’re we’re we’re we’re if we’re not asked already, we will be asked to leverage and start justifying, you know, again, all of our tool spend and all of those things. And so I’m just wondering if anybody has, and I can connect with the group as well. That’s kind of the context of my question.

Denise Tilles: Sounds like a group question. Do you guys want to share out in the chat? How you respond to that? So Jeff, to summarize your question is. What are the tools and how do you justify them.

Jeff Duncan: Yeah, because I think, you know, a lot of, I mean, right now we use, we use chat GPT and we’re building, you know, custom GPTs all the time. We made a tool that kind of will create PRDs, you know, as an automated tool that they can actually put in all this information and it kind of spits out like so like, documentation creation, I think we’ve kind of hit the stride there but when it comes to development and, you know, again, I think measurement is probably the biggest thing, right? Measuring our impact, our ability to manage KPIs, et cetera. That’s kind of where we’re falling short. So I think those are some of the kind of over the horizon type things that we’re looking at.

Denise Tilles: Mm-hmm. Anybody got a suggestion?

Ana Andrade: We have a couple. Kate, are you got an idea?

Kate Mortenson: Hey, sorry. I feel like I’m monopolizing this here but Jeff, that’s actually something that I’m I’m trying to work on as well. My organization recently revised some of our North Star metrics. And we had a whole session where we presented them to the company and we started talking about like, not necessarily this is how we’re measuring or this is how we’re tracking, but this is why we track it and this is why it’s valuable.

And I wonder if there might be an opportunity for you to establish like North Star metrics for that work that you’re doing, like a North Star metric for operationalizing documentation. And then if you have any kind of like scheduled company-wide meeting where you’re discussing where you’re at or what you’re doing, that would be the time to unveil it and say, this is the work we’ve been doing and here’s why it’s valuable. Yeah. I don’t know if that makes any sense.

Denise Tilles: So the why versus the how. Yeah, yeah, that makes sense.

Ana Andrade: Thanks, Kate. Okay, so we have… a bunch of them. I think we will need to answer some of them in Slack after. But maybe we still have time for one or two more. So Marta is asking, can you share best practices on how to leverage the use of AI but keeping it safe in terms of the company’s sensitive information? Which we already covered a little bit but if you can I’ll call it a little bit more.

Denise Tilles: Yeah, especially as you think about life sciences companies and pharma and all that, you know, PII information, that’s really sensitive, right? How do you make the decision of what goes in, what you may run through some sort of generative or agentic AI tool. I typically see companies building that internally, but even still, there’s always a slight concern, I think, from the executive team and even the board of like, “How are we protecting that? And if we have to get audited, what does that look like?”

I think that a couple of principles would be probably building it yourself or at least leveraging some of the things and understanding what those sort of guardrails are, and establishing those and not necessarily trusting that if you’re using ChatGPT and paying for the paid model, that your data is safe. I’ve always sort of questioned that. So thinking about it that way and understanding if it really is worth putting some of the more sensitive data through AI to leverage it. Does it make sense there or could it be better used elsewhere? But I’m curious from the group what’s a key principle for you in terms of sensitive data?

I don’t have all the answers.

Ana Andrade: Anyone want to jump in? I guess not. Okay, one more question before we wrap up. And I’m sorry that we didn’t have time to cover all the questions.

Denise Tilles: Yeah, we’ve got a lot here. This is good.

Ana Andrade: Yeah, we can make some kind of a round of questions later in the Slack. So last question. Our product ops organization is trying to move from being reactive to PM development leadership needs to being more proactive. Feeling a bit lost on how to prioritize and defend prioritization. What is a key framework you might use to help guide decisions for what product ops work on near term, long term, or not at all. And what are some key criteria?

Denise Tilles: I think it’s about… being really intentional about what you’re working on sort of above the line and then what’s below the line that we will not get to. So I’m guessing you probably don’t have a huge team. Most of us don’t. And focusing in on basically, we’ve done discovery work. This is what the product team and leadership has told us are the gaps and opportunities. And we are going to leverage our resources here.

Down here. Sounds like, you know, here are all the things that we’ve also heard in a less order of importance, we are not going to be doing these things. And as time goes on, if these are fine, you know, seeming to be more important areas to cover, then you start talking about what that would look like and possibly do we reallocate somebody? Is that a new headcount but based on what we’ve heard number of folks we have and the opportunities at hand, this is sort of the the box of capabilities we can offer today. So it’s hard to say no when you’re pushing back, but eventually folks do end up understanding, you know, just kind of like, here’s what we work on. Here’s what we are not doing right now. We will reassess every quarter and understand that.

I think the faster you’re able to do that and get back to people rather than, you know, sort of gnashing your teeth and trying to figure out, could we do this? Does this make sense? If you keep trying to chase all those things, you’re getting nothing done. So the more focused you are on that specific set of capabilities and delivering and sharing that back proactively versus reactively, you start demonstrating the value, you start sharing the quick wins, you start talking about the vision of where we’re going.

And it takes time. I wish I could say do this and it becomes non-reactive to proactive, but it takes time, and the quick wins of sharing back what you were able to accomplish. 

But you also start to understand the trends of the kinds of things you’re going to get asked in the seasonality. So anticipating that. So if QBRs are coming up, get ahead of it. Don’t wait for someone to say, “well, what are we doing? What do we need? What are the templates? I haven’t gotten anything. We’ve got one week”. Put that in your calendar and figure out like how far ahead do we need to get it at? So if you start showing those types of things that you’re leading and driving towards outcomes versus reactive, oh yeah, we forgot about that. You will start building up, I think, more credibility and political capital, so to speak. So that would be one of my suggestions.

Helpful Resources

Ana Andrade: I love that. Thanks, Denise. So I think in the interest of time, I will share my screen with some resources, and we will wrap for today here. Again, thank you so much to everyone and special thank you to you, Denise for driving this amazing conversation. Let me share my resources here: Share screen. Okay, so Denise, can you talk about these resources?

Denise Tilles: Yeah, yeah. As I mentioned earlier, still looking for really amazing case studies and it doesn’t matter if you haven’t like gotten amazing results, you’re still in progress. We’d love to hear about how you guys are leveraging AI, the case studies to potentially include in the book. So if you grab this QR code, I have links to this survey, to the quick form if you can fill it out and then I’ll get in touch with you. But it would be great to have so many voices from the community, new voices to include in the update of the book and the audiobook.

The second thing is the Product Ops book. It was lovely that somebody said, “This is our Bible”. I was kind of like, oh, that’s lovely. But it would be great if you guys would leave a review on Amazon. It sort of helps get more visibility and helps the community by making sure everybody’s aware of it.

And then feel free to get in touch if you’ve got questions, ideas, thoughts, challenges. Happy to talk it through. So all of that contact information is on this page here, just grab the QR code. And thank you guys. Wonderful questions.

Ana Andrade: Thank you, Denise. Leah is also sharing these links with you in the chat so feel free to see them there as well. And I think this was all amazing, and we have a great session today. So if these resources are not enough, you can also check the Product Ops Playbook by our sponsor, Dragonboat. Leah is also going to drop that link in the chat and it’s free, by the way.

And I think we are wrapping up, so apologies for not getting to all the questions. We will try to publish them in Slack so all the community can jump in and answer. Thanks again to everyone who joined us for this lovely discussion, and a special thanks to Denise.

I hope to see you all in the next meetup, and I hope to see you all in Slack. Have a wonderful day and until next time. Bye-bye!