We picked the brain of Klaviyo's Chief Product Officer, Conor O'Mahony, on the hurdles companies face when it comes to building products. He stressed the importance of data-informed decisions to determine how much to invest in a product—driven by qualitative and quantitative data, and gathered from user research and MVP feedback.
From O'Mahony's perspective, he shares that many companies often fall short in data collection and integration—but he predicts a future where the data process will be less labor-intensive.
We've included the full transcript from our interview for you below:
Margaret Kelsey:
Welcome everyone to Voice of the Product, an interview series where we talk with leaders of organizations who are embracing the product-led movement. Today, we have Conor O'Mahony, chief product officer at Klaviyo. Conor has a rich tradition of leading cultural transformations and digital transformations among organizations. So thank you for being here.
Conor O'Mahony:
It's my pleasure.
Margaret Kelsey:
And congratulations on your series B. Can you tell me a little bit about Klaviyo's product and the stage of company that they're at right now?
Conor O'Mahony:
Sure. So Klaviyo is all about growth. We are growing like crazy and primarily because we focus on helping our customers grow. Klaviyo essentially marries a deep analytics engine together with marketing technology to help our customers provide the right message to the right people at the right time in order to help them grow.
Margaret Kelsey:
And so you have been with the company now for about two years. Is that correct? A year and a couple and seven months or so?
Conor O'Mahony:
That's right. I started off as an advisor, and then after about a year as an advisor, I came on as the chief product officer.
Margaret Kelsey:
And at the time, you were advising the CEO and the chief product officer as well?
Conor O'Mahony:
And the head of products.
Margaret Kelsey:
And the head of product.
Conor O'Mahony:
Yeah.
Margaret Kelsey:
And so tell me a little bit about what that advising work was looking like or what kind of problems they were working throughout that time.
Conor O'Mahony:
In my experience, a lot of companies experience the same kinds of challenges, how to figure out what to build, when to build it, how to make decisions, how to distribute decision making then through the organization and how to have that decision making be as data-informed as possible.
Margaret Kelsey:
And so you say data-informed, but not data-driven.
Conor O'Mahony:
By being data-informed, you can figure out the right degree to which you want to invest in a particular part of the product or not. There's a big difference between building out a fully-fledged feature and a feature that's good enough to meet the needs of the market.
Margaret Kelsey:
And is data the only thing that tells you whether or not to build out maybe an MVP or good enough versus a big product? Or are there other sort of indicators of which way to go?
Conor O'Mahony:
Well, user testing is a form of data. There are a set of data points there. Feedback from an MVP are a set of really good data points. So those all go into the mix. It's not just user research ahead of time. It's not just market penetration for products. It's not just usage levels of certain capabilities. It's not just incoming feature requests from customers. It's the marriage of all of that together.
Margaret Kelsey:
And so it sounds like there's both qualitative and quantitative data married together to make these decisions?
Conor O'Mahony:
There is qualitative and quantitative data. A lot of the qualitative data can be positioned in a more quantitative manner, if you have enough data points.
I've got a story from a previous job that I worked at where early on in the job, I came on board and I started digging through the data. And what I did was I basically measured how everybody had spent the last six months of their time. And then what I did was I went and I turned that into a view of the different projects and where everybody had spent their time. I worked with finance to translate their time into a dollar amount.
And then I got up in front of the company and I said, "Hey, you know what? For the last six months, the company has spent $400,000 in this particular feature, $20,000 in this other feature, and so on." And it was real interesting because the people who had worked on those features had an immediate visceral reaction. They talked to themselves, "Oh, $400,000 on this feature that had no impact on the business? $20,000 in this feature that had a big impact on the business?" And I got an immediate reaction. And the immediate reaction was that people then volunteered saying, "We need to do this on an ongoing basis." And what it did was it actually started to shift the culture of the company where they had a much greater appreciation for the fact that their time was an investment from the company. And that, in turn, we needed to not only measure that investment, but we needed to measure to return on that investment. And over time, the company became very focused on ensuring that we use the right data to inform the right kinds of decisions.
Margaret Kelsey:
Are we drowning in data right now as folks were worried about growth in product?
Conor O'Mahony:
I don't believe we are. Sometimes, it feels like it, but in my estimation, we're actually not gathering enough data. The reason I say sometimes it feels like it is, because a lot of the data we're gathering is fragmented. It's fragmented and it's not easily combined with other sources of data. And because of that, it can feel very unwieldy, first of all, just to gather in one place, so it can be worked with, and so it causes a lot of pain. And because of that pain, it feels like we're drowning in data. If we're to look at the inevitable that's coming, the inevitable that's coming is that we're going to continue to leverage data more and more in our decision making. And if we're to do that effectively, we do need more data. I think it's inevitable that there's more data coming our way, but we just need to find better ways to handle it.
Margaret Kelsey:
And what are your biggest recommendations for folks that are handling a lot of data or feeling like they're drowning in data themselves to either clear it up or to somehow gain some sanity around all the data?
Conor O'Mahony:
So right now, Klaviyo is investing a significant amount in people to go and clean that data, integrate that data, be able to analyze that data. We have ops resources, BI resources, engineering resources that are all being put to play around us right now. We're making that significant investment because we believe in data informed decision making, and we believe that it's necessary for us to do that in order to grow effectively.
Margaret Kelsey:
And it's a people process right now. Do you think in the future, it will be... We'll have better tools or better software, or where do you see this world of data in the next 10 years or so?
Conor O'Mahony:
It has to evolve beyond a people resource intensive process. If you look at a lot of the announcements in the world of data and business intelligence, there are a lot of exciting things that are happening right now, a lot of interesting acquisitions, a lot of interesting product developments that's only going to accelerate, as we move forward.
Margaret Kelsey:
It sounds like product-led growth is not necessarily automating every human touchpoint. And if that's the case, how do you think about growth with product-led growth?
Conor O'Mahony:
Yeah, so here's literally the way I think about it, as I work at the C-level. I look at, in Klaviyo's case, us growing the team in line with the growth of the business for the next several years. So if the business revenue is growing at this rate, I expect a team to be growing at a very similar rate, but then I do expect an inflection point where the business will keep on growing at the same rate, but the team size will grow at a lower rate. So I always expect the team to be growing as we're going forward at Klaviyo, but just not at the same rate as the revenues of the business.
Margaret Kelsey:
So Conor, you have worked at both companies before Klaviyo that were non-product-led. And now at product-led companies, what are the differences that you've seen between the different kind of sets of companies
Conor O'Mahony:
At companies that are not product-led, now what you can often have happen is large customer opportunities end up driving the roadmap and the product strategy. And then what ends up happening oftentimes is you end up with a product that's not very cohesive. It's meeting specific needs from different customers that are not necessarily aligned. So what you end up is you end up with a bit of a Frankenstein of a product. It does a pretty good job. But in my experience, what ends up happening then is you also end up accumulating a bunch of product debt along the way, because oftentimes you end up with multiple underlying platforms that are each designed to meet different specific needs of particular customers. So what I often find ultimately the big difference is that non-product-led companies end up having a ceiling, a point above which they struggle to grow. They end up plateauing in a way that growth-driven product companies don't.
Margaret Kelsey:
And it almost seems like if a company is building a product just for a particular enterprise company, and let's say that that company leaves the product itself, what has happened and have you seen that happen? How does that change the product, if you've built a product for a company and then they leave and are no longer a customer?
Conor O'Mahony:
I've had direct experience with that, where it can cause a lot of turbulence, because what'll sometimes happen is you'll build a particular feature for a specific customer. Let's imagine they're paying you many, many millions of dollars for that, but you'll also start accumulating some smaller customers who can take advantage of those same features. The problem is then if the big customer pulls out, you're left supporting a bunch of smaller customers in perhaps a non-sustainable business model.
Margaret Kelsey:
And to go back when you were talking about product-led growth companies not having the ceiling that non-product-led growth companies can have, can you talk a little bit about what that ceiling looks like and why product-led companies are unique to not have that ceiling?
Conor O'Mahony:
Yeah. So I actually experienced this in a couple of companies where we literally hit a revenue plateau above which we really struggled to move. And now, when that happened, we realized that we had a couple of options. One was to reformulate the product and make some really hard decisions about the customers that we wanted to keep on that. Two was to look at alternative ways to grow. And in those cases, we looked at the alternatives ways to grow, which included growth through acquisition, which is a very different dynamic and has a big impact on the business on a few dimensions, including how the company is valued.
Margaret Kelsey:
So Conor, word on the street is that you lead the product team that also is the design and product together, customer success, customer support, and a few other ancillary teams. How do you go about creating team alignment throughout all of them in terms of both projects that they're working on, but also metrics that they're looking at?
Conor O'Mahony:
So it's interesting. As a product-led company, we want the product to be as informed by our customer touchpoints as we can. So that was part of the rationale for having me also lead the customer success and the customer support and the academy teams. Those are really important customer touchpoints. We want to make sure that there's as little friction as possible in getting information from our customers back into our product and design teams.
Margaret Kelsey:
Interesting.
Conor O'Mahony:
So there are a lot of synergies there. It's interesting. I've talked to one or two other product leaders in North America, and this is something that other companies are also doing. I think it's too early to call it a trend, but I think it could emerge as a trend.
One of my personal biggest frustrations has been around what are called Voice of the Customer programs. This is where a company will typically try to gather from their customer touchpoints, information to be used by the product team. A lot of those programs I've found have been very superficial. Yeah, they've maybe tagged some information in support tickets and in customer contacts, and they've done some reporting on that. And perhaps, there's been some level of analysis and writing around it. What I believe is needed is you need some glue to tie it all together. So one of the things we're doing at Klaviyo is we're creating these people in our customer support organization who are what we are calling sages for the product or for a particular part of the product. And now, their job is, first of all, to compile all of the customer feedback for that area to [inaudible 00:15:05] sage for.
Then what they're tasked with doing is getting together with their product manager for that area. When the two of them are meeting regularly, it's a two-way exchange. The sage is representing the voice of the customer for that particular area. The product manager is informing the sage on what's coming next. And there's just so many benefits beyond that because that sage is then going back and spending time with the rest of the folks on the customer support team, educating them better on what's coming in that part of the product. The sage is also gathering more input from them to return to the product manager. It's little programs like this. It's programs where you have people that are acting as the glue here that are representing different parts of the product. That's the secret to making these programs effective and actually enacting change in the product.
Margaret Kelsey:
So one of the hallmarks of a product-led company is an easy way to get into the product and to get your hands on the product, so a free trial or a freemium experience. What has Klaviyo done to create that easy experience that folks can get their hands dirty with the product immediately?
Conor O'Mahony:
Yeah, we do have freemium product that we have, that we offer, but far more interesting is an internal initiative we have. It's all about reducing friction. What we're trying to do internally is we're trying to have it, so that our customers can be fully up and running in our product within 60 minutes, and we call this Klaviyo in 60. It's an internal initiative where we're looking at every step along the way.
Margaret Kelsey:
Does that user experience, is that holistic of the marketing experience into the product and the product itself, or is this more of a product team initiative?
Conor O'Mahony:
It has to be holistic. We need to look at our touchpoints from marketing through sales, through the product, through the ongoing account management, and make sure they're all consistent. We also need to make sure that at each of those touchpoints, we're providing a set of user experiences that are cohesive and then consistent. We're setting the right expectations, and then we're either delivering, and those expectations are exceeding.
Margaret Kelsey:
And with a complex product, like Klaviyo, how has it been to reduce friction so far? Has it been easier than you've expected or more difficult?
Conor O'Mahony:
It's a real challenge. It's a real challenge, but it's a real challenge that the people who work at Klaviyo embrace. A part of the reason some of them have come to join Klaviyo is because we want to solve this challenge.
Margaret Kelsey:
Conor, is there something happening in the market or in the world in general that you personally are very excited to dive into deeper, learn more about?
Conor O'Mahony:
Yeah, absolutely. I'm really excited about the potential for machine learning. It's an area that at Klaviyo are making big investments in. There is a lot of hype around machine learning, but there's also a lot of very real value that is being delivered. In some cases, it's obvious that it's being delivered, but in more cases, it's not. And that's a lot of the key around how to seamlessly integrate it into a product is to have it and be integrated in such a way as people are not even aware that you're doing smart things for them under the cover based upon the results of some machine learning.
Margaret Kelsey:
And so I think that speaks to what you were talking about, about reducing friction, so it shouldn't even be a thought in your mind that machine learning is the reason that this personalized experience is happening. It's almost magical.
Conor O'Mahony:
Absolutely. Absolutely. And when you have that seamless experience, it's just so smooth and it tends to delight, but people don't put their finger on the fact that it's actually delighting them, but they have those positive associations with the brand. And then, in turn, it leads to word-of-mouth references, which are the most desirable references you can have.
Margaret Kelsey:
Absolutely. Well, Conor, thank you so much for being here today and sharing your valuable time with us.
Conor O'Mahony:
Thank you for having me.
Margaret Kelsey:
And thank you so much for spending your time with us today as well. Don't forget to check out other articles and videos on the product-led growth collective and have a wonderful rest of your day.
