Talk Companion | The State of ResearchOps: More Than Just Theory by Kate Towsey
Presented on June 7, 2019 at UXRConf 2019 in Toronto, Canada.
Watch the full talk on YouTube here. Want to listen audio-only? Find this talk on Apple Podcasts here and Spotify here.
Talk Overview ✨
As one of the faster growing fields of work, Research Ops is on the rise! But what is ReOps? Who does ReOps? Am I ReOps?
Watch as Kate Towsey, founder of the ResearchOps Community, answers these questions and more.
Key Takeaways 🔑
1. Invest Early in Research Operations
As UX teams grow, the complexity of managing research processes increases exponentially. Senior UX researchers should advocate for the early establishment of dedicated research operations roles to ensure the team’s scalability, efficiency, and effectiveness. The recommended ratio is one operations person for every five researchers to maintain optimal productivity.
2. Research Ops is Essential, Not Optional
Research Ops is not just about organizing tasks; it is a critical function that supports the entire research process, from participant recruitment to technology management. Understanding and implementing Research Ops allows researchers to focus on what they do best—conducting research—without being bogged down by operational tasks.
3. Clear Leadership and Strategic Alignment are Crucial
Successful Research Ops require clear and consistent leadership from UX research heads. Operations should align with the broader research strategy, ensuring that resources, tools, and processes are in place to support the team’s goals effectively. Changes in research strategy should be communicated clearly to allow the ops team to adjust accordingly.
4. Designing Effective Research Operations
It’s not enough to just handle research tasks; operations need to be thoughtfully designed to avoid inefficiencies. Senior UX researchers should work with their ops teams to create processes that are streamlined, scalable, and adaptable to the evolving needs of the organization.
5. Foster Collaboration Across Ops Functions
Research Ops should not work in isolation. Collaborating with other operational functions like DevOps and DesignOps can create a seamless pipeline from research to product delivery. This holistic approach ensures that the insights generated by UX research are effectively integrated into the product development process.
Transcript, Per ChatGPT 🤖
I work for Atlassian, and today I'm here to talk about one word, research ops. Two words, research operations. Three words, operations for researchers. And this morning when I woke up, I was thinking, oh, I could make this into 5 words and even 7 words, things like services for researchers or support for researchers and various things like that, although support for research is really 3 words if you're mathematically inclined. So the agenda today is what is research ops?
I'll talk about the rise of research operations and I'm going to talk about who are the people leading the way and I know we've got a few in the organization, I can see Matt over there, companies leading the way, Atlassian Research Ops team that I've been building up, very exciting to share that news, what research ops is not, and this is one of my favorite topics at the moment, and the stuff that we're grappling with as a new practice. So what is research ops? Who in the audience feels that they know what research operations is? Hands higher. Okay.
Not a lot of you. I've got some work to do. So here's a definition I wrote possibly about 6 months ago and, it's interesting because I look over it now and I think I obviously wasn't feeling very concise that day because really, you can bring it down to this. Research ops, it provides the roles, the tools, and processes needed to support researchers, and as I've just shown you, I could really put that into 3 words, support for researchers. Here's the research ops framework, which, some of you will have seen.
Who's seen this before? Ah, okay. Cool. So probably the same people who know what research ops means. It's a lot to look at on the screen.
I get very chuffed when I walk into an office, and I see that someone's printed it out and stuck it on their wall. It kind of fills me with a sense of pride and warmth and happiness and makes me think we should do it all over again, and then I think we don't need to. Here's a a list of all the things that are really included on that framework, a lot easier for you to kind of absorb in a in a setting like this. And you'll see it's everything from research participant recruitment, I have Sarit from Atlassian who works on my team in the audience, she can talk to you about that, it's research library management, it's program management, it's space bookings, it's a ton of stuff. More technical things too, like delivering AP's eyes, it's about migrating and sunsetting of tools.
I've got someone on my team working just on that, on making sure that our tech stack is is doing what it should do. So hopefully that's given you a better sense. It's a super quick look at it. Yesterday, I ran a one day workshop for 6 and a half hours on what is research ops and what does it mean to you. There's a whole lot more on the web, more and more coming out, I hope, as as we start to really, as a practice, get together and share much more and do more talks and publish much more about what we're doing and learning as we go along.
So the rise of research operations, I'm sort of introduced occasionally and it still kinda stuns me as, like, the queen of research ops or the leader in this field, and it feels like it's, been quite a journey for me as well. So 2018. Alright. Let's get to this first. Research ops is kinda interesting because why now?
Like, why is it this interesting conversation right now? Well, teams are getting bigger, it's and and they're getting more sophisticated. The bigger they get, the more sophisticated they get. I don't know if anyone's read Yuval Harari's book, Sapiens, and he talks about communities growing and they hit a certain point and now they need to have legislation and all sorts of things built around them that didn't need to exist before when people were just able to chat to one another and organize themselves. And this is very true for our research teams.
So 7, 10 years ago, you'd find 1 researcher sitting in the corner, and they'd deliver a PDF report, and that would be kind of their job, quite lonely. Now it's very, very common to find teams of 10, 20, 30. 80 is pretty common too. I'm sure many of you would have big teams like that in the audience, I know some of you do. Up to 100, 300, 400 in the very, very big tech companies.
And as Yuval Harari says in Sapiens, now suddenly, you're not just 5 people needing to chat to one another and coordinate, you're a much bigger bigger space, bigger organization that needs stuff around you and needs efficiencies built in. So I've been working on operations kind of stuff, since 2013. I cut my teeth at, Government Digital Service, built them a lab, I built them a research panel, did all sorts of things for them in that space, but I felt super lonely. It was around 3 years of just being like, oh, am I the only person who really cares about this stuff? I was writing blogs and kind of communicating with the world, and people were coming back to me and saying, hey, like, I'm doing this stuff too.
I'm really interested as well. So I kind of thought, well, why do I keep having these conversations 1, 1, 1? Why not make a community? A super small community because not many of us are interested in this stuff, it's just me and maybe someone from Etsy and maybe someone from some other big tech company and so I started the Slack channel and within 2 months it was 200 people. And not just 200 kind of like, well, it's sort of another Slack, I've dying to have another Slack.
I just needed more communication and more information. But 200 people were like oh my god, this is amazing, I really wanna talk about this stuff and this has been really bothering me. Now that community is 2,500 people and it's kind of a year and a half later which is quite astounding. I no longer run it because it became very heavy on the admin front and, full time job building up research ops at Atlassian was kind of more than enough for me. But we got together as a group of people, as 200 people and, 200 enthusiastic people, and we did something very meta.
We researched ourselves. We were researchers who wanted to research researchers to find out what researchers want from wanted from operations. Obvious, right? And we called it hashtag what is research ops. I was really big on that this didn't become a USA, UK, European kind of conversation and that we were really global, and so I went shopping around the world to find other people who were as enthusiastic, and so we ended up with, spread around the world, Japan, India, Russia, South Africa, and I made some very good friends as well, which are now very personal friends around the world and got to know a lot of people.
In the end, it was 34 cities and 17 countries took part which is quite amazing. And all of this driven by volunteer effort just kind of the industry working kind of across companies, putting companies aside and just saying, hey, we're an industry and this is stuff we care about. We care about what we do, and let's work together to make it better. We also ran a survey and, all in all, between workshops and surveys, we spoke to around a 1,500 researchers, and that's how we got this framework together. So some of it was me on a one fine day when I just moved into my new apartment in Sydney.
I had no furniture. It was kind of a few rental pieces, very minimal. I thought I was a minimalist until that actually happened, and I thought, oh, shit. Okay. I've kind of said we're gonna make this framework and now I've got all this data.
We thought we were gonna get very little data, but now we've got this ton of data and, what are we gonna do? So I kinda got into MURAL and, a bunch of people had done a ton of work on analysis and given me some kind of process data and, I spat this out and it seems to have been taken quite nicely by everybody. So let's look at people leading the way. This talk is about the fact that research operations isn't just theory. It isn't just 2018 and this framework and a community that has come up with an idea.
It's much more than that. So here's Airbnb, and we've got Matt in the audience and I know a few other people from Airbnb, obviously, a sponsor in San Francisco. I speak to Tim Toye quite regularly and, he's got 9 people at the moment, growing to 12, I think, in the next short while. They've been going for 2 years and Tim's quote is it's challenging scaling a team with limited resources to provide comprehensive coverage to a fast growing and fast paced research function That way scaling comes up time and again in research with researchers, never mind research ops, and the interesting thing is as research scales, research ops need to scale too, and so your problem becomes our problem as usual. We like that.
We like it to be that way. Here's Deliveroo in London. They've got 3 people. It's been a team that's been going for a year. And this is Saskia, who has also become a friend of mine.
We spend a lot of time as operations really kind of talking to one another. In fact, when I gave up the big ops community, I decided I still wanted community, which is why I started it in the first place. And so I started a much smaller club, I call it the Illuminati of Research Arts. It's the ChaCha Club. ChaCha stands for cheerleaders and chums because you definitely eat cheerleaders and chums when you're working in this this new emerging space and there's no slack, you'd be surprised to hear, and we just meet once a month.
We also don't do talks to each other, we just talk to each other. I think there's a as a complete side note there's a lot of, this kind of situation which is great, obviously I wouldn't be here otherwise, talking to people but we don't spend a lot of time actually just talking with one another, so that's what we do. So here's Saskia, and, she, has built up a team to 3 people in 1 year. Her quote is really lovely, and I agree. It allows me to indulge my geeky planning spreadsheety side as well as my people loving supportive side.
It's the best of both worlds, and it's incredibly fulfilling. Booking.com in Amsterdam, 10 people, going for 6 years just shows you that this is not a new discipline. It might have emerged and become a thing in 2018, but it's not new, and this is Carrie Peters, who says, the spark behind scaling research operations is recognizing and communicating the need as soon as possible. The longer you take to do that, the harder it becomes to amplify the value of your research investment. So true, I recommend that as soon as you've got 5 researchers, start to think about operations.
Anyone who knows is nodding their head right now which is always validating. I have an algorithm, actually, now. I used to think it was 1 ops person to 10 Now that we've been running around like crazy people for the last 10 months at Atlassian, I realized that it's really, 1 person to 5. Spotify in New York, this is Lucy. Lucy has actually moved on from the role and into a research role, Every ops manager's worst nightmare.
Please don't move. It's not as nice as they make it look. So it's been one person at Spotify. If you want a research ops role in New York, it's a great office. They have excellent snacks.
There is a job going there as far as I know still, and she says the potential scope of research ops is vast and, therefore, requires working closely with the team to understand their needs and to develop and communicate a plan of attack that is both realistic. It's worth pointing out that word realistic. Realistic to existing ops bandwidth when you've got one person that's fairly limited, as you'll see, and flexible enough to evolve with the needs of an organization. That's really a very, accurate quote, that, from Lucy. You've seen this.
She's been there for a year and experienced it, and this is her war story in a short amount of words. And finally, here's Atlassian in Sydney, the team that I've built up. Atlassian's had research operations for 2 years. I've been there for 10 months and prior to me being there someone, was running as just as one person, and we've now built up to 6 people which is changing a little at the moment with strategy. I'm sure you're all in kinda coming into new financial years but with 6 people.
And my quote, I love it when I can quote myself, don't underestimate the resources you need to get the job done. So you'll see with those quotes, it's really about scale. We're in start up territory as research operations. We're always trying to do more than with what we've got trying to stretch our resources, and it's very much around how do you scale? And some of my thinking is reading books on start ups and business.
How do you scale and not fall apart? How do you deliver so you can get more? But that's not a comprehensive list, these are just people that I've managed to get a quote from and get them to agree that I can talk about them on stage. Microsoft has a big team and they've been going for 20 years with research ops specifically. LinkedIn have a team, Facebook, Google had, I know, they had 50 people working only on research participant recruitment just participant recruitment Stripe has Rowan has 1, she she's the only one working at the moment, Dropbox has 6, HMRC and actually the Home Office is not on this list, and the Australian government have invested and are investing more and more in research operations which is very exciting.
I don't know how to say this word. If anyone does, shout it out. It's Vrbo? Vrbo? I don't know.
Not sure. Vbr0? How do you say it? Vrbo. It is Vrbo.
Okay. Cool. There you go. You learn something every day. Workday and Indeed, and more and more.
I actually learned that Oculus have a good research ops team yesterday and Mailchimp, of course, as well. So, I love to meet people who are working in this space. If you happen to be one of those people and you have not had a conversation, come find me. So let's look at the Atlassian research ops team, my team, and and here's all their faces. Yesterday, I ran, as I said, a one day workshop on research operations and I have the squeezey chicken, if you look on Twitter, you will see it, called Henrietta and I've been told that I need to put her picture in here as part of the team.
So this is a, a graphic of, my team and the structure particularly going into f y 20. This is not necessarily from, at least my studies, unusual. A lot of the research ops teams will have a very similar structure. So we have the head of research which in my case, very thankfully is Lisa Raykal. It's incredible to work with.
I've worked with her since 2,012 actually. And very very important to a research ops function to have a really strong research leader in place. Then there's me, the research ops manager, and, program manager is someone who sits side by side and they're kind of, traffic controlling and defending as well, looking for spaces and also saying there are no spaces for research projects going. Then we have our research and insights. That's what we call, research at Atlassian, research and insights.
And we have our service desk team, and, actually, Sereet's sitting here in the audience. She she's leading on that team, and they look after research participant recruitment, a really big, big, big job, and after all sorts of other tooling requests that come for through our Jira service desk. We've set up a Jira service desk that researchers and people who do research, can use to submit requests to us so we can send them services and look after them. Then I have Teresa in research technology, and she's actually recently moved over from marketing tech to come and build out our tech stack for research and she's the person looking after our massive big survey tool, our our decline from mentioning the name and, because I could get nasty, and, how the data moves from one place to the other and how it goes and sits with the massive big set of customer data we've got and so on and so forth. This is a full time job.
Migrating things, renewing contracts, looking after all these various things is a big job. Then we've got engagement and insight. So researchers have done all this work, and then what happens with it afterwards? There's things like that, but also branding for the team. When you're trying to build up your team in an organization, it's pretty cool to look sexy.
Right? Have some nice stickers, have some good brand, make your content look nice, your, Internet home page should look nice, blogging internally and externally, these things need someone to look after them. It's a full time job. So that's event and communication, and next to that is our digital librarian. I've just recently, someone has converted from researcher into digital librarian, so it works both ways.
Just be careful. I can steal researchers as well, and I do and I do try. And I'm very, very excited. I've been working on library type stuff for around 6 years. I have new hypotheses at the moment about the fact that it's not a library, it's a log book.
And we're gonna start experimenting with that. So I hope within a year's time that I can come back with either a massive success story or like, oh, shit. I was wrong. And then, I'm hoping to get in a research ops coordinator who can cover all the paper cut type stuff, the procurements, the budget tracking, the space bookings, the very important stuff of cake. People need cake.
I am often called cake ops because I spend a lot of my time ordering cake for birthdays, for 1st year anniversaries, for you stubbed your toe, that's okay, we'll get a cake. It's fine. So this is the list again, the laundry list you saw earlier. If I go to here, this is really when it was just Sereet and I, we could cover about this, and we were hot under the collars running around like crazy trying to get this stuff done. Participant recruitment, celebrations and condolences aka cake ops, vendor management, service design management, and thank you gifts for our participants.
This does not look like a long list, but it is a lot of work, and that's why I said do not underestimate. Now that we're 6 people, we can do something more like this, but we're not covering all those things, and so you start to see a team of 6 sounds like it's a lot of people to do operations, but really, that's what you can cover. And if I get another couple of people in, I could start to cover those things, and I start to get a bigger picture. But there are areas if I go back a little bit, to there. You can see the plus signs there, and that means that if there are more researchers, I'm gonna need to do more research recruitment.
And if you're not dealing with PWDRs or people who do research, you're gonna need more recruiters as well. And if you have more coordinating to do, you'll need another coordinator. So I put kind of plus signs into the teams where I need those to grow, and that's where we get to that that ratio of 5 to 1, and so I've got an agreement with Lisa, and it's good that I'm now saying this publicly so she doesn't, change her mind, that every time we get 5 researchers, we get one operations person so that we can actually have a life, which would be great. Okay. So one more back.
Here we go. So there's this little word in the corner, ops service design. Now, so easily forgotten, and I've actually put it in a separate line item because it's easy to forget that you've actually got to design your operations. Otherwise, all you're doing is you're moving the ton of inefficient work from the researcher to operations and then you're just doing a ton of inefficient work again which, if I'm honest, is where we've kinda been for the last 10 months in a sense because we haven't had time to design our services and so my work as a manager is trying to carve out the time and get us in a space where we can actually design our operation service so that we are effective and efficient and we're operationalized and not just doing all the researchers' work. It's like this distinction that I probably couldn't have gotten to unless I'd actually been running around for the last 10 months realizing we don't have time to design and we're gonna start to move into that as well in FY 20.
So here's my favorite, favorite part, what research ops is not I spend a lot of time ranting about this on Twitter. It is not research leadership. That's Lisa's job, and there are so many things where I just say don't ask me that's Lisa's job. She's the person who decides who is in the team, what skills they need, how they're gonna improve their skills, where they should be going, all that kind of jazz. That's Lisa's job.
What kind of research we're doing or not doing is her job. It is vital for operations to have very clear research leadership. Without that, I'd be lost at sea. If Lisa changed her mind, in 3 months about the fact that I'm not saying this is our strategy, but she said we're gonna focus on doing lab research, and I run around for 6 months and build labs in all our geolocations, and then she goes, you know what? Actually, lab research is so yesterday.
We're gonna do most of research. My entire research ops strategy changes. I'm now suddenly a travel agent looking at visas and putting together field kits as opposed to building labs in geolocations around the world. So you really need a strong research leader who who knows where they're going, has a 1, 2, 3 year plan, and something that you can track against as operations. From a philosophical point of view and a methodological point of view, research strategy can shift and change.
From an operational point of view, we're putting contracts in place with vendors, buying equipment, hiring people, who are committing to developing skills in an area so things can't change all the time for us. Research strategy relates to the previous point. We are not about research strategy. I respond to research strategy all the time and I try and put processes in place to make that research strategy happen. Now, if you get annoyed with me I I want you to wave your arms at me because it would just inspire me to be more anti it is not researchers being more organized and and I love it that research operations can inspire researchers to be more organized, but to be honest, you're all good people.
I've worked with you for years. You wanna do a good job. If you were gonna be more organized, you would be doing it already. You can learn from us, and we can help you, but it's unlikely with the massive amount of work you've got as researchers that you're gonna suddenly find inspiration to be more organized just because we're here. Besides, I like my job, and I don't want you to have it.
It cannot be done by one person. I think you would have seen from the laundry list that I had and from the the organization that we're building in Atlassian or the team we're building, we don't build teams I mean, we love teams at Atlassian. We're all about teams and and openness and all of that, but at the end of the day, we're not gonna build a team just because we wanna have fun. I mean, that's nice too. It it just cannot be done by one person.
It's not physically possible. And so a researcher doing ops on the side is what I call just being more organized. Right? It's akin to you might be a wonderful home chef, a foodie, you might go home and make wonderful, incredible meals that could be in a Michelin star restaurant. It doesn't make you running a restaurant.
It's a very different thing from cooking at home and running a restaurant, and this is where our conversations as operations become very different to that of researchers being more organized. I love to be shot down over this. I often come up and I sound like I've, like, decided and I've got a serious opinion. It's it's, like, it's not the case. I can be argued into other ways.
I say that to encourage you, because we're a new practice, and we're still forming, where we are. So here's the stuff we're grappling with. It's wildly underestimated. I'm gonna fairly quickly go through these because I've covered a lot of these already. It's proving you're worth a very little resource.
It's that scaling thing, it's that start up space. There's confusion as to what it is and is not, and that's within industry but also very much within the organization. A lot of my work is, yeah, yeah, but that's our job and why are you doing that? Give it back here. Or no, no, no, this one's not our stuff.
It's theirs. And it is also figuring out because we're learning figuring out what is our stuff and what isn't our stuff. Heavily reliant on clear and consistent leadership and reliant on close collaboration across the entire organization. We need to make friends with, and I do, make good friends with, estate management I need spaces With the legal team, I work with them all the time. With procurement, with finances, I make very good friends with them.
Even with building security, if you are building labs, you need to know them, and so on and so forth, we get to know the entire organization. In terms of opportunities, we're building a practice from scratch. How super exciting. I've shown that it's not entirely from scratch because there are people who've been doing it for a long time, not to mention medical research and research ops in those spaces. They've been around a very long time.
But as a technology space, digital space, we are building from scratch. And as part of that, we're a pretty small community. In Cha Cha Club, there's, like, 15 of us. Anyone else who has a research ops job title and does it full time, get in touch. I'd love to have you on board.
But it means we get to know each other very well. These people are my friends. I've gotten to know them over the last year and a half, and that loneliness that I had when I sent up that first tweet has been fully looked after, which is a very wonderful thing. Aren't you pleased? We're exploring working with other ops in the organization.
Okay. Fine. Research ops. But there's design ops, there's dev ops. Fine.
There's all these ops y things, which is people kind of, like, get shocked when, as someone who's, like, seen as the research ops person, that I'm, like, who cares? You know? We are delivering a service to users, and there are certain there are pathways to get there, and all of them need some kind of operational support. And I really wanna work with DevOps and DesignOps to understand what is that full pipeline and how do we build something that is going from we did some research and then we did some design and then we developed something and now here we are. We centralize the research effort and costs.
A sort of quick point is, it's interesting you start to centralize what is costing money anyway it's not you necessarily doing more research possibly but not necessarily and it can be in the 100 of 1,000 a year and that's very interesting from an executive point of view where everyone's like oh we're just doing some research and now it's holy shit, it's costing a lot of money. Maybe don't be so close to customers all the time, like, just a little bit, like, skillfully. Only people who are skilled should be close to them. I think it's a very interesting conversation for the next year or so. So my last call to you is help your researchers focus on research.
You might be thinking this ops thing seems like it's an expensive extra effort. They are making the effort anyway. Help them do what they do best. Not every researcher is an excellent travel agent as well as an excellent recruiter and so on and so forth. They don't all know about governance, and they don't all know about GDPR and legal and procurement and shouldn't have to learn all those things all the time to get the job done.
As I said, one ops person per 5 researchers is what you get is what will get you going. I suggest that when you've got 5 researchers or even 4 researchers, look at getting that 1st ops person in. Stay connected with me. Here's a variety of ways to do that. I do a lot of work with Rosenfeld Media Design Ops Community.
There'll be a lot of speakers from ChaCha Club and from elsewhere coming in to do talks on what we're actually doing in this space.
And thank you very much.
Thanks for reading! You can find more about our upcoming events and content by signing up to our mailing list, following us on LinkedIn, or subscribing to our YouTube channel.
See you soon! 👋