Director of Product, IFTTT
Ten Tips for B2B Customer Research
Yana Welinder: Great. So I'm going to go ahead and get started. Uh, first of all, I want to thank Sasha and all organizers for doing an amazing job with what looks like it's going to be a fantastic event. It's a beautiful, uh, San Francisco Day here out on Presidio, a gorgeous, um, venue. I'm really excited to be here. I'm Yana Welinder. I head up product at IFTTT. We're working to unlock remarkable experiences by connecting services. Um, that could be anything from um, getting, uh, users to open their garage door automatically when they come home or being able to tell Alexa to turn on your Roomba, um, variability any time you're trying to connect to services and have a really seamless experience from B2B perspective. We're empowering businesses to offer richer experiences to their users. And that's sort of connected to my own personal passion, which is to solve core everyday problems with incredibly simple UI.
Yana Welinder: Um, and there's no way to know if you're actually solving anyone's problems without doing customer research. Um, so that's what I'm going to talk about today. Uh, specifically kind of the tricky art of doing B two B customer research. So why do customer research, right asks no one ever, uh, or, or at least not anymore. So who here is some product? Okay. So whole lot of folks, so I don't necessarily need to tell you that doing customer research is really key to save time, resources and your team's morale developing something that no one ultimately wants to buy or that customer research helps you with. Uh, having that flexibility to figure out what the right solution is instead of kind of, uh, uh, is playing catch up with, with your competitors. Um, so if it's so important and key for product development, why don't people always do it in, uh, B2B situations?
Yana Welinder: It is just more trickier than doing B2C customer research. So I'll talk a little bit about why it's more tricky, uh, and then talk about some ways to get around that. So, uh, the first, uh, kind of issue with B2B customer research is that it's really not just a product team project. Um, unlike with B2C, uh, research where you can really figure out your own channels for that, you can, you can do it in product surveys. You can follow up on those surveys with, uh, with interviews. You can, uh, reach out to folks on social media. You can talk with friends who kind of fit the, the target target persona. Um, with B2B you really have to do kind of cross functional collaboration and team up with your, uh, sales team, your business development team, your customer success team, um, to ultimately set you up with the right people to talk with.
Yana Welinder: Um, and then sometimes you're not actually going to have the opportunity to, with your customers directly. Um, and those cases, I tried to figure out ways to parse the feedback I get from the customer facing folks in my organization to be able to tell what is, what are the actual customer needs in that situation. So I'll ask about things like, who was the customer, um, you know, what, what sort of context the feedback come up and what's their use cases, that kind of stuff. Um, and then usually it's not going to be a whole lot of information, but it's going to be enough to get me started and building up personas that I can then use to create a, to, to kind of take the next steps in my research. Um, and then in addition to setting you up with the right people, sales can also be helpful in just kind of managing that relationship in setting up the customer for the conversation.
Yana Welinder: So I'm usually in nothing more frustrating them. Um, you know, having figured out what the right customer is, get on the phone with our customer. Um, just to have them tell you that they can't tell you anything about what they do. Uh, they can't tell you anything about how they plan to use your products. Um, that really sucks when you get to that point and nothing more frustrating than hitting it a confidentiality wall. Um, so that your sales manager can reach out to the customer ahead of time, make sure they've signed the appropriate NDA to be able to answer the type of questions you're interested in asking the customer, and then they can get the customer comfortable talking with you by having sort of inappropriate introduction. So may seem like a minor point, but it's actually kind of something you come across quite often. Um, so I'm super fortunate to work with an amazing sales team at lift.
Yana Welinder: They really understand the value of targeted product development, uh, and what kind of information they need for that. But from previous experiences, uh, both for myself and with talking with my counterparts at other companies, I know that collaborating with sales can be kind of one of the sticky points in doing B2B research. Um, it's quite natural. So sales teams will sometimes try to, uh, sell things that are close to deal with things that are not necessarily on your roadmap. Um, anyone have that experience. Yep. Quite common. Uh, and then, uh, and then you come in and you provide all this clarity about what's actually on the roadmap. Uh, in a, obviously they're going to be nervous about that. Um, or you could have a situation where actually, um, you know, if they're, you're, you're not, you're telling all the right things, all the right things for, on the roadmap and the customer gets so excited about the things that will come that they wouldn't wanna close the deal, then they would just want to wait until the thing launches next year.
Yana Welinder: Right. Um, so for all those reasons, sales are naturally kind of nervous about customer research. Um, one way to try to get around that is to try to avoid perspectives altogether. So you like talk with, with customers who are not in the moment where they're going to be evaluating your product offering and they're also not close to kind of that renewal cycle. Um, that's a kind of a good, good time to, to, to talk with, with customers. Uh, obviously that's not necessarily gonna work if you're just starting out and you don't have a whole lot of customers. Um, so you have to talk with perspectives or you're, you're going to be shifting focus to a new market segment. Um, you don't have the right kind of customers and you want to talk with prospective customers because they're the ones you're going to be interested in.
Yana Welinder: Um, in any event, whether you're talking with perspectives or uh, or your existing customers, you want to get sales comfortable with what you're going to do. So don't try to get into a one on one situation with a customer in till you're a sales manager or, or customer representative that, uh, um, you know, to not join the conversation. Um, instead try to get them to see you in action. Um, I found that it's really helpful to catch up with sales ahead of time. Ask them about any kind of like sensitivities that the customer may have. Um, share my questions ahead of time. Um, have them provide feedback on the questions. Is there anything here that you think may be problematic? Ideally, if you're actually, if you're asking the right kinds of questions and I'll talk a little bit about what those questions are in a moment.
Yana Welinder: Um, you're never really gonna get to the point where you're exposing the gaps in your product offering cause you're not going to be asking specifically based off of my product what are the things you, you need, you, you'll be asking slightly different types of questions. Um, and then you'll never want to outsource the customer interview to someone who may not be mindful of the customer relationship. I've done that mistake. I've outsourced, uh, a, uh, customer interview wants to a design consultant, uh, who ended up pushing on the customer too much. Uh, and the customer kind of wasn't happy about that and sales had to step in and try to mend the relationships. So, um, learn from my mistake. Don't do that. Um, and then, um, what happens if the, if you're a customer facing folks or seals, uh, asks you to join a conversation or a visit, uh, when the customers is in town but won't let you do a proper kind of customer interview.
Yana Welinder: She joined that. Yes, yes, yes. Do it, do it. So I usually tried to jump on every opportunity to talk with customers. I think PM's can sometimes be a little bit kind of, uh, hasn't been about that because, uh, you know, you're, if it's not your responsibility to pitch and sell, uh, and, uh, you know, you want to save your time to the most impactful moments where you really get to ask the right kinds of questions in yeah. Customer interviews, like the really targeted ones will be most impactful, but there are just fewer of those opportunities. And so for going a moments where you can ask, uh, some, a few tactical questions here and there, but generally just carefully listening in and, and, uh, catching moments of, uh, insightful moments from, from the customers is going to be helpful too. So I'd definitely do it.
Yana Welinder: Um, and then if you absolutely can't collaborate with, with your sales team, try to, um, try to get around them, right. Sign up for industry conferences, uh, or, uh, you know, where you can kind of just talk with folks informally. Um, reach out to friends of friends or friends of friends on LinkedIn. Um, there's, it's gonna feel awkward to be reaching out to random people, but ultimately you cannot, you can get really useful information out of that and, and, uh, essentially just creating a better network for yourself. So usually it's just the one, one situation. Okay. So I have this picture on the slide because trying to figure out who to talk with or talking or doing B2B customer research oftentimes feels like playing with nesting dolls, right? You kind of your talk with one customer only to discover halfway through the conversation that you really need to talk with whole different part of the org.
Yana Welinder: And it's really not that you could have just skipped the first person and just talk with the second person. There's so many different stakeholders, but you need to satisfy it that you really want to talk with, with all those people. Um, so obviously the most important person to talk with and get get requirements for from is the user. Um, if you're not satisfying the user, the, because the company's never gonna sign up and pay, uh, and they're absolutely never gonna renew. Right. Uh, because by then they will have tried it out and, and realize that the product doesn't really do what what they wanted to do. Um, and then, um, there's a few other interesting stakeholders and B2B, right? So there's the, there's the buyer, there's the contract approvers. Legal and finance will have lots of interesting, uh, requirements that you need to satisfy as well.
Yana Welinder: Um, it teams will have a lot of requirements on SAS, um, uh, B2B offerings. So you'll need to make sure you get those folks checking off their checklists without creating sort of a Frankenstein's monster product. Um, so that's going to be important. Um, and you really just want to figure out what is the list of different stakeholders I want to talk with in this particular company and give that list over to sales. Try to set you up with all those people, uh, and that can be your starting point. And then don't get frustrated that you get kicked around, uh, in the company and get to talk with more people. Usually just a good thing.
Yana Welinder: Um, so there's a ton of preparation that goes into a customer interview. It's not just kind of getting, getting on the phone and in, in chatting, uh, though sometimes if may feel like that's what you're actually doing when you're in the moment. Um, so the things you want to look into us, uh, obviously like who's your ideal customer? So you know who to ask questions to you, uh, what are, what's the general problem that you or your product can be solving for them? Like what's, what substance do you wanna talk about? Um, researching market trends, just to understand where your product fits into the market is going to be important. And then, uh, your competitors, right? So it's not just gonna be who is specifically competing against your company, but what are some of the competing types of solutions out there? So like if I, if I were a pm for a Skype, I wouldn't just care about, uh, Google hangouts.
Yana Welinder: I would also care about just the use of the telephone. Like what are all the difference us around that. Um, and I would want to kind of think about that as much as possible before I go into an interview. Um, and then you want to learn about the customer, right? Like what do they do? Uh, what products, um, do they have? What things have they released recently? Um, you want to learn about their competitors? Um, learning about their competitors. Uh, for me, usually I know what my customers looking to do, what's on their roadmap, uh, by looking at what their competitors have in their product. So if I see that certain type of customers have a particular type of Iot integrations and their products, I know to ask about those kinds of integrations to, to the the customer I'm talking with. Um, and uh, and then you want to use their products as much as possible, um, just to kind of inform your questions and be able to ask better questions and, and have some context for that.
Yana Welinder: Um, I have a ton of IoT devices at home just to be able to try out all the different things. So if you look at my, um, uh, my phone, I've downloaded a crazy amount of like apps that I, I'll never use in the future, but I'll, I'll, I'll, I used in preparation for customer interviews as kind of helpful thing to do, both in terms of being able to ask better questions, but also to signal to the customer that you really care. Um, you're really care about making them successful, um, that makes them open up and tell you much better information in the interviews. That's sort of like more of a psychology track I guess. Um, and then once you have all of that information, you want to form personas, um, based off of both kind of the company and what they do and also the individual you're going to be talking with, uh, their role at the company.
Yana Welinder: And with those personas you can usually formulate better questions. So I try to have a combination of super general questions like, uh, what does your normal day look like? Um, and then very specific questions like how many, uh, browser tabs can you handle before you, um, before you go crazy. Right? Um, and then I tried to write down all the questions that I can think of, uh, in this super long list of questions as you can probably tell from my weird Swedish accent. Um, and from San Francisco at, no, I've, I've lived in San Francisco for awhile, but, um, English is not my first language. So I tried to write out every question that I have so that I can make sure I can get it right. But I think that actually applies even beyond that. That for, for, for native speakers, you want to write out every question and make sure that you think through what are the possible ways the customer could interpret that question.
Yana Welinder: And right. There's, there's a lot of different ways, uh, that could go. Um, and then I prepare a super long list of questions and uh, and then I'm prepared to just put that list of questions to the side and not use it at all if the conversation goes in a different direction, which it surprisingly often does in B2B customer research because you have seals in the room and they will ask questions. Yeah. And they will try to pitch in. There's gonna be a whole lot of that going on in, you're not gonna have as much control over the conversation as you would in the, in the, in the kind of traditional customer customer interview. So be prepared to do that. I find that it's helpful to just grip questions very tactically by topics so that when the conversation does go in a different direction, I can sort of Skim my topics and figure out is there anything I can kind of still dive into, uh, based off of the things that, uh, that we've just discussed with the customer.
Yana Welinder: So ideally, I want to bring no one. I want to go to a customer interview and have a one on one conversation. Um, if I bring folks from, from, uh, from my company, the customer will want to bring someone from their company and we'll be in the group setting, uh, and everyone's gonna see things to impress each other, which is not at all what I want. I want the customers to be as honest as possible and as open as possible in telling me everything about their needs. Right. Um, so ideally you, you really wanna have, if you could, you'd have a one on one conversation. Having said that, seals will often want to join because they put you in touch with this person. They have the responsibility of tracking all the things the customer's gonna ask you about in follow up on those conversations. So that kind of makes sense.
Yana Welinder: Um, and you don't want to exclude them from the conversation. You really want to make sure that they feel comfortable about what you're doing. Um, but it usually, if you have a good relationship with seals, it usually shouldn't be too disruptive or it has in my experience, hasn't been too disruptive. Um, unless they kind of steal the show and don't let you ask any questions, which does also happen. Um, so the other type of person you may want to bring, uh, to, uh, to the conversation is, um, sometimes there's going to be folks who have particular types of expertise that you may want to bring. Um, I will often bring where one of my engineers, um, when we're talking about kind of some of the future APIs and the customer is represented by one of their engineers. Um, it's, it's just helpful to have an API engineer in the room for that kind of conversation.
Yana Welinder: Um, interestingly, I found that, uh, seals, uh, are, uh, somewhat uncomfortable when pm spring engineers or designers, um, in certain contexts, there's the stereotype that's, they may ask less that can, they can be less tactful in how they ask questions and may kind of expose you or your gaps are more. Um, obviously it depends on the person, depends on their stereotypes. There's just some of the most thoughtful, uh, engineers out there who will just be able to maneuver a conversation in an amazing way, but just be mindful of who you bring in, try to keep, uh, the participants too and minimal. Okay. So the kind of questions that I wanna ask, uh, I want to uncover as much as, as much detail as possible, um, about how the customer will use my product or a equivalent solution and, uh, really get the full picture. So that I can understand their needs.
Yana Welinder: Um, if my product were a tool that they would use in their day to day, I'd want to ask as much as possible about just what happens during their day. Um, if it's a product which is [inaudible], which is the situation make is, um, where they're going to be incorporating it in their product. I want to ask about when they've previously done that. So, um, like what were the different steps for creating an integration? Uh, what were some of the pinpoints during each step? How long did each step take? A, why did he do things in certain ways? That kind of things, um, may seem silly, but I find it helpful to ask the same question multiple times in different ways. Um, it just forces the customer to think about what they just heard in a different way. And they may kind of essentially answer the question the same way, but add a little bit more detail or a little bit more nuance and you kind of get a fuller picture that way.
Yana Welinder: Um, and then one kind of somewhat controversial thing is site visits. A lot of PMC that you absolutely have to go and visit the customer and watch them in action. And that's true for a lot of products. Um, so if, if your product is a tool that they're gonna be using in their day to day, you really want to go there, sit next to them, see what their work set up is, what hardware and software they use, kind of how they move their arms. I don't know. [inaudible] you want to do all those things, um, you don't necessarily need that for a lot of other products where they're not, they're not actually going to be using it then their, their, their day to day. So traveling to, um, exotic destinations may kind of just distract you from talking with more customers. So just something to keep in mind.
Yana Welinder: Um, so after asking a ton of those kinds of open ended questions where I just tried to get the customer telling me as much as possible, um, I'll try asking a different type of question, which is usually kind of assumptions that I've formulated about their needs based off of the personas that I prepared ahead of time or the kind of the information that they've given me up to that point. Um, sometimes that, that point in questioning comes really early. If customers aren't great at responding to open ended questions, so you can be kind of jumped to this stuff much, much faster in the process. And then, uh, and then usually the reaction will be to confirm or deny my assumptions. Right. Um, I found that when my assumptions are slightly wrong, that's when I get the most useful information out of, out of the interview, uh, because customers will, we'll try to course correct me and the process of course correcting me, they will see a ton of things about kind of, um, details about how the and they're going to be using my product. Um, and then the other thing is that's when I am slightly wrong. I'm not so wrong that they feel like this is hopeless. I don't want to talk with this person. Right. But just like you're wrong. They feel like they're more knowledgeable than me and they want to share their wisdom. And so kind of just like from a natural kind of a human nature perspective, you'd really just want to teach and that's, that's the mode they get into. And then they, they, they can give you more useful information as well.
Yana Welinder: Um, I usually try to save my pitch or am I kind of a product demo, uh, for a followup call. Uh, sales or the customer, uh, may get really annoyed with me that I'm wasting the customer's patience, not pitching the product. Um, so it helps to tell them ahead of time what I'm doing. Uh, I'm going to share the pitch in the followup call. Don't worry, it's coming. I now we need to talk about kind of what, what, what your needs are. Um, I find that, um, I get a little bit more slack if I promise to incorporate the information I uncover during the first call in my demo, uh, in the followup call and then obviously have to follow through on that promise. Um, sometimes though the product you're talking about is so abstract that it's really difficult to get useful information without some sort of demo to the customer.
Yana Welinder: Some are very [inaudible] are, are that way. Um, and so in those use case, in those situations, I find it useful to just formulate, uh, use cases that, uh, that the customers can interact with. Usually those use cases work in the same way as kind of presenting assumptions. Um, customers can sort of confirm or deny what you have on the slide, uh, and kind of, uh, slightly change how the flow would go or some of those things. Um, the, it also helps them to uncover some of the kind of very specific needs that they have. Like, you know, this part of the flow needs to be logged for it. That part of the flow absolutely cannot because we're in Germany and there's, you know, no, that's a no go or something. Right? Um, uh, or this part of the product needs to be wildly abled. I don't care about that part. You can, you can, you know, brand with, with your brand, uh, all you want. So those kind of specifics are difficult to get to without actually having a demo. And then, um, sometimes I do all these things, uh, and talk with a ton of customers and don't really get to really useful information. And I talk to this one customer, it's just amazing. They tell me everything, um, everything about their needs, about their pain points. I get super clear priorities. It is so incredibly tempting to take all of those priorities and run with them. Don't do it. Don't build for just one customer. Right? So, um, if you get that really strong signal from, from, from one customer, uh, you really want to validate that with, with multiple customers. So, usually what I try to do is to set up interviews after that point and use the strong signal or the information I got from one customer to set up assumptions or kind of use cases that I can present to more customers than they can pick those apart and give me a much more nuanced, uh, view of what, what we need to build. Um, I also found that it's actually easier to formulate better surveys after you've spoken with a ton of customers because you know how customers are likely to kind of, uh, misunderstand questions or take, uh, questions from a certain direction. So you can, you can formulate something that will get you to the core much easier based of, of interviews you've already had.
Yana Welinder: And then we want to feel like you never want to feel like, uh, you know everything about one customer's needs. Right? Uh, I think it's kind of tempting to have a list of customers and then talk with each of them and then just cross them off the list and feel like, oh, I know what they, what they need. Uh, but actually customers' needs will evolve over time, um, as the, you don't have to do different tasks or are introduced to new technologies and that over time can be literally like next week. Right. Um, so, uh, I usually tried to talk with customers periodically. Um, it helps to also be able to tell customers, um, how I, how the things I've learned from a previous conversation got incorporated into the product. Um, that makes them feel like they're being listened to the conversation that's worthwhile. It's not just the waste of their time.
Yana Welinder: Um, so that's something to, uh, to, to try to do if possible. Okay. So I did say 10 simple steps. Maybe not that simple, but hopefully if all you came away with, uh, from today, I think these are kind of the most important things I think. Um, so you want to collaborate with, uh, with your gatekeepers, with your customer facing folks. You want to, um, kind of both in terms of having them connect you with the right people, but also set the conversation up for success. Um, take every opportunity to talk with customers. Uh, and if you can't do it through the folks in your org, just cleverly get around them and, and try to do it anyway. Um, when I do ask open ended questions as possible, ask them in a tactful way and listen and in, in same your pitch for another day. This is not about pitching your product though. That base is oftentimes what customers are looking for you to do. Um, and then you want to talk with customers periodically. So this is not just the one time thing. You want to see what their latest needs are over time. Cool. So I think I am right on time so I'm not sure if we have time for questions. Um, but if we do, let's ask them that.
Speaker 2: Yeah. Um, I wonder if this is Megan Good. Cool. So thanks by the way, great talk. Um, on your point number three, nesting dolls. Uh, have you, like, I've found that to be extremely important as well. Like sometimes talking to sort of a manager on the other side can give a misleading signal, but talking to the end user is just a totally different game. Yeah. What are some like tactics you've seen or you've used yourself that like actually lets you get past the fear of wasting the managers employee's time and getting that really good feedback?
Yana Welinder: Yeah, so I mean, uh, I, a lot of in, in my experience, I don't feel like I'm, there's, uh, there's ever anyone who has, who you shouldn't talk with if you're, if you're, if someone is sort of the, the, the, the first nesting doll in the conversation, you can actually uncover a lot about needs even if there is kind of a report who will be direct report, who would be interacting with, uh, with the product and much more directly. So I would say talk with all of them, uh, without kind of skipping, skipping the manager if there's absolutely nothing to learn from the manager. I think just being open about that to the manager is, is fine. Uh, like you kind of explain, explain it to them from the perspective of, you know, you want your employees to be as productive as possible. You know, I, I want to talk with the people who will be using the product the most. Um, who would that be? Is there someone, is there someone in your, on your team who is a particularly good, like, uh, providing really good feedback, get them to be kind of partnering up with you and feel like they're being productive too.
Speaker 2: Yeah. Ah, yes. Do you want to, I wonder if we want to send around the microphone. Could you just talk a little bit about how to do research on your direct competitor other than going through the customer out?
Yana Welinder: Yeah. So your direct competitor or the customer's competitor are your direct competitor, your a direct competitor. Um, so you're, I mean, so there's, there's a ton of research you can do just online in terms of figuring out what are, what are you offering now and what are they talking about, uh, in like conference settings and things like that. Um, but I think, uh, talking with the customer is going to be really important because Muslims, oftentimes they will have, someone will have pitched things to them. Um, and they will have, like, even if they can't tell you this cup company does x thing, the way they're thinking about your product is going to be biased by what they've heard. So a lot of the things they will be asking about like, can your product do x? Uh, right. Uh, we'll, we'll tell you a lot about what they've heard from your competitors. Other questions?
Speaker 2: Yes. How does your team aid all this interview, research and then [inaudible] product?
Yana Welinder: Yeah, so a good question. So the question is how, how do you gather all of this information and actually get it to form the product? Uh, it's, um, I think it's always a tricky piece to figure out how much time do you want to spend? Like documenting things and, and uh, and get it in a way that's actually like actionable. Right. Um, and, uh, I found that I don't want to just like take my notes and give them to the engineers. That's always a bad idea. Um, but, um, but try to like get the most important piece out of each conversation and see is there a way I can inform the engineers? I will like literally get on slack after each interview and like write three bullet points. One was like my most important things, but I think my team really needs to know, um, in that, and then in a, later on you can kind of think about how do I consolidate all of these different things across, across the different words, the different customers that I've talked with. Um, but just in that moment in time, you can be like, oh, this customer asked the same thing as these other three customers that I just talked with. You want to know that other questions?
Yana Welinder: Cool. Um, I am going to be around during the break, so feel free to reach out to me then as well. Um, just wanted to mention that we're hiring. Um, so if you or someone you know, uh, is, uh, interested in IFTTT, uh, ping me either via email or on Twitter. Thank you.
Keep me posted on Empower 2019.