Insights vs Observations


We discuss the difference between an Insight and an Observation with regard to Research. The distinction might sound semantic, but in practice it’s good to be clear.


Matt: Make that your new opening clip.

Matthew: and I'm definitely gonna leave in the part were we're talking about the opening clip in the opening clip. Including this part right here. Cue music.

Matt: Hey!

Matthew: Hey!

Matt: And we're back!

Matthew: And we're back!

Matt: We're in, we're here, welcome

Matthew: Season two, episode two. Insights versus observations.

Matt: Dun, dun, dun.

Matthew: It's gonna be... We're gonna do a little bit of... What's it called, Thunderdome for ideas? I don't know.

Matt: Insights versus observations. You're gonna put me on the spot.

Matthew: I am gonna put you on the spot. Because I have a very specific question about this that I've been, you know, ever since you brought this up as something to talk about on our show, I've been wanting to ask you this very specific question. What distinction, what line are you drawing between, what is an insight, what is an observation, and if you would be so kind, that little of context as to when these things come into play.

Matt: Okay.

Matthew: Go, I'll be back.

Matt: And I don't get any time-- I have no prepared words for this.

Matthew: Okay.

Matt: Where do I draw the line? I draw the line between an observation, I draw the line where an observation is something you observe, so when you're doing research, you gain a literal observation between what somebody says, what you see in an environment, what you read in an artifact from other documentation, versus an insight into how you as a researcher interpret that observation into more of a, what we call scientifically nuggets... where you can look at something and say, okay, this participant or these participants said X, Y, Z. This is a problem or a pain point, so the problem or pain point would be the observation. The insight would be, these people are not able to do their job because of these other external factors that are impacting them, that they might not even know to express as part of their pain points. So that's kind of a vague, but specific example.

Matthew: And the point why research is so valuable in having someone who is a trained researcher to be doing the research, so that they can make those observations and then have those insights the people themselves wouldn't necessarily be able to express on their own.

Matt: Right, or the business be able to recognize. You have the client, whoever's paying you to do the research. And to be fair, observations, I don't wanna say they're easy, but they're easier, because you're just watching. Sometimes insights, personally speaking, are almost impossible to figure out without enough information. Like, you make guesses, it's almost like looking at cause and effect. This is the end result of something, here's what's observed, what's the cause? And sometimes we're not able to figure that out based within scope, parameters, other factors of the project, so sometimes you're making educated guesses. Sometimes you do know for sure or reasonably sure. When we're talking about communicating research to a client, observations are fantastic, and that's great, having that real, actual data that you can go back to and say, this is what I saw, this is what I heard. But then showing that next level down, the insights, is really where we earn our keep.

Matthew: Yeah, this is what your customers think of your product, for good or ill, and this is literally the words that they said, this is literally the things that they do. And then the next question, of course, in our minds and certainly from our clients' minds, is, well, what do I do with this? Like, what do I do about that? What is the thing that ties these all together that's an actionable thing? And that's where the insights come in.

Matt: And the other insight is, usually they'll ask why. Like, why are you observing this? And can we dig in to get some rationale? Some reason of why, behind it.

Matthew: Just the basic constraint of the kind of work we do is typically, no one's hiring us to go sit in a park and observe people and become inspired with some research moment where we're like, "A-ha! "Roller blades for everybody!" You know, or moving sidewalks everywhere, instead of regular sidewalks!

Matt: Segways for everyone.

Matthew: Segways for everyone! Segways for... By and large, no one's gonna hire us to be 100% generative. Go to Thailand and just be inspired, and tell us how we should change our business or whatever. There's always some hypothesis that, something is wrong, or that there have been observations made, sometimes, and this is where we can get into quantitative versus qualitative research, where there might be someone who's involved with customer experience type stuff, where they're getting a bunch of data. Not to pick on this, but it's the easy thing that comes to mind, is shopping cart abandonment, or whatever. They know what is happening, but they don't necessarily know why it's happening, so they've observed that this statistically significant event is occurring. And now you need to have some insight into that. For me, I don't think you can get that insight just by looking at the observable data when you're talking straight-up quantitative.

Mat: It might...

Matthew: You can have, going back to your educated guess perspective, absolutely. But you still--

Matt: Right. You see what they're doing, they're stopping, they're navigating, to use your example, they're leaving the website.

Matthew: Right.

Matt: But to get to that, that why, or the actual insight is why this happens, is harder, and usually that's where you get into the, again to your point, more qualitative data and research methods of talking to people, not just looking at data.

Matthew: And not just looking at data, and then saying, okay, we'll just A/B test some new things and see how well those work. If you have the means, if you're a big enough company, and you can iterate that quickly, maybe that is a good approach, but not just because I am a qualitative person by training. I think always pairing qualitative and quantitative together is, um, is something that leads to those insights that are actually well-informed and going to be of high value to the client, or to your organization if you're an internal person. What brought this up? What was on your mind that--

Matt: So this has been bouncing around in my head for a while, like months, maybe longer, as a potential talk topic. Something you said in the last podcast triggered it, and I don't remember what.

Matthew: It must have been a really good thing, though.

Matt: No, you said something, and I'm like, "Oh, yeah, that!" And I just literally went over and typed it into the deck, because I put that Airtable, because I didn't want to forget again. But yeah, I don't know what triggered it originally, I feel like maybe I had to explain it to someone, and I didn't have a really good answer off the top of my head. And it was just in my head, it was like, oh, this might be a good topic for a talk, a conference talk or something. I don't know if there's enough meat to it, but it's just one of those things, almost like a workshop thing, like, all right, let's all practice writing, what's an observation, what's an insight? And maybe it's obvious, maybe it's not even a good topic, but in my head, I thought maybe there's something to it.

Matthew: I think there is something to it, in the sense that, I feel like I've, I've been adjacent to and involved in conversations with people who are, to some level, decision makers within an organization. Whether I've been the internal person doing some research or related work, or it's been for a client, where it's not so much an observation, then an insight, it's an observation, and then, what's the first thing that comes to mind to me, as that individual decision maker, that is in scope and in budget with what I have to work with? Basically, what's the cheapest way I can fix this fastest?

Matt: Right.

Matthew: Hey, how about a popup? Or I feel like, every time I'd be on some site, if my cursor went outside of the viewport of the browser, I would get a popup, and that would say, "Hey, stay here!"

Matt: Right, "don't leave!"

Matthew: Right?

Matt: Yup.

Matthew: That's so annoying, and it probably works, which is why they do it, but...

Matt: No, but they don't do it anymore, so...

Matthew: But they don't do it any-- Right, I haven't seen it much anymore, but--

Matt: The other thing I haven't seen lately are those pops up that says, "Need help?"

Matthew: I still get the "subscribe to our newsletter" one when I'm trying to click on something else.

Matt: Right.

Matthew: But...

Matt: But those ForeSee ads, like, "Oh, do a quick survey!"

Matthew: Right, right. I feel like those are the kinds of solutions that you get to that, because I've observed a behavior or I've observed an outcome that may not even be reflective of the larger experience that most of my customers are having, but I'm then going to come up with an idea that... is shippable this week, to solve that problem. Okay, maybe you get lucky and it solves the problem for the small number of people, but maybe it creates a bigger problem for everybody else, and I think that that's why working with someone who is a researcher, to help you go from observation to actual insight, you know, as our good friends across the Pacific pond say, the sense-making. Having someone help you from that observation to insight moment, as opposed to just the educated guess, I think is really important. I think it's important to make that distinction. So it's, potentially as practitioners who are watching the show right now, you may have thought, "Well, of course. "Observation, insight, that makes total sense." But looking back on all my clients that I've had, all the jobs that I've had where I've been a researcher, I would say it's 50/50 as to whether people, that really resonates with people. I think there's a lot of people who, especially in decision-making positions, or who are under a lot of pressure to deliver, and it almost doesn't matter what they're delivering, as long as they're seen as delivering.

Matt: Right.

Matthew: And I think that's how you get to really half-assed solutions or poorly thought through solutions that solve for, you know, a small percentage of people who have the problem.

Matt: Yeah, anyway. All right, catch you tomorrow, go team.

Matthew: All right, go team.

Related Post