#062: When and Where Does the Role of the Analyst End?

It seems like a simple couple of questions:

  1. When and where does the analyst’s role start?
  2. When and where does the analyst’s role end?

And, do the answers to either of these questions change based on the type of organization you’re in (in-house versus agency)? As it turns out, Michael and Tim largely agree on the answers to these questions…but their agreement is pretty expansive, so this could be the episode that infuriates you, dear listener! Give it a listen, and be prepared to shake your fist at your earbuds!

People, Places, and Things in This Episode


Episode Transcript


00:05 Announcer: Welcome to the Digital Analytics Power Hour. Tim, Michael and the occasional guest discussing digital analytics issues of the day. Find them on Facebook at facebook.com/analyticshour and their website analyticshour.io. And now, the Digital Analytics Power Hour.


00:28 Michael Helbling: Hi, everyone. Welcome to the Digital Analytics Power Hour. This is Episode 62. Hey Tim, I’ve got a question for you. Do you vape?

00:41 Tim Wilson: [laughter] I do not.

00:45 MH: Well then, that’s fine. If you are a smoker though, you should check it out. But seriously, there is…

00:51 TW: Maybe I’ll start smoking then. Ask me again in a couple of weeks.

00:55 MH: Yeah. No. But there is a question that I do want your help to answer and that is, where exactly does the job of the analyst start and end? This came to us via Justin Goodman, also known as Electricmice on the Measure Slack and he proposed that. So thank you Justin. And what is it like? Do we just chuck insights over the walls of the business and then go back to our shack on the data lake? And, or are we chasing waterfall development methodologies and trying to make them more Agile? This is the show where we’re gonna talk about that.


01:38 MH: Tim, you have been called…

01:42 TW: Mixed metaphors and butchered cliches.

01:44 MH: Yeah. Now this is where it’s gotta be. Tim Wilson, you are the co-host and notable industry people have called you the quintessential analyst.

01:56 TW: Did they ever put that in writing and who were these people?

02:00 MH: It was just me in Episode 60 but…

02:04 TW: Okay. [laughter]

02:04 MH: I think people agree and so I am, of course, Michael Helbling, the non-quintessential analyst. So we kinda have both of those bases covered. So let’s talk about the beginning and the end. So Tim, where do we start? Where do we start as analysts?

02:27 TW: I think we certainly start way before the data. I mean, I think that’s… I’ve certainly seen cases where analysts get, on the one hand say, “We need to be involved really, really early” and then they get pulled in really, really early and they came out of the meeting saying, “That was a waste of my time. We’re nowhere near needing any data yet.” So I think it’s, it is almost… Cliche is probably not the right word but analysts, we like to pull our hair, gnash our teeth about not being involved in the planning stages of things. I actually think we should be involved there, it’s just we haven’t done a very good job of defining what our role is when we’re involved there. Because a lot of the discussions are much more strategy and vision and ideation discussions and we can’t just say, “Yeah, but where’s the data? I want to do something with the data.” But I think we should start there.

03:26 MH: Well, and it’s always… It always feels, from my non-quintessential point of view, it feels unfulfilling to get asked for data as opposed to asked to help solve a business problem.

03:41 TW: Yeah. And I guess the… But the challenge is when somebody comes and even if they frame a business problem, a lot of times it feels like that’s… If the first you hear is, “Hey, a bunch of us were sitting around chatting and we’ve been pulling our hair out on this for the last three months,” and hey, somebody finally said, “Why don’t you go ask Tim?” And they arrive to the… It’s very, it’s getting… It can feel like you’re thrown into the deep end. You haven’t been on that journey with the business to say, what are we really trying to do? We’re trying to ask good probing questions but they’re at their wit’s end by the time they got you involved.

04:21 MH: So at a strategic level, that goes all the way up to corporate goal setting and planning. So what is the big plan for the year? And then as the analyst, you cascade down to, “Okay, what am I doing in my plan that supports that bigger objective?” And I wonder, is that a common thing for companies or is that something that’s not that common?

04:49 TW: I feel like there is just a woeful lack of clear capture and articulation of what that overall plan is, how we’re thinking about our customers. Just in the last couple of weeks I was having a conversation and this, I guess this is a question of like, “Is this the analyst role?” ‘Cause I was talking to an analyst who’s a client at a very large company and as we were talking I was like, “You know, it almost seems like, fundamentally you guys are thinking about your customers as in two major groups and one of them has sort of this path that you want them to follow and the others had this other.” And we were just kinda talking, trying to work it out between the two of us and it wasn’t anything she had seen. It was just kind of both of us picking up from the questions we were hearing and the way that people were trying to look at the data and I proposed at the time, I said, “Well, maybe that’s what we should do is we should literally; one, map that out and go and check and see if it’s right.” Which I think is, I’ve never seen an analyst get shot down that they said, “Hey, I was just drawing on my white board trying to understand how we see the business fitting together.”

06:08 TW: There’s nothing about data in that. So I don’t know how much is this is what I like to do. I like to have the one clear understandable diagram that says, this seems to be the way we’re thinking about this, is that because I wanna do that? Is that because I think the analyst should do it. I’ve certainly tried to coach analysts to do that and watched many of them either resist or just struggle, ’cause they don’t seem to see the value of it, but I think there’s enormous value. Somebody’s gotta do that and to me, I see a lot of companies maybe there’s a high level strategy by the time it gets down to a certain maybe marketing area, that whole cascading down and view of the big picture has gotten kinda lost and now we’re just executing campaigns or executing projects. So I don’t know, I feel like that is valuable. When we had Christopher Berry on, that was a little bit of what we were talking about with systems thinking I think, wasn’t it?

07:13 MH: Yeah. Well, sure and I think that only helps analysis when there’s context both in terms of problem pursuit or like what should I be analyzing and in terms of inside delivery and framing.

07:30 TW: Yeah. And I think even in the third which, it’s related to both of those, is that fostering the dialogue with the people who are the actual stakeholders to say, I think this is what’s going on, and if they say, “Well, not quite,” it’s really this other thing or once you have them saying, “Hey, we’re on the same page and we’ve got that framing. We’re aligned on the framing.” Now I’ve got a lot more confidence when I go back and pull data that I come back and deliver it and it’s within that framing. But I feel like, how much is that? This is where we play the gray hair card. How much of that is a experience, more senior, more background with marketing able to play that role? Is that too much to ask of, I’ll say it, typical analyst? I think it’s clearly not too much to ask of somebody who is a senior seasoned analyst, but at what point in evolution, I feel like after a year, you should be trying to do it because even if you’re dead wrong with the framing, you’re; one, learning yourself because when you’re playing it back, the person you’re working with is saying, “No that’s not right at all” and that means you’re rapidly accelerating your ability to do that and you’re not hurting yourself.

09:00 MH: And you’re finding out through that process if there is someone in the organization willing to partner with you, mentor you in thinking about the business in the right context. Which is really helpful in understanding what your future at that company could be like?

09:17 TW: Well, and I guess I’d even say if you don’t do it, if you don’t map it out, you’re still gonna be making decisions when you’re doing the analysis that are gonna be making those assumptions, like, I’m now marching down the path of saying, “If you aren’t making that attempt consciously and deliberately, then you are very very potentially gonna be doing analysis that’s gonna miss the mark.” Because you’re doing the analysis with the wrong framing because you’re making assumptions, even if you haven’t consciously thought about that framing. Oh boy, feeling pretty abstract right now.

09:53 MH: But I think it’s good to kind of build the structure and then we can try to move towards more practical application and usage. But it’s… I think we’re both in agreement on this one and that means as an analyst getting started, the beginning is as early into the processes as you can possibly get. I mean, even maybe as early as ideation in terms of any kind of initiative or…

10:19 TW: Which seems like it’s easier, it’s a smoother ask or a smoother thing to do in-house because you don’t necessarily have the time pressure, “Oh this is another body that we’re having to bill for.”

10:35 MH: Right.

10:35 TW: Now you still have to have a management support and whatever that your… And organizational support to get to be in the room. But actually I think Justin, ’cause he’s been in a couple of agencies and I think he was initially even asking from the agency side and I definitely do remember that being a tougher lift, both through the account team who’s trying to manage the profitability of the client saying, “We can’t show up with an army of people. We have to have our creative person. We have to have our account manager. We have to have our strategy person. We can’t show up with five people for every meeting.” And so they will push back and the client may push back as well saying, “Why is the analyst in the room?” And maybe that’s just… Well that’s why there are some aspects of being an analyst in an agency that really really suck, but do you have thoughts on the agency side from your days back doing that?

11:34 MH: Well, yeah. And I think for the agency side you get brought in at a specific juncture, unless you’re round in the corner into year two after the first year working with a specific client which is something that hopefully is happening. But then you can look to participate earlier in the process. I think you just go in and grab up any information you can get to make yourself understand the context. And a lot of times, I try to be really upfront with clients about, “Hey, in the first analysis, it may feel somewhat contextually challenged because there’s still gonna be some growing in terms of what we can bring to the table.” Now, some stuff is just really easy to analyze, so you get the Jim Cain’s favorite quote, “the low-hanging fruit.”


12:31 TW: Wait a minute. Are you speaking… I was thinking more from being on the creative agency side. Well, I guess that’s…

12:39 MH: Oh yeah, even if you’re on that side of it. So maybe, your agency is engaged to create a brand new experience or redo the website that the client’s using, whatever. I guess in those cases, yeah, I’m always trying to push to at least include somebody from analytics all the way back in the design phase. Because you have to hear what the motivation, objectives, plan, goals for what this is all doing from the client. You’ve gotta hear that stuff because that’s how you design the measurement and know what to go look at from a metrics perspective anyway.

13:27 TW: Yeah. And I think I’d agree. There’s… I got to the point where I was often in the pitch and that was actually useful because the pitch, you’d have conversations and you’d figure out what the client was lighting up on. But I was always pushing to say, “We need to be involved early, we need to be scoping in time to be participating, we need to be doing a measurement plan.” I definitely remember coming out and sometimes doing measurement plans as an analyst at a creative agency that was actually putting in those diagrams of saying, “This measurement plan is based on our understanding that this is the place that your business is.” And then, because we’ve come in with the strategy that we’re proposing, I’ve sometimes felt like I was trying to be the, “Let me actually translate from this 74-slide strategy deck that is death by PowerPoint to try to… No one ever really framed it, what are we trying to do here?” Now I felt like I was being very not what was expected. And it really varied. Even that, there were the account teams that were like, “This is awesome. You’re really helping.” And there were the account team saying, “Get the fuck back in your little cubicle, shut up and give me the Google Analytics tags.”

14:54 MH: Yeah. Well, right. And I think that just goes to show you there’s that spectrum everywhere of stakeholders, people you work with and things like that. And you…

15:06 TW: Except there’s two layers, ’cause that you could say, say 50%. 50% were in the, “What the hell are you doing?” The same thing happens on clients. 50% of them are gonna be, “What the hell are you doing?” And in order for it to work on an agency side, both of those have to say, “Yes, we actually want you here.” So, it’s only 25%.

15:28 MH: Exactly. Aha. I was about to come out with that number.

15:32 TW: That’s good.

15:33 MH: I second guessed myself ’cause I’m not good at math. No, I’m just kidding. [laughter]

15:37 TW: I wrote a little R script that told me I was…

15:39 MH: There you go, perfect. So, yeah. And on the agency side, I think it’s tougher. You mentioned the constraint of how you allocate time and those kinds of things. And so that’s part of that. Obviously, when you’re doing a full service thing, you can have somebody who’s 100% committed to that. And so that’s nice when you have a structure like that where someone can really focus. And then, I think you can still build that same level of insight and quality over time, but that’s probably not 100% of the kinds of engagements that people do.

16:18 TW: But I think that’s the other thing. When you were bringing up, “The longer you work, the more you’re understanding the business.” Now having had cases where I’ve had multiple clients for multiple years, it does seem like there’s sort of… You can easily slip into the comfortable, “I know what they ask for, I know how to give them the stuff that meets what they’re asking for.” And kind of forgetting that, “Hey, I should be getting more familiar with the business. I should be more able to elevate my thinking a little bit further and have a better understanding of the business to actually do that framing.” ‘Cause it can feel a little awkward to say, “Well, I’ve been working with them for two years, why am I framing this stuff for them now?” Or, “Why am I asking these questions?” Because I’m now suddenly confused about, “Wait a minute. Are we doubling down on our emerging products or are we really going after our enterprise clients?” Like which one of these is it?

17:30 TW: But I’ve also seen that, “Hey, asking the big question.” As long as it’s not like, “I’m not gonna do anything until we’ve answered this big impossible question.” You can always do that whether you’ve been doing something for six months or for five years. So, we’re agreed. As early as possible, put on your strategic thinking hat, ask smart questions. And those questions need to be questions you genuinely want the answer to because you can see or understand how that will inform what types of analysis you’re doing and how you’re framing what you find. Right?

18:10 MH: Yeah. And the key to staying invited to those early sessions is be careful what you say. [chuckle]

18:20 TW: But don’t just be a bump on a log. Right?

18:23 MH: It’s fair. You gotta contribute but you gotta… It’s a delicate thing ’cause it’s sort of like, when you don’t have anything to… You don’t have data to bring “yet” but you certainly have insights and things you’ve observed and if you’ve been in the industry for a while or done things you know some things that will or won’t work or are good or bad usability or things like that, so you’ve got something to say. That’s sort of one where you just gotta know yourself a little bit I think.

18:56 TW: Well, and I think it’s often fine to not say anything but listen like crazy and then follow up. Like, you’re listening to this, nobody knows this, what the answer is here, to go off and say, “I spent half an hour looking at… ” I realized this other campaign that was maybe from three years ago did have some similarities in this one area so hey, I went back and looked at this and then whether it’s the next meeting of that group or whether it’s just a quick note blasted out to him and say, “Hey, there was a lot of discussion around this and occurred to me I could look here or here.”

19:35 TW: Which is the sort of thing that you watch on the measure slack or in other maybe more private forums where the question will get thrown out, “Hey, does anybody ever seen or done X or Y, tapping into your network?” And saying, “We have this weird thing we’re doing with Pinterest.” Or “Has anybody done this or seen this or did they learn anything?” And so, I think you didn’t leave the meeting with any action items, nobody looked at you and said, “We’re counting on you to go deliver something.” But if you actually follow up after and say, “Nobody asked for this, but I realized there was something that I could help move the ball forward in the planning stage.” That rockets you up to the credibility they’re like, “Oh yeah, Tim was a… Didn’t say a word,” and I walked out of the meeting saying, “Why was Tim there again?” And then, the next day, Tim shot a note and it was super helpful.

20:37 MH: See, that’s quintessential right there. Actually, what it is, it’s proactive and I think that’s just a good look on anybody. So now I love that tip, that’s a really good tip, is ’cause, maybe you don’t have anything you’re able to say right then but if people see you having put thought and action behind that in contributing like that, that’s pretty, that’s a good idea. So, that’s interesting okay. So, this one frankly is super boring, we both agree. So let’s get to maybe the more interesting or the more nuanced which is, where do we stop? Where do we stop and honestly this one has probably a matrixed answer because some analysts they get done pulling some data out of Google Analytics and then they jump into AdWords to modify the bids ’cause they’re wearing lots of hats. And so for them, the job of an analyst is also the job of the paid search marketer and the email marketer too, maybe or some combination of things that they’re doing as an analyst. So that’s… And as you grow in the organization I think maybe there’s more and more specialization and more kind of focused stuff. But, I don’t know do we want to pick a spot? Maybe at the enterprise level where should the analyst job end.

22:04 TW: Well to me, my perspective on this is simpler but also seems like a lot of people push back. I had a client once that literally would go in and make a custom report, when I asked a question would make a custom report or multiple custom reports export them, throw them in Excel and ship them back to the requester and say, and then they would check it off as done. To me that was like, that wasn’t even debatable that how… What a piss-poor job that was…

22:36 MH: That is not doing analysis.

22:38 TW: So, even if he went through and said I had delivered, I’ve done the analysis and answered your question to me, that is still stopping short of what the analyst should be doing and that… It’s so easy to say, “I’ve sent this.” People get 50 a hundred emails a day and even if they asked for it, just the asynchronous nature of analysis, I had a business problem or a business question I asked the analyst and then I had to wait to get that back from them. And I may look at it, but I’m not in the same spot I was when I asked the question. Somebody to not be pissing away resources, somebody needs to kind of own the entire life cycle of an analysis, which goes from capturing the right question, which was what we were talking about earlier, all the way through “Did we make a decision? Did we take an action? And did that actually drive impact?”

23:49 TW: And none of that… It’s not like I can sit down and say, “From 8:00 to 9:00 tomorrow morning we are going to make a decision, take the action and measure the impact,” it’s just inherently complicated that you’ll make the decision and then at that point, plus five hours or three months before you take an action, and then at that time plus a week or three months, you might be able to quantify it. And I just don’t think it’s… It is insanely naïve if an analyst thinks that the business stakeholder is going to own each of those spots, even if they should, but we’re not living in theoretical land. I guess, clearly I have some strong opinions on this one. [chuckle]

24:40 MH: I don’t disagree with you Tim. That makes a lot of sense to me. I probably take a nuanced view as an analysis kind of reaches its point of passing into the business, of sort of proving it and evaluating the impact and those kinds of things become tasks after the action as opposed to necessarily attached to the original analysis. But I do like the idea of there being an ownership of the overarching process. That makes a lot of sense to me. And I think that has to be from within the analyst organization in some context, because I think the ownership has to live there. And the only thing I would say is in terms of where you pass that off, it’s at the point that you take your data, you synthesize it into a recommendation. Hopefully, with some kind of a forecast of what expected results will happen because of that, in some kind of business terminology… Then that’s where I see the… You’ve done it. Now, the next thing, and this is where you transition from being an analyst in a certain sense and to being an organization builder or, perish the thought, a growth hacker.

26:09 TW: Oh my God. [laughter]

26:10 MH: But I did it. I was in a meeting a few days ago, and I managed to say web trends about 10 times in one meeting. It was great.

26:19 TW: That’s outstanding.

26:25 MH: Yeah. So anyways… No, but you see what I mean, like I don’t see that as an analyst skill set as much as it has to be, because almost nobody else is doing it. But is it really the analyst’s job?

26:38 TW: I think there can be organizations that have built out their processes such that it may bubble up on the analyst’s radar to say, “Oh now it’s time to go measure the impact,” because we are clicking, it is such a high functioning organization that is so driven around a Lean or Six Sigma or whatever it is, process that we’re… Or even agile in a fully, fully, functional way, that seems like it is such a rare thing to see that if you get out of the… I can piss and moan all day about it’s not happening, but hell that’s something I can do. Maybe it does come down to kind of a racy mentality that, okay maybe I shouldn’t necessarily be accountable for measuring the impact, that should be the project manager, the marketer, whoever. But the fact is, if nobody is really holding them accountable to it, then do we want to just let the company throw money away that we’re not doing it? Or do we say, “You know what, I, as an analyst, am going to take that on.” So I think it’s fair.

27:43 MH: And I don’t disagree with that either because I am a big time take-ownership and take-initiative… It’s like you see a piece of trash on the ground, just pick it up, you just don’t walk by it because you can. That being said, I have also… That has hurt me in my career as well, especially on the agency side early on because I was doing so many things to help train people, help with business development, all these things. And then you get to a time and they’re like, “Michael, you haven’t spent enough time billing clients.” You’re like, “But I did these all other,” and they’re like, “We don’t care.” So there’s this balancing. Obviously, if you’re on the industry side, then hopefully there’s a little more understanding of that time gets spent in all these different ways. However, that’s an interesting problem and I’ve seen it go out past the analysis to the action where literally the analyst takes on the job of managing the action across the organization. And I’m not advocating for that, but it’s like well how badly do you want to see the change happen? Well, enough to go and wrangle an IT group and align everybody and go run the project yourself.

29:04 TW: I guess if we agree that it is a major value drain, if there is analysis being… Even by the time the analysis is done, if it’s a good analysis there’s been a lot of investment of multiple people’s time getting to the point where the analysis even started, much less finished. And, if it shouldn’t be the analyst job to actually drive, that should be a project manager or the product manager or the marketer to actually drive beyond that. And if that job is not happening, there is the risk that if the analyst says, “By golly I’m going to step in and I’m going to try and make sure I drive that. Well, is that actually just perpetuating a shitty business process and reducing the bandwidth to actually do analysis? And that’s the risk.

29:53 MH: Well, that is the tradeoff, right. The opportunity cost is that the analyst isn’t doing their primary function of doing more analysis at that point, ’cause they’re over here trying to get tags on a website or something like that. The only thing I’ll say though, is up to a point if you wanna see it go anywhere you might have to put on a hard hat and go do that. Because I just fundamentally believe something about analytics generally, which is as organizations see the value of what happens when they do take action, they’ll make the adjustments to start making those things, and that’s what I’ve observed over time. And so, in a certain sense, it’s like there’s some of this that’s maybe just priming the pump, it’s like, “Hmm. Nobody’s taking action, well, how far could I drive an action?” Okay, and if I show positive results and actions over at a course of period of time, and I can sit down with my boss and my boss’s boss, and other groups of people and say, “Hey listen, I wanna talk to you about a prioritization thing ’cause here’s how I’m spending my time. I think you’ll agree that we found some really good recommendations and if I’m over here making those come to life, I’m not back here finding new ones. What can we do to continue to drive the success forward?” And if you’ve had success you can point to, somebody’s gonna cough up another resource at that point.

31:21 TW: Yeah. And if you… A lot of this does go to the life cycle, the life cycle of analysis or be one of the little phrases I use where if you say, it starts with a problem that leads to the hypothesis all the way through, it’s put in place and measured impact. If you are tracking all of those… If you’re just trying to track that entire life cycle, that inherently is gonna draw you in to asking questions of, “Is this gonna get implemented? When is this gonna get implemented? Do you want me to manage getting it implemented?” But what that also gives you is… It gives you the data to say one, “How long does this stuff take?” Two, “Who are the partners that I’m working with? Where we are actually successful? What are the ones where I’m clearly just spinning my wheels?” So, I think that winds up getting all the way back to the drum I beat all the time about actually owning the tracking and the rigor behind the entire life cycle of analysis projects.

32:28 MH: Well, and for the times when you can’t take immediate action, storing that somewhere and keeping it current and in the view of the business, so that it can be prioritized and acted on. And I remember in my days as a practitioner analyst, we managed to get ourselves all the way up into the steering community prioritization meetings with our big list of “Here’s things we think we can accomplish on the website,” because we’d built up that backlog. I might’ve said I disagree but I don’t disagree with the idea of keeping track of what is getting done and then going back and doing the measurement to fix that. So, you’ve mentioned that a couple of times, and I think it’s not more… What I’m trying to say is, there’s distinct parts or phases and I view that as a phase or as part of the analytics job, but not necessarily one that has to happen in sequence, like if you’re documenting where and when things are happening, then you have a running list of what you need to go back and quantify. Also, if that’s happening and those are… People are gonna come ask you too, so you better keep track ’cause someone’s gonna come ask for quantification and like, “Hey, you said we’re gonna make $5 million at this. What actually happened?” And then you come back and be like, “Well actually it was seven and a half million, ’cause I’m terrible at forecasting things.” No, I’m just kidding.

33:53 TW: So, quick little tangent. What’s you’re take on… Because when it come to that, it’s managing a process.

34:02 MH: Yes.

34:03 TW: And it is managing a process that can take a lot of time and a lot of rigor, and not necessarily the sort of person who is necessarily perfectly geared towards analytics, so where do you fall on the… If you’ve reached some level of scale, be that as an agency or a large organization to have a full time or maybe half time role that is just someone who’s trying to manage the intake prioritization and tracking. So ideally not just a PM who knows what a work breakdown structure is, but somebody who’s saying, “We are actually trying to shepherd this but probably not actually doing any analysis ourselves.”

34:51 S?: I think that’s the third person you put in that team, department whatever you wanna call it. The first two… ‘Cause we talked about this on a prior episode, if you could build a analytics department with three people, who would you hire? And so, I always chose the technical analyst, somebody who can manage the implementation of the technical aspects of business analyst who can crunch the numbers and work with the technical analyst and a successful manager that has already demonstrated success moving projects through the organization. That would be my ideal team.

35:31 TW: So, that’s still… That’s like two… ‘Cause on the one hand, that person would be the knowledge of all business knowledge, organizational knowledge, the ability to get things done, but also potentially the “You know what, I’m gonna roll up my sleeves and my tactical things, I’m gonna have my spreadsheet or my JIRA system or Basecamp or whatever what I’m tracking.”

35:51 MH: They’re gonna own that process and obviously, the analysts can help with that. They can update just whatever it is that’s gonna be the system, but if you’re drawing the racy matrix, this is the person at the end of the day who is accountable for it, right?

36:09 TW: Well ideally they’re, with the team that’s that size they are… They aren’t doing it just for the sake of ticking off boxes, which can happen if you say, “Oh we got a project manager we put into this role.” They just go with something abstract and if it’s tied to the person who’s actually trying to drive organizational change, they’re like, “Look we’re capturing the information that we need, when we need it and then we’re using that to actually feed to drive the organization.”

36:35 MH: Exactly. And the thing is ideally if they’re having success doing that, then they probably understand how to get good results from their people, right?

36:49 TW: Yeah.

36:49 MH: In terms of doing the blocking and tackling out in front to allow the analyst to actually get work done, and kind of picking up those side random requests and giving them this priority that they need. And again if they have credibility ’cause that’s the thing, right there is analysts have to go in and build credibility. And if you start with a PM or a project manager or a leader who isn’t an analyst, but already has some of that credibility in the organization, you’re so far down the road already. You’re gonna have a ton of success and your analyst…

37:27 TW: Maybe, I don’t know. I’ve seen the PMs who are well regarded PMs, yet they’re total box checkers and so they come and they tell you what data to deliver.

37:36 MH: That’s fair because… I forgot about the extremely, way more dysfunctional than I’m picturing in my head, organization. It’s like oh yeah, they’re the the best PM ever and its like, “Best PM?” It’s like, “Oh, no!”

37:53 TW: I had a case where every time this one PM was on it I’m like, “Are you kidding me?” And it was funny because apparently that PM thought I was great and I’m like, “That PM, I want to put my head through a… ” Even now I’m getting… I’m no longer working on that client and I still get through back channels. They’re like, “Oh, guess who’s back and guess what that PM just asked for?”

38:17 MH: See, that’s because Tim, you are the quintessential analyst.


38:25 MH: PMs love you. No, so this is really sad for our listeners ’cause we couldn’t come up with really a single point in which we violently disagree and I think…

38:40 TW: I think we disagree with… I think there are plenty of analysts that actually will violently disagree with… Because we basically said it starts super, super early and we went as broad as possible and…

38:52 MH: Yeah that’s true.

38:53 TW: Definitely, the analysts I’ve ran, into there were some that will violently disagree.

38:56 MH: There are some, for lack of a better word, there are some places that are kinda like union shops where it’s like, “Oh, I can’t pick that up. That’s for a local 102 to do.” And again, nothing against unions. It just means the way that happens in reality can sometimes look dumb on the ground. It’s just sort of like, “What? That doesn’t make sense to me.” And it’s like, “Well there’s arcane rules behind it all and you have your box and you stay inside it.” And then at really large scale, when you’re talking huge teams and massive organizations, some of that makes sense ’cause no one person can manage all of that. So I can get that to a point, but if you’re sitting in a less than five billion dollar business then I think this is decent advice.

39:52 TW: Yeah.

39:53 MH: Alright, so Tim. You’re a pretty smart guy and on the show we do this thing called the last call where we go around we just talk about stuff we found recently that is interesting to us. What have you got?

40:09 TW: So I’m gonna do it. I’m gonna have to do a twofer ’cause I cannot decide. So, one is gonna be very short. It’s just really really cool and this is over a month old at this point. But I don’t know if you caught this, but on April 1st, for about 72 hours, Reddit did a thing on Place that was a pixel collaboration. Basically any user on Reddit could go in and change one pixel on this massive canvas to any color, but they could only change one pixel every five to 10 minutes.

40:45 MH: Yeah, I did see that.

40:47 TW: It’s just freakin cool and some of the analysis were done of it. So it’s a visualization. It’s not a data visualization, but even some of the analysis that were done of what happened and the different types of strategies that were taken in. So that was just too cool not to mention so we’ll link to that. My real last call is probably gonna ramble a little bit. So I was in Russia last month for the Go Analytics Conference and spent a lot of time hanging out with the Yandex people, specifically the Yandex Metrica. And where I was kind of struck was… So Yandex, which if anybody goes to analyticshour.io, they’ll see we’re running Yandex Metrica cause I was working on a little demo. And Yandex Metrica, turns out it’s kinda cool. It’s a basic web analytics thing plus it has heat mapping, ala ClickTail. It’s got session replay and I was like how are they… What is Yandex? And so Yandex as a company is literally like the Google of Russia and that they have higher market share in some aspects of the exact same things Google does.

42:05 TW: I don’t know exactly where they fall, but they have display networks. The have paid search. They have the go-to map, destination finder for the Moscow metro system. They have taxis, there’s Yandex taxis. So, they just have all this stuff goin’ on, but they have a… Their analytics platform is just like Google Analytics started out. It’s free because it drives the other aspects of their business. They don’t have sampling. I think their UI’s a little farther, maybe not quite as advanced as GA, but it’s a pretty cool tool, and it just got me looking in the US, we’ve got, basically, Google and Adobe. And in Russia, they have Google and Yandex. Adobe has just a tiny, tiny footprint. And they can duke it out in those local markets, but there is innovation that isn’t happening because they’re in the markets… They’re not trying to sell an analytics product. They’re trying to sell all these other services, and analytics is backing it up.

43:22 TW: So, I don’t know. It was kind of a rambling, to me, I was there for almost a week, so I was pondering it a lot and literally see Yandex everywhere, and it was just like seeing Google. Everybody knows Google, but does your second cousin know Google Analytics, and it’s exactly the same with Yandex and Yandex.Metrica. So, it’s like being on a another planet that had an insane number of parallels to my home planet of the US, but it was a completely different company. It was just really a mind-bending thing, and if you’ve got a site and you… That’s not like a corporate… If you’ve got one that you can throw another tag on, we’ve implemented it in five minutes through TTM on our site, and it’s fun to poke around with a different tool. It’s fun playing with their little tools, especially ones that don’t require a whole lot of time.

44:16 MH: No. Well, I, actually I think that’s pretty awesome, and as someone who’s had the opportunity to go to Russia once before, I’m super-glad that you got that experience ’cause it’s…

44:24 TW: It was an experience. [chuckle]

44:27 MH: It’s a unique, and it’s valuable for us Americans to get over there. Alright, so my last call…

44:34 TW: What’s yours. Yeah. Whatcha got?

44:35 MH: So, a couple months back at Adobe Summit, you meet all kinds of people during that time, but one person I met was this woman from Singapore who runs a small analytics consultancy called Attribute Data. And she and one of her team were there at Joanna Teo, and she was showing me this little proof of concept that they built leveraging Adobe and IBM Watson’s API, and it’s actually really cool. So, what they did is that they basically took, they ingest your Twitter feed, they run it through Watson’s API from IBM, and ascertain your personality type, feed that back out to you and then deliver via Target a personalized message based on your personality characteristics from…

45:27 TW: What? ENTJ?

45:29 MH: No, no. It’s more. It’s not Meyers Briggs, okay?


45:37 MH: It’s just a proof of concept, but it was just really neat, and it was just really fun to meet innovative and creative people from around the world. And so, it was just… Anyways, it’s just something that stuck with me from my Adobe Summit experience couple of months ago, and it’s a pretty cool little proof of concept. If you go to attributedata.com, you’ll see it on their home page or at least, as of this recording, you would… I imagine they might leave it there for a little bit longer.

46:04 TW: Where are they out of? You said…

46:05 MH: They’re based in Singapore.

46:07 TW: Oh.

46:08 MH: Yeah. And so that’s also pretty cool, so if you’re lookin’ for some good analytics talent in Singapore, it seems like they’re a pretty sharp crew.

46:19 TW: Nice. Awesome.

46:20 MH: Anyway, just throwin’ that out there. I thought it was pretty neat, really fun to meet those folks. Alright. Well, listen, you’ve been hanging with us so far listening to us just destroy your concept of where an analyst’s job starts and ends, and it’s time for you to get back into this conversation via the Measure Slack or our Facebook page or Twitter. We would love to hear from you, so don’t hesitate, stop the car, pull over, pull out the phone, and just start Tweeting. No, we’d love to hear from you. I think this is something that if we continue to push this conversation and grow our understanding of what we should be doing as analysts, I think it will make us all better off. And I think that’s the show. I think that’s all we got. Tim? You got anything else? Are we just…

47:08 TW: Uh. No. As always, give us a review. Give us a rating on iTunes.

47:13 MH: Ratings and reviews on iTunes. It’s a… Damn it. We’re begging now. No, I’m just kidding.


47:22 MH: Anyway…

47:23 TW: “It helps others find the show.” That’s what they always say.

47:26 MH: Is that what it is? Okay.

47:27 TW: I don’t know. I don’t know if it does or not. I think that’s just an urban myth at this point.

47:30 MH: Well, we’re both good and bad at this self-promotion, and I guess that’s okay. I guess that’s okay. So anyway, Tim Wilson as the quintessential co-host. [chuckle] I think you would join me in telling our audience to keep analyzing.


47:53 Announcer: Thanks for listening, and don’t forget to join the conversation on Facebook, Twitter, or Measure Slack Group. We welcome your comments and questions. Visit us on the web at analyticshour.io, facebook.com/analyticshour or at analyticshour on Twitter.

[background conversation]

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Have an Idea for an Upcoming Episode?

Recent Episodes

#249: Three Humans and an AI at Marketing Analytics Summit

#249: Three Humans and an AI at Marketing Analytics Summit

https://media.blubrry.com/the_digital_analytics_power/traffic.libsyn.com/analyticshour/APH_-_Episode_249_-_Three_Humans_and_an_AI_at_Marketing_Analytics_Summit.mp3Podcast: Download | EmbedSubscribe: RSSTweetShareShareEmail0 Shares