Or use one of the embeds at the bottom 🚀.
This is a longer one, really in the weeds, it's a deep dive. (I'm saying it too many ways because we're feeling out of we should do more long ones like this.) What do you think?
Anyway, on this one we talk through one of the messiest topics that marketers and honestly who companies deal with, Demand Gen Workflow.
We start with a broader convo about why this stuff matters, why this stuff is tricky, and (the usual) why this is the transition of marketing into engineering. Sure, a CRM is easy to use, but its actually just a graphical interface for a regular old data database. It's complicated like a database, for which we have engineers. This is part of that, so we talk about it.
Then in the back half (if you want to skip there) we actually run through our demand gen flow, mostly focusing on marketing. It's laid out in the spreadsheet below, so, if you were smart enough to pull up the notes. Follow along! 👇
More Netwise:
YouTube | Twitter | Facebook | Linkedin | Web I Blog+Newsletter
Adam Kerpelman:
I was out and about, it doesn't matter where I was, and I was watching somebody check their appearance with their selfie cam. It made me think like, "As a photographer, I understand the warping that's happening as the light passes through that lens versus just a mirror, which is flat and shiny." Do you think that's a whole different set of image issues that we're going to have because people think they're a little fish eyed?
Brian Jones:
I'm looking forward to the camera just taking control and making everyone look beautiful all the time. It's already doing it in Zoom. Everyone's got that little slider [crosstalk 00:00:37].
Adam Kerpelman:
I was going to say, wait till we've all got glasses, and then we can all just live in ... You just pick your avatar.
Brian Jones:
It's going to be wonderful.
Adam Kerpelman:
I don't ever have to see you again.
Brian Jones:
It'll be wonderful picking out those new avatars.
Adam Kerpelman:
Unless I take off my glasses.
Brian Jones:
Then it's Halloween every day, which has always been a dream of mine.
Adam Kerpelman:
Hey, it's Data-Driven Marketer sponsored by NetWise. I'm Adam.
Brian Jones:
I'm Brian.
Adam Kerpelman:
Welcome back for another hang in the Data Basement. How's it going?
Brian Jones:
It's going pretty good. Just frantic as usual.
Adam Kerpelman:
I do legitimately want to ask how it's going at the beginning of the podcast every time.
Brian Jones:
It's what people do.
Adam Kerpelman:
It's good for the conversation. Also because this is a work podcast, I always feel a little weird. The answer's always like, "I don't know, man. It's Tuesday at 3:00. I'm at work." Not like you're going to be like, "Well, it's Saturday, and I spent the morning working on a model of the Enterprise D."
Brian Jones:
I'm currently working on a model of the Millennium Falcon.
Adam Kerpelman:
Perfect, I was close.
Brian Jones:
I was debating an Enterprise ship when I bought it.
Adam Kerpelman:
What are we talking about this week?
Brian Jones:
We're talking about the marketing, sales, and customer success funnel steps, which as anyone knows are a bunch of haphazardly named things that everyone interprets differently.
Adam Kerpelman:
We have been working on this internally for a while now, and today, a representative from LinkedIn told us that this is the single hardest thing to figure out-
Brian Jones:
All companies are struggling.
Adam Kerpelman:
... in the data-driven stack. Every company's like, "What's an MQL? What's an SQL? I don't know man, just name it something."
Brian Jones:
Everyone thinks it's something different. There's a really good set of reasons why this is so hard, and so we're going to talk through the concepts and then how to make a mental model of this that's functional for the business because again, you can always come at this from an engineering perspective. This is a very engineering problem. There's a bunch of data that we need to classify and organize and put concepts around and be very explicit about.
Adam Kerpelman:
Historically when you talk about sales funnels or marketing steps, it's not been as binary as computers want stuff to be. As you transition from this thing that's essentially like a humanity over into, I mean, you would study it in your humanities degree at your liberal arts college, over to an engineering department problem is because you want binary failure states in an engineering system. You want to know if the thing progressed from one step to the other. Marketing and sales have traditionally been like you all get in a room, and the guy goes, "I don't know, man. I don't think that one's going to close. Stop chasing it."
Brian Jones:
I think it's at step three, which is a reasonable way to handle things until, like you said, you bring in the computational complexity of modern, digital, data-driven marketing. When all of a sudden, for something to qualify being an SQL or an MQL or a lead or whatever step you're talking about has just massive amounts of complexity behind it that's nonlinear. It's not a list of things. It's not like an opinion. It's quantifiable aspects of your business that you're trying to digitize. It's just really complicated to organize this stuff and to find it and to put it into functional practice.
Adam Kerpelman:
We talk about funnels, and we talk about pipelines and stuff like that in the context of these conversations. Everybody's kind of used to this idea of a stepped process. What I've come to realize as we've been setting up this one recently is the difference is that those are actually the reports that you're used to looking at. You have a funnel report, and it's mostly manually filled out. It shows each step of the process, and it's kind of like those are cobbled together by the managers at the end of the whatever time period so they can report up the chain how this thing is working.
Adam Kerpelman:
Everybody is aware of this idea that it goes step by step, but they don't yet think of it as a program. That's the difference because if you're talking about programming a funnel, you have to be really explicit about definitions, like literally what in programming you'd call like variables and classes. It's easy to say to somebody, "Yeah, but if the variables don't match, then the program breaks." Then everyone goes, "Yeah, totally." Yet because of this structure we're used to with leads and all that kind of stuff, we have eight different words that mean the same thing.
Adam Kerpelman:
Now that we're trying to put it into a computer program more explicitly, it's like the computer can't handing any one of these eight things could be any one of these eight things, unless it's the first Thursday and I haven't had my coffee.
Brian Jones:
You've got the extra added humor here that every marketing tool and every CRM does it differently. It's been hilarious. Our company uses HubSpot as its CRM. It has for years, and now we're using HubSpot for marketing and for customer success and a bunch of other stuff. It's hilarious how often some of the teams we're working with will say, they'll mention another CRM or another stack and be like, "Oh, you're talking about it because this one does it this way. They call this that. They use this-
Adam Kerpelman:
[crosstalk 00:06:41].
Brian Jones:
... variable and that transition state."
Adam Kerpelman:
They can call an opportunity this, and that's an opportunity whatever.
Brian Jones:
It's like, that's not why we were confused, but I'm glad to know that all the [crosstalk 00:06:49].
Adam Kerpelman:
Well, but then hilarious thing with me and you as we've been working on this plan is increasingly, because we're literally talking about trigger states to move a thing from one thing to another, and so increasingly, our notes look more and more literally like code. This if/then statements that we then just by thinking of them like programmers, we're then like, "Oh, we need to be more explicit about the failure condition for that if/then loop." It's like this is just engineering. The same old, beating the same old drum.
Adam Kerpelman:
That being the case, we'll talk specifically about our funnel and HubSpot and stuff in a minute, I think. What really is the challenge that we're dealing with up front there. What's the broad instructions? Because what platform you're using and how they designate stuff doesn't matter. It's all just-
Brian Jones:
It doesn't.
Adam Kerpelman:
... variables. It's variables. First thing is like what are those variables for?
Brian Jones:
You're totally right. What the platform is and what people call things and how it's set up doesn't matter. What does matter is that understand how the platform is using these things, like what their definitions mean. That's been complicated for us with HubSpot, and I've been actually surprised how complex their setup is because it's got this multilayered. It's got like multiple flow layers that don't actually align with each other. I still don't know that I like all the things they have. I still think there's some fundamental issues there. I can't quite put my finger on its because I'm not the one in it day to day.
Brian Jones:
Much like when you're implementing a new piece of software or a new API system or a new service as an engineer, you've got to understand how it works to interface with it. You can't just start building stuff on top of it, which to be fair is how most engineers do stuff too. They just, "Here's a thing. I'm going to start using it," and it ends up being a mess. It's really important that you first go into the application you're using. I think we're mostly going to be talking about CRMs right now because-
Adam Kerpelman:
Yeah, for sure.
Brian Jones:
... sales enablement, I think is some other terms for some of these platforms that are like sort of CRMs plus sort of lead gen, demand gen platforms, like kind of help you send email sequences. A bunch of different names for them, but systems that kind of manage customers and where they are in a flow. The best thing you can do is just what's it doing out of the box? How does it work? What does it call things? What does it want you to do? Because you got to remember, these are inventions. CRM as a concept is not 100 years old.
Brian Jones:
It's not what salespeople were using door to door selling globes in the 1940s. That was simpler for a reason, and it must be joyous to go back to that time and just be like, "My sales team sells globes door to door, and they give me a report at the end of the week." We don't have that anymore.
Adam Kerpelman:
Why'd you skip that house?
Brian Jones:
We've got 10,000 different signals coming into the business about how people are interacting with you.
Adam Kerpelman:
Well, and so I mean, this is the thing that I think people don't like doing but engineers understand is part of the job. You got to look at the manual. You probably need to read the whole thing. It's like programmers, when I say engineer, I mean, literally all engineers, mechanical engineers wouldn't just start cranking on a giant machine that does a thing in a warehouse. They would get in trouble. They sit down, and they understand the pieces.
Brian Jones:
It's a good thing HubSpot can't cut my fingers off.
Adam Kerpelman:
Yeah, exactly. Engineers understand that you need to understand the definitions and the typing and the classes. You have to look at the docs if you're going to adopt a new programming language. That's what we're talking about here, really friendly to use but hyper complicated database management software.
Brian Jones:
[crosstalk 00:10:57].
Adam Kerpelman:
If you start messing up the variables in your database management, you're going to end up having to burn the whole database down and just start again eventually.
Brian Jones:
Which everyone just heard that and they're like, "Oh yeah, we talk about our CRM being a complete disaster zone." Everyone's CRM is a dumpster fire. You get that joke constantly. Part of that's unavoidable, but that's also something that has to constantly be done. In any engineered system, our engineering system is constantly going back in and redoing things they just did. You finish a week of hard work, and you got to go spend a couple days, go clean up what you just did or a month from now, you won't remember and everything will be a dumpster fire again. It's a tremendous commitment.
Brian Jones:
That's really funny you said engineers I think like doing this stuff. I actually think engineers hate everything they do secretly because it's all really annoying. It's why engineers are annoying, right?
Adam Kerpelman:
I wouldn't say they like doing it.
Brian Jones:
They like having had done it well. That's what most engineers like.
Adam Kerpelman:
[crosstalk 00:11:51].
Brian Jones:
They're like, "I did it more right than you did it." That's what engineers like.
Adam Kerpelman:
Oh, let me show you what you messed up and where in the manual you could've found the answer.
Brian Jones:
It was awful, and they hated the whole week, and they were unhappy-
Adam Kerpelman:
For sure.
Brian Jones:
... at the end.
Adam Kerpelman:
That's the first thing. Sometimes it's hard to do, I think, because projects will be underway already, and so for me, for us, for the last month or so, that's been a lot of me saying, "Okay, is this the thing that we can change by understanding our definitions and our process, or do we have to change the way that we do that to align with a thing that's rigid within the software that we can't change?" There are a few names in HubSpot that every time I see them, I have to substitute with the name that I would've called it if I were designing this product. That is-
Brian Jones:
[crosstalk 00:12:47].
Adam Kerpelman:
... what it is, and you can't change it and so whatever.
Brian Jones:
What's that file called again? Can you send me a link to it? I can't find it.
Adam Kerpelman:
You need to understand that ahead of time because it's going to happen that way. It doesn't matter what the CRM is. This is the part where you get this mismatch with the consultants going, "Well, in Salesforce, they call it an opportunity, but in HubSpot, it's a deal." You need to know which because for me, as the CRM admin currently, if I can just rename it and solve the problem at the customization level of the software, I might just do that. If I can't do it, now I got to starting the real fight, which is I have to retrain a whole team of people to do a human thing, which is hard [crosstalk 00:13:30]-
Brian Jones:
This is why you'll see if you search our data and dig into different jobs in kind of the sales and marketing space, you'll see a lot of people who are HubSpot admins or Salesforce admins. It's a very complex job. I would guess at a large company, you have whole teams whose job it is just operate and put processes around and limit access to things inside of these systems. Again, it's an engineering role. You're not at the terminal writing code although are you behind the scenes on a lot of these systems. HubSpot has a lot of custom code spots where you can plug stuff in, and we even have some.
Adam Kerpelman:
That's step one, which may or may not have happened if you find yourself midstream with a CRM, but backing up to it is always important to facilitate the next steps. Step one is understand the CRM software that you are using and how it views the world. Also, because that can teach you. I mean, it's a funny way to personify it, but literally, that's how I say it in our meetings all the time. I'm just trying to understand how HubSpot categorizes the world so I can conform our processes for a lot of things.
Brian Jones:
Well, you're hitting on something really fundamental with software, and I'm especially seeing this shift in general with software that's available for people. We had a period 10 years ago where very specific software was launching for very specific tasks because software was just coming to be in industries where it had never been before. Now, a decade later, we have all this software that you can't quite tell what it is. It's like a bunch of things. It services five different tasks at a business or 10 different things that you do at home during the week. They overlap with five other pieces of software.
Brian Jones:
We're in this squishy phase where the software is a personification of the team that's building it, of the engineering team or the founder of the business. It's some vision that's based around how that team needed to solve a problem. You really are running into personality. This is an important step to realize with any modern software system. We're using something called ClickUp now for our project management, and it's wildly complicated. If you didn't go in and read the docs about how they intend you to use it, you're not going to use it well.
Adam Kerpelman:
You're not going to use it right. Then you're going to be frustrated that it didn't work. This is the whole customer success part of the thing. These things are so complicated you need a department that makes sure people know how to use it so they don't churn.
Brian Jones:
Exactly, right. It's especially interesting in the sales and marketing space because there's so many new products. There are 10,000 companies in this space based on Chief Martech's giant graphic that I think a lot of people have seen. Those are all abstract, custom software to access data and manage stuff. They're all different, and the concepts are all different, and their goals are different. This is an increasingly important concept when leveraging software to do your job efficiently.
Adam Kerpelman:
The other funny thing, I think and this may fade, it will fade given the sequence you just talked about, but for our generation, we are still stuck personally knowing and that means everyone older than us also knows what it was like to have software that was like, "That's simple as shit. It solves exactly the problem I need it to solve. It wasn't hard to learn. Moving on." It's not that way anymore, but people always expect it to be.
Adam Kerpelman:
People are constantly mad at their software because it doesn't do everything exactly as cleanly as Windows 95 apps used to. That's the world we live in now. HubSpot is an operating system for 10 different things, and it sucks at four of them. It's all integrated, and so it's the best we can get. You put up with it.
Brian Jones:
I mean, you hit a really funny split there. I don't know if this is age or experience or job function or whatever, but there does seem to be a group of people who are kind of like, "Why isn't it like it used to be? Why isn't it a simpler thing? Why do we have to use this new piece of software when I have this old one that worked well?" Then it gradients over to the other side, who is like, "This still is not the right piece of software. I need to go find another one to do the thing that I need done in here." Neither of them are the right approach quite.
Adam Kerpelman:
Still missing the ones. It has everything except the one thing that I need. The last thing I think on that, I think a lot of people sometimes run into this with the attitude of like, "I'm reading the docs. It's research. It's boring." You really can frame it as learning. Me trying to get my head around how HubSpot thinks about the world has taught me things about how this world works because they are successful software. Their worldview can't be that far off something that's functional for thousands and thousands of companies.
Adam Kerpelman:
If I'm like, "Well, they shouldn't do it that way," what do I know? They're the ones successfully selling SaaS into a giant market that does X, Y, and Z. If they want to use a word I don't like, I at least got to leave in my head, "Well, okay, there's probably a reason for that."
Brian Jones:
Are you specially targeting my distaste of subscriber as a funnel category?
Adam Kerpelman:
No, no, I'm just saying like it's how I try to reframe it sometimes when I'm like, "I got to watch another video to learn how this thing works?"
Brian Jones:
Totally, totally.
Adam Kerpelman:
It's like, "Well, maybe that video teaches me something about data-driven marketing that causes me to go, 'Oh shit, lead scoring. Okay, I hadn't thought about that. Woo.'"
Brian Jones:
That is a really neat aspect. You're hitting on something really neat here, and it's continuation of what I was saying earlier about how the software has gotten kind of broader, like less specific but more specific at the same time. Anyway, you can, you can go into these system. The point is the software is getting to the point where the teams building it are saying, "This is the process we saw our customers at our previous CRM company doing all the time that worked the best, so we're taking it to the next level where we control how you send emails or we control how you send ads so you don't have to think about it."
Brian Jones:
The whole point is to bring more efficiency to more companies. If we were all Fortune 1000 companies and had 1,000 people on a marketing team, we have the manpower to take. I just did a webinar with AWS. It was like 10 people coordinating for a month to film a one-hour webinar. Nobody has time for that.
Adam Kerpelman:
Well, and I think the apt bit in the data-driven space is it's not just about how they about the world. It's about what do they say. The pitch for the company five years and a billion dollars ago was, "We all worked at a CRM company, and we realized that people weren't paying attention to this data point and this data point and this data point, so we built a whole universe around that data." You see them evolve in that way because the ones that have funding continue to evolve. It's like, "Oh, it's all net promoter score."
Adam Kerpelman:
They can see how the CRM evolves in order to collect the right data to make NPS work for you or whatever metric you're chasing at this point. It's not just about that worldview thing. It is largely about what data you're getting, which so that gets us to the next step. Once you understand the programming language of the system that you're dealing with, the next step is still don't go in there and do it. Sit down and outline the process for how each one of these steps that the software is envisioning aligns with how your company wants to do it for your use case. In our case right in the middle of the funnel, we kind of have some overlap between sales and marketing that wouldn't be there if we were just selling t-shirts on Shopify.
Adam Kerpelman:
We have a sales process and a complicated product, so there's overlap. We have been talking about just those two steps where everything overlaps for like a month and a half to try and figure out the right way to classify this stuff so that the program works and so that the program casts off usable data. I think the only way around that is you just got to get all the stakeholders in a room and do flow charts or whatever works for you guys, right?
Brian Jones:
Yeah, I mean, the flow chart is a great segue into the actual funnel because it's really easy to draw a funnel shape on a piece of paper and slap some colors in it and put some names. You can go search Google for sales funnel right now, and they all the look the same. Everything's labeled differently. Every single one has different labels and different notes and different thoughts. It's interesting to see how complex that gets. A lot of that is because business is complex. Every business is very unique, even businesses that sell the exact same thing. You have 10 companies that just sell bike tires. They're going to go to market very differently. They're going to be structured differently.
Brian Jones:
Their sales team operationally will be differently. Their branding is different. Just because of who they hired, things will be structured differently. There's a tremendous amount of customization that's really important here. Beyond being just functional, it's very strategic. How you go to market is extremely strategic. How you handle customer conversations, when you send out contracts, when you let customers in, when you first send out samples or products or renewal notices. A lot of that has to be put in, so there is definitely a major analysis paralysis aspect to setting all this stuff up because it is complicated and because there is a handoff point to multiple departments at different steps in the flow.
Brian Jones:
As your business grows, you are constantly running into squishy areas where it's like, "Wait, I know we were doing this this way before, but my department now needs this other data point that we didn't have to think about." That's where we are right now. We've been using HubSpot as a CRM, and we have a very efficient, well-designed sales process in there. Now we're trying to layer in a marketing funnel at the front of it. What we need to capture and what we need clarified and where data overlapped between those departments and between concepts is a little different. In those squishy spots, you've got to reimagine things, which gets really complicated really fast with modern software.
Adam Kerpelman:
Well, and so that's why you get to the flow chart piece. You can't just look at the funnel and say, "Oh, well, it's lead, and then it's MQL, and then it's SQL, and then it's whatever." Each one of those is steps. Each one of those isn't even steps. Each one of them is designations for the progress of a contact or lead or whatever you want to call it through the system, and in between that, there is another computer program running that is assessing the progress from one step to the next.
Adam Kerpelman:
You need the flow chart because you need to have, here's this one stage. Here's all the stuff that happens in between. If X equals Y, then they move to the next step. That's what every phase of a funnel looks like, particularly with the data-driven approach and if you're trying to scale an automated marketing machine.
Brian Jones:
I think it'd be fund to talk through our funnel as it sits right now. I'm not even using the right terms now, talking to you about all the terms we've been trying to define.
Adam Kerpelman:
It's funny because it's like a man gen like contact flow basically I guess, whatever you'd call it.
Brian Jones:
There's intelligence like in the conversation. It's really important, and you define a lot about the business just walking through these concepts because like you keep saying, even knowing how complex this stuff is while we're in the midst of setting it up for a new product, I still will leave a conversation and later think, "Oh yeah, that's just this step." Then I come back and talk to you or our VP of sales and marketing, and I'll say something. I'll be like, "Shit, that is not what we talked about it, and that's not actually how I know it works. Why do I keep thinking of it that way?"
Adam Kerpelman:
Well because, I mean, and some of it, again, that's why you start with understanding the system because some of it is counterintuitive because the goals, the data and the goals and the stuff that you planned out don't always align with how a human brain even thinks about it.
Brian Jones:
It's part of the reason why you have 10,000 companies in this space because everyone thinks they can build a better version of all this.
Adam Kerpelman:
I think the last thing to hit before we dig in our funnel as the use case, whatever, case study, I guess. The last step I have written down is educate and enforce. You have to make sure that everyone else who's going to touch this funnel understands the rules, understands the processes, at least as far as they impact them, gets on the same terms and terms of the language. If you're busy rolling this out at another company, that's going to be really uncomfortable. The best you can do is apologize for it upfront and then keep being that guy that I am constantly apologizing and then saying, "I'm sorry, what you mean is a lead."
Brian Jones:
You're the sales funnel police.
Adam Kerpelman:
It sucks, but it'll go away eventually. If you don't use the right language among the humans, then this tenuous computer system we're trying to set up that's hybrid, it's like half computer and half humans, breaks. Then the data is trash again.
Brian Jones:
Do you identify more as a computer or more as a human?
Adam Kerpelman:
Personally?
Brian Jones:
Yeah.
Adam Kerpelman:
I guess a computer to be completely honest.
Brian Jones:
That's the right answer. Another good comparison to engineering, which I guess I'm getting tired of saying but will probably continue to be a theme, sometimes you just have to be obnoxious and be like, "No, that's not ... Wait, what did you mean? Did you mean this, or did you mean that? Hang on, one second. Wait, let's go back. Oh wait, I don't remember which one is which. Can we go look at the definition that we wrote up and agreed on?" Which is a really important step with this. Again, we're defining concepts. These are like in a legal document, at the top of the contract, there's a bunch of terms to find so that you can reference back to them.
Brian Jones:
Again, it just gets so complicated so fast, and the payoff to having this so organized and agreed upon and centralized, like you said, it's as important that the person putting this together write it down for themselves as it is that the other team members have it to share just so that you can communicate effectively about it because it's only getting more complex.
Adam Kerpelman:
Then you can just long to be a computer where all you would have to do is not work if the right variable is not used. Instead, I got to go, "What do you mean by that?" Then keep trying to do the job. It'd be beautiful if I can just not answer-
Brian Jones:
You need a little flag that's like-
Adam Kerpelman:
... until you use the right word.
Brian Jones:
... "404 error. Variable not" ...
Adam Kerpelman:
Then you just have to think about what you did wrong and try again.
Brian Jones:
Computers are really annoying party guests.
Adam Kerpelman:
Yeah, they're really annoying party guests. Cool, so let's do it. Our demand gen flow, what's the way to do it? Should we talk about the steps first?
Brian Jones:
I think it's to try to use HubSpot terms right, and this'll be kind of an entertaining bit to see how often we screw up. HubSpot's lifecycle stages, I think it's like the-
Adam Kerpelman:
Lifecycle stages is the highest.
Brian Jones:
... those are like the high-level ones that actually flow between marketing and sales and customer success the way we're setting it up, right?
Adam Kerpelman:
Yeah. I mean, that's out of the box, that's how it behaves. The first stage, which already is the confounding part they call subscriber.
Brian Jones:
Hate it.
Adam Kerpelman:
You can think of it as a pre-lead. I don't know where that terminology came from. I can't change it in there. Subscriber it is.
Brian Jones:
It has nothing to do with subscribing to something.
Adam Kerpelman:
Go ahead.
Brian Jones:
We think it's a legacy from another company they bought, and they had to tie it in to get their marketing software working.
Adam Kerpelman:
Something like that, or it's something about a type of funnel that I don't understand. It doesn't apply to our business, so it doesn't matter.
Brian Jones:
Totally.
Adam Kerpelman:
Maybe someday I'll go, "Oh shit, I get it now."
Brian Jones:
How are we using subscriber?
Adam Kerpelman:
It's more like you could call it pre-lead.
Brian Jones:
Pre-lead.
Adam Kerpelman:
It's a catchall for anything that ends up in our CRM for one way or another but has not yet crossed the bar into saying, "Okay, this is"-
Brian Jones:
It's kind of like we have a bunch of people because everyone has this in their CRM. There are tons of people in there. No one knows where they came from. They're not associated with anything. We throw them in subscriber, and then they're at least bucketed. We can look in the data and be like, "Hey, all our data is at least in a category in a customer lifecycle or lifecycle stage." Customer lifecycle stage?
Adam Kerpelman:
Yes, customer lifecycle stage. If you follow our flow chart, you got content and campaigns. They'll run into subscriber by default. Then anytime there's any kind of proactive engagement, we bump them into the lead thing. You click on a link, you click on an ad, fill out a form, those all end up as "leads," which is the next step.
Brian Jones:
As an example, like when our sales team is prospecting on their own, they're going out and finding people and companies that might be relevant. They want to just really target a company specifically, and they want to put a company and a person in, that company or person has never done anything. There's no real signal that they're interested or engaged in the business. They'll go in-
Adam Kerpelman:
They're not a lead.
Brian Jones:
... as a subscriber. They're in there. They're organized. They're in a flow that a computer can understand. They've never done anything yet.
Adam Kerpelman:
This is a funny aspect of the struggles here because in the effort to make it data driven, we're redefining lead compared to what the person selling the snow globes house to house used to think of as a lead. Because the people programed for that type of sales behavior, if you ran out of leads, your manager flopped the phone book in front of you and said, "I just got you more leads. Start making calls."
Brian Jones:
In a modern marketing funnel, a lead can be a completely anonymized signal from somewhere in the ether online. It's an IP address that your computer dropped a cookie on. There's no identifiable information. You can't look it up. You can't go call the IP address. It wasn't in a book somewhere. That's important. It's gotten so granular that it's beyond human conception to chase.
Adam Kerpelman:
We kind of have to redefine lead because if a lead is just a person who might buy your thing because they're on a list, like I got 90 million records for you.
Brian Jones:
It's a lead for the computer salespeople now.
Adam Kerpelman:
Have fun with the phone.
Brian Jones:
The computer salespeople are running Facebook ads and retargeting.
Adam Kerpelman:
Lead, at least in our system, means that it's someone from an audience. If sales comes up with a list of people and they go, "I bet that marketers at mid-sized ad agencies would be a great target," awesome, they feed that audience to us. They're saying, "Based on my exchanges out in the world, I think this could be a really good target audience." We use a tool like NetWise to look up a crap load of people that fit that description, and then we put them in at the top of the funnel. They aren't even subscribers at that point. If they have to go in the CRM, they're subscribers. They are target audiences for the campaigns. Then if somebody clicks on a link, then they're into our system.
Brian Jones:
Now we're seeing the signal.
Adam Kerpelman:
If they view two pages, now you're a lead or whatever. This gets to the next step after the lead piece, which is lead scoring.
Brian Jones:
Very specifically, because I used to use all these terms even like a month ago. I was using a lot of these terms interchangeably and incorrectly. Lead scoring is very specifically the way we're using it at the lead step. It is the lead step scoring. It's not that like all these things are leads. Everyone in your system is a lead. That step now is when it's considered a lead.
Adam Kerpelman:
I mean, and this is sort of the pre-lead lead designation piece. If you wanted to call everything a lead, yeah, you could, but a lead with a lead score of zero is not a lead. That's another way to think of it. Yeah, fine, they're all leads, but subscribers all have a lead score of zero.
Brian Jones:
That's fair.
Adam Kerpelman:
The second you have a lead score of five out of 100, okay, fine, you're a lead, whatever. We need to put tags on things so we can track them through the system and graduate them to different levels. You click on something, now you have five points, so you're a lead. This is where the marketing automation software that everybody is trying to sell you if you're a marketer right now comes in. Now you run people through nurture campaigns, and you do different things to try to figure out whether or not these are good leads.
Adam Kerpelman:
Every time you get a signal that's good, you increase their lead score. Every time you get one that's bad, you decrease it. If they hit a certain trigger score, it's like we're just playing an RPG here, it level up to being a marketing qualified lead.
Brian Jones:
For instance, someone might see a Google ad we're running in Google search, click on a link, come to our website, browse-
Adam Kerpelman:
Five points.
Brian Jones:
... three pages, right? Maybe they get-
Adam Kerpelman:
15 more points.
Brian Jones:
... five points for each page. They don't fill out a form. We don't actually know who they are. We dropped a cookie on their computer though, so now maybe we're running retargeting ads on Facebook and LinkedIn, meaning that person might see an ad two weeks from now, click on our site again, come over and fill out a form and say, "I'm interested. Can I reach out and talk to someone?" All of a sudden, that combo of points, they're going to continue going up in value until they hit a threshold where we say, "Okay, there's enough engagement here that we want to invest a salesperson's time on reaching out to this person."
Adam Kerpelman:
Business development that we're [crosstalk 00:36:16]-
Brian Jones:
We no longer want automated systems. We want to start going after this person.
Adam Kerpelman:
That's the first step. That's where you're trying to take it from a million down to 10,000 or whatever because you can't just have your sales department call a million people a month. That was a proper use of the word "can't" grammatically.
Brian Jones:
There's an important piece here too in that the lead, the way we're using a lead and the scoring method, it's where you're able to throttle how much you push through to your sales team. Your sales team's capacity will change day to day, and hopefully will change over time. As your marketing gets more sophisticated, as your sales team gets more sophisticated, as marketing hires more people and then sales hires more people, you go up and down. You're able to in this lead section throttle the scoring and change the scoring and change what makes a score or what that threshold is.
Brian Jones:
That will let more or less sophisticated leads through to your sales team. If they're having a week and they're way ahead and they need more leads, drop your threshold down, and more stuff will jump over for them to start to reach out to. If they're engaged with a ton of great leads, dial it back and let your system rise more cream to the top.
Adam Kerpelman:
Well, and that's the thing, that gets to one of the aspects of this that's tricky if you're not used to computer systems and stuff like this. This whole flow chart has to run as a sequence, or else the program also breaks. It's a computer program. It can run through five steps in a split second. You don't even have to see it, but it can't skip those steps. A lot of times, you'll go talk to sales, and you'll say, "Well, what if I meet them on the street, and I'm like, 'Hey this is an opportunity. I'm going to bump that to SQL.'"
Adam Kerpelman:
Doesn't matter. The computer is still going to think about it as new contact. Subscriber until told otherwise. Then a script is immediately going to tell it lead, and then a script is going to immediately tell it MQL, and then that same script is immediately going to tell it SQL. It's going to move through the steps one at a time. People say, "Well, why can't I just jump right in at SQL?" I was like, "That's not how computer programs work. They run sequentially."
Brian Jones:
It's also as important is the fact that that's how you get the complete picture of what's going on, which is really hard to set up now because you do immediately, you run into these. There are so many edge cases. We're not a big company or a very sophisticated sales process or marketing process. We're getting that stuff going, but already, there are dozens of edge cases when someone like our CEO will bring in leads sometimes where he met someone or he knows someone through his network who's a great lead for us. They're primed already.
Brian Jones:
How do you put that in the system? What's the process so that all of the data still tracks, but the salesperson who's relevant can follow up that same day. You don't want these processes to take a long time, but it's why you need the system or the process steps to find.
Adam Kerpelman:
Well, and I call it out because I think it's one of those things, it sometimes is hard as you're trying to get a system like this in place because people that are trained on old funnels see each step and go, "Okay, that's at least two weeks, right?" I'm going, "No, it's literally a millisecond. Stay with me here." Anyway, moving step by step through the thing. At MQL within our system, we graduate to a BDR or a sales rep actually running through sequences to talk to our prospective customers.
Adam Kerpelman:
Now, the same way our marketing system tried to qualify that lead and then tagged it as qualified by marketing, sales is now trying to qualify the lead for their purposes to say, "Okay, now this has the markers that we need human people to listen for, but once we see them, okay. We know this should move through the funnel." Then sales manually has to accept that lead for now in our system. There might be ways to automate that in your system or whatever.
Adam Kerpelman:
Sales accepted, so that ends up being kind of like the metric that's the last thing that marketing cares about. I'm just trying to get leads that sales agrees are good. After that, it's on the sales department, so we've just executed a handoff. Now it's on the sales department to take that sales qualified lead and move it through the pipeline.
Brian Jones:
This is a very pivotal step for us because we've purposefully defined the transition, and often it's the transition that's really important to understand here, the transition from MQL, marketing qualified lead, where our automated system said, "Hey, we think this is relevant," to SQL, sales qualified lead or sales accepted lead as you called it a second ago. We want a person on our team to talk to the customer because that signifies things. It means there is a person there. This could have all been something else. This could've been bots traffic on your website that you accelerated too fast or an email open that was not really a person.
Brian Jones:
It signifies there's a real person. They're interested enough to talk to your company, and the salesperson who's infinitely more intelligent than our SaaS software scoring model can actually engage and say, "Well, are you in market for our products, and should we continue to put time and effort into closing a deal with you?" Then if all of those things jive and they're correct and it was right, that person qualifies it, and now we have a very solid, very consistent metric at that point that we can then optimize all of the rest of the funnel up and down to.
Adam Kerpelman:
I mean, every step here, you optimize everything above it on the funnel based on conversation to the next step. Some of that is easier than others. For us, the sales accepted step is the important one to build in because it stabilizes being able to say, "Okay, the real goal for the marketing department should be one-to-one." Every MQL, everything that marketing qualifies as a good lead, sales should agree.
Brian Jones:
I like how you-
Adam Kerpelman:
That's not going to happen.
Brian Jones:
... defined that-
Adam Kerpelman:
Not going to happen, but that's the Utopian.
Brian Jones:
To make it even a little more academic, you want it to approach one-to-one because you want to know that you didn't leave some good ones in your dataset that should've been MQLs.
Adam Kerpelman:
You should always be testing. It should be approaching one.
Brian Jones:
You want the occasional MQL to be thrown back at you and said, "It was no good."
Adam Kerpelman:
It's even funny because that happens, and it's represented on our flow chart. They can mark a lead nonresponsive, and it gets kicked back into the nurture funnel up at the marketing step. That's them actually kicking it out as MQL.
Brian Jones:
Realistically, there's a lot of philosophy here, and different teams and different products and different processes will have very different results. No one will ever be one-to-one on this. Something weird is going if you're one-to-one. You're probably more like 20% of your MQLs get converted, and it's actually why a bigger team with more leads has a different type of salesperson. They have the business development rep. Is that BDR stands for, business development rep?
Adam Kerpelman:
Yeah.
Brian Jones:
Where that person gets a quick five-minute call, qualifies something, then pass it off to the sales team. You start to then see the economics of staggering your sales team so that you can afford to test more MQLs. You can afford to test a larger funnel to make sure you're not missing anyone.
Adam Kerpelman:
As this is the Data-Driven Marketing Podcast, we'll gloss over the next steps just because they'll come up if you're trying to get these funnels in place. That's kind of like up through the marketing piece and our reasoning for the handoff. There's another tricky bit in here, which is all of these apps include, or if you have one like HubSpot that has so many different functions, there's also a sales pipeline. The sales pipeline is a completely different thing from lifecycle stages. It's collecting data that may kick back and impact the lifecycle stages programmatically but unrelated to things.
Adam Kerpelman:
In some CRMs, they use terminology that overlaps versus deal pipeline and lifecycle stages, so a lot of times, people from sales will go, "No, but it's not." You just kind of have to keep saying the same thing. Lifecycle and pipeline, completely different things. Pipeline is owned by sales, and it's the phases that a deal moves through, which is sales' version of that same process where they're going, "Okay, this just graduated to the next step of getting them to eventually closing as a customer."
Adam Kerpelman:
We have signup for a trial. Someone's done a demo with them. We've sent them this or that white paper. These different steps that as they move, are sales, the sales department's way of following the pipeline steps and going, "Okay, deals are moving. This is working, blah, blah, blah."
Brian Jones:
All of this stuff, we keep saying is customizable and will be different for every company, and they'll also be very different depending on the product line or the region or the industry that you're selling that product in. At a very large company, they're going to have lots of different flows that correspond with different verticals within the business. Even at our company, we have two primary sales pipelines once you get to that point. That depending on what your engagement in is kind of how what we think you might be interested in or could potentially buy from us.
Brian Jones:
You go through two very different sales pipelines, and at this step, it is. It's what you just said. It's the same concept of steps with many different tasks in between them, but it's specifically the area where the sales, you have your sales team that's an expert at that product and that sales pipeline executing to move a customer through towards a closed deal.
Adam Kerpelman:
Then in HubSpot at some point in the pipeline, you trigger a lifecycle stage change to opportunity, which some CRMs will call it deal. That just means that now sales is doing their thing and for their internal tracking. Other departments can see, this one that made it from MQL to SQL now has progressed to opportunity. It's made it through even more steps of the process, and you're moving forward.
Brian Jones:
This is the step where your pipeline starts to have value because pipeline traditionally from a sales perspective, when your CEO's asking about what's closing your pipeline shows you the potential dollars that could be flowing into the business. In our setup, opportunity is where that gets defined. Now when you look at a pipeline report, there's dollar values in the system. As you expand the other areas, marketing can backfill to this when you start to have enough data. When your sales pipeline is sophisticated enough and you've had data for long enough that you've had consistent trends, you can now take those values and start to push them upstream.
Brian Jones:
Now marketing can assign a value to all the way up to like a click from a Google search ad is worth this much because your data you've gathered on customers and then flowing through process and your success rates is so much so that you can interpret that stuff. At an early step though, that stuff is almost impossible. For the first year of you setting up a system like this, you'll have tons of information, and you can use it to intuit things and make decisions and set up reports and understand how things are going to work. It takes such a long time to get this data to a functional place.
Brian Jones:
It really starts inside of that sales pipeline because that's where the dollar values are really happening and where you have the functional thing that matters for the business, which is a closed deal. It doesn't matter how many people come to your website if no one ever closes a deal. It doesn't matter what that SPC cost is.
Adam Kerpelman:
This is an example of a place where our system is still broken. [crosstalk 00:48:47]
Brian Jones:
Totally, we don't have enough data to [crosstalk 00:48:50].
Adam Kerpelman:
It'll be a thing that's hard to figure out. We can't yet link all of our data up to say, "A lead is worth this much ROI because we close at whatever rate." It's not that we're not closing deals. We're just not getting the data feed off of it right now. We're busy diagnosing why.
Brian Jones:
Volume is important. Our business has historically been an enterprise company. Sales volume, you can have a very large company with not a lot of volume. Starting to get data signals off of the volume of an enterprise company versus a consumer brands t-shirt company, very, very, very different. The volume's higher, so you're going to get normalized trends that you can analyze and use to predict other things much quicker. A lot of challenges there too.
Adam Kerpelman:
Then past opportunity, you got a deal that closes, and now lifecycle stage, you have a customer.
Brian Jones:
Yay!
Adam Kerpelman:
Yay!
Brian Jones:
Most people think you're done now, right?
Adam Kerpelman:
Nope.
Brian Jones:
No, not yet.
Adam Kerpelman:
I was about to say, "Oh, you're not done yet." This is the bow tie. There's more flow chart past that because you need to have processes for supporting the customers, educating the customers, reaching out to the customers, getting feedback on your app, all that other kind of stuff, prodding them for referrals, all that kind of stuff. Then you even have another lifecycle stage past that, which is evangelist. If you do your job well at customer support, bam, evangelist, there's your NPS. It took us that long to get you to data wise having a functional NPS machine.
Brian Jones:
Those last two sections are almost always overlooked at businesses, especially that traditional go back to the 1940s, I'm selling globes door to door. There wasn't a mechanism that was affordable to go back to that customer and be like, "Did you enjoy your globe? Can I get a recommendation of a neighbor that I should sell a globe to?" That concept, it wasn't functionally affordable. Now depending on the type of business you are, it still may not be. A company like ours that is high dollar, B2B software sales essentially, SaaS sales, there are all kinds of digital means, just like in the sales funnel, just like in the marketing funnel, where you can scale that process.
Brian Jones:
You can understand how to sell more to a customer. You can sell more to someone who's already a customer. You can watch their behavior in your system. You can watch their purchasing behavior and find ways to help them do more with your products. At evangelist, there are all kinds of means to do that effectively. You know who people are. You can reach out in a bunch of ways, and it's why you end up having in a SaaS-based business model a whole department called customer success that I think was very specifically rebranded from customer service because we wanted to make a strong point that this is different than how we traditionally looked at operating this part of a customer lifecycle.
Adam Kerpelman:
Damn. That's the funnel. I hope the rundown was helpful. That was a doozy of an episode though.
Brian Jones:
The funnel was good. I'm surprised we didn't trip up more. I hope we're aligned.
Adam Kerpelman:
Well, I'm not. We spent the last three months getting aligned on this, rehearsing. Hopefully, hopeful and hopefully, not too much of a drag for the listener at a longer episode.
Brian Jones:
This was a good in the weeds. This was an operational episode.
Adam Kerpelman:
Absolutely.
Brian Jones:
At least the second half.
Adam Kerpelman:
Which maybe we kind of try to do-
Brian Jones:
Maybe we split it into two episodes.
Adam Kerpelman:
... more and more. Maybe.
Brian Jones:
Because then people can just get right to the meat if they want.
Adam Kerpelman:
See how it feels, a twosie. People know how to work a slider on a phone.
Brian Jones:
I don't know, man. Write up the process and share it in the blog post.
Adam Kerpelman:
Oh geez. Fair enough. Anybody who made it this far, if it's the hour-long version or the half-hour version, whatever, thanks for joining us. This has been the Data-Driven Marketer sponsored by NetWise. I'm Adam.
Brian Jones:
I'm Brian. Keep the focus out there everyone.