August 30, 2014
Demand Engineering: Building a Go-to-market Execution Engine
Marketing is a mix of science and art for which there is no playbook. Every company needs a marketing engine specific to their product, mark...
Personas are the cornerstone of everything we do in marketing. They're the very first thing that every campaign should start with, every go to market should start with. They are really, really the cornerstone. I can't think of a more important skill that every marketer should have. I think that the way that you build the capability to make great personas, is incredibly old school as a craft.
You don't need fancy technology, you really don't need anything other than a pencil and a piece of paper, and it makes it very powerful. It's something that marketers have been doing for hundreds of years. It's really about understanding your buyer. I want to talk about the way I do personas. It is super detailed, and something I love doing. The most important thing to remember about personas, is that they are not abstract.
They are not numbers, they are not anything other than people.
You should be able to point to your personas within your user base, and be able to identify people who represent that persona in a very clear and immediate way.
Here's my persona. I'm the Senior Director of DemandGen for Developer Marketing at Okta. Until very recently, I was VP of Marketing at StormPath. Before that, I was at a bunch of other early stage and growth stage companies. I love marketing quite a bit. I also love early and growth stage companies. It's fun, and I like the hustle. I have an MBA. I am a Digital First marketer, even though we're not going to be talking about anything digital tonight.
I put a picture on all of my personas, because it helps you identify the person that you're looking for. That is my favorite picture of me, because I knit that sweater. I consider myself a knitter, maker, builder, crafts person. I'm 38 for a little while longer, married, and I have always, since I was freed from the shackles of my adolescence, lived in an urban environment.
Interestingly, different parts of this make me attractive to different types of organizations. These demographics are very simple, right? They're things that you could easily glean from my LinkedIn profile. But they're important, because they help identify what I care about, where my proclivities are, other than orange yarn. The agenda we're going to talk about tonight is pretty straightforward.
We're going to talk about how to gather qualitative data that matters for a persona. How to then put together the persona, and I'm going to walk you through a very simple developer persona, based on the categories and the types of information that I use. Then we're going to talk about how you use those personas, basically within a go to market context, and then action steps for what you can go do next.
So the first section is gathering data.
One of the things that I love about persona development, particularly for early stage companies, is that it looks very similar to the way you chase product market fit.
Your market is reflected by your power users. It's reflected by people who are actually very very likely to pay you money for your product. They either should be paying you, or they are very likely to. The earliest way to find these people, is through your power users. This is particularly good when you are very early stage, or when you have very little data. You don't need 150 people building trading data for an algorithm, to tell you about Karthik. Karthik was one of the very early power users at StormPath.
You don't need a lot of data to tell you who he is, or that he's using the API a lot. You need even less data to tell you why. What you need to do is get on the phone with him. One of the things that we use these early power users for is not only to help do a lot of interviewing, and fill in the persona itself, but they're a great place to test programs. I would call him, literally, and say, "Hey Karthik, I'm thinking about doing this thing. What do you think about that?"
It was helpful, because he could tell me whether I was moving in a direction that would appeal to him or not. So it was sort of a feedback loop, as we we refined our product. It was also really helpful, because they're highly qualified. So, when you're looking for that repeatable go to market, that repeatable thing to chase, focusing on the people in your user base, even if it's just a handful who are highly qualified, and then tailoring your go to market off of them, is very efficient.
We talk about that a lot, in early stage marketing, but this qualification is really about understanding very deeply who this person is. The most important thing, is to go through your user base, identify representatives. If you have an API, it's pretty obvious, because these people bash your API quite a bit, they use your product heavily. They might also express enthusiasm in other ways.
People who are enthusiastic on Twitter. People who are active in your Slack channel. People who are really just abnormally engaged. And they're shockingly willing to give you their time. The most important thing is to get to know these people. The great thing is, is that your early power users, they tend to stick around.
The people that form the best part of your personas are active enthusiasts for the product. So you get to work with them, as they grow through the course of your company growing.
Karthik, I think, used the product for three or four years, which was a pretty big growth pattern at StormPath. I think we went from five employees when I joined, and when he started using it, to 35 by the time he got a new job. So that's the first step. But when you're a little bit further down the pike, you have this other phenomenal resource. The great thing about this resource, is that they are highly incented to figure out who cares about your product.
One of the things that I love about working with sales teams, is that they are very easy to work with, when your motivations and your incentives are aligned. I like using this free mark, which we'll talk about in the last section, of "ignore, focus and cultivate".
It's a really easy thing for a salesperson to tell you: Who do they ignore? Who do they focus on? Who do they cultivate? With those, you know the middle section, is where you really want to focus your efforts for building personas.
They can also walk you through the buying motion. Very often, personas don't operate on their own. The example that we're going to go through does, just for simplicity sake. Almost nobody buys any kind of serious software by themselves, particularly enterprise software. These people can really help you understand who is influencing around that persona.
What is going on in the context that they work in. They can help you understand, not just the team around the buyer, but what are the steps that they have to go through in sales? Is this person going to have to talk to procurement for the first time? Is this person going to have to explain what the hell your API does to his CFO? What is the real challenge that this person sees, when they're actually in a deal? What does that mean for how you enable them as a marketer?
They can also define roles, as anybody who's ever done marketing to any size of organization knows, when you have multiple people on a deal, they have different roles. It is harder for marketing to see that, than it is for sales to see that. Sales lives and dies based on whether they can identify those people and what their roles are.
Sales can also tell you what the buy indications are. I like to call these the triggers. What's an early trigger you look for? It could be an event. For a lot of developer services, it could be that they're building a new application, or they're ripping out an application.
These guys are highly incented to make sure that you know everything about how to find those people to focus on, because it makes their lives very efficient. The other thing is, it can put you on sales calls. That's really really beneficial. Being a fly on the wall in a sales call is one of the things that I think many many marketers do wrong. In the sense that they don't do it. It's really easy to completely avoid sales altogether, and particularly to avoid the grinding work that they do.
But it's important to be on sales calls, I make my teams do this. There are things to look for. You look for key words. You can also find these in other places, which we'll talk about, but the words that a person is actually using, that your target persona's actually using, those are important things to write down. Those are important things to include in key messages. They are important things to make sure proliferate everything that you do, which starts with the persona.
They also can identify objections. We'll talk about this a little bit more, when we talk about the actual persona itself, but the objections that you can squash early, tend to come up on sales calls, and it gives you a great opportunity to move things quickly through the sales funnel. You can hear about motivations and excitement, which is always a great way to start campaigns.
The things that people get excited about are hard to see in Google analytics, but it's really easy to see on a sales call.
Most importantly, you can hear about confusion. If there's something murky, in the way you're going to market, you're going to hear it on a sales call faster than you'll hear it in your marketing data. There is an interesting point here, because a lot of us have worked for small companies, where the marketer may be the sales person, or the CEO. They may be some blend, you may not have that separation of church and state, or separation of roles yet.
I think it's really important in that situation to have someone who is a separate third party from the sales person in the room, documenting stuff. Looking for these three things, you give them a list of what to look for. What words stick out to you? What key words stick out to you? What were they excited by? What were they pissed off by? What were they confused by?
It prevents you from having to be distracted from being the sales person. But also, it gives you this really important third perspective on that call. Typically, those people see very very different things, than people who are in the action. It could literally be your office manager. Someone who is just very good at listening.
I think internet research is great, but it's last step. It is very very flat, in terms of the amount of data that you'll get about a person. You can find things like key words on their LinkedIn profile. Key technologies that they use, communities that they're involved in. And demographics. It's really easy to get that stuff from LinkedIn, Twitter and GitHub. They won't tell you as much what your users care about, in the space around your product, and why.
I think it should be the very last step, and maybe just the very seed. But Internet research is not the most important thing. It is, in fact, the least important part of the process. It's an easy crutch. You can do it while watching the Walking Dead.
I'm going to talk about, you've gathered all this data, you've done all these interviews, you've written it all down, you've got copious notes in Evernote, what do you do? How do you go about building a persona? Let's first talk about what not to do. These are two examples that I poached off the internet. My apologies to these humans. The one in green is the one I like.
It's really super detailed. It's got this woman's salary and age on it. You know exactly what someone's trigger pricing is going to be, you know how much money they make. You know exactly where they are in their career, if you know that their age. It's from health care so I don't even know what it's about, but it's incredibly detailed, and it tells you a lot that help you construct campaigns that'll be meaningful for this person.
On the flip side, you have John Tyler, who is a CMO, who is organized, strategically minded, adaptive, confident, and has communication skills. Which just described every single senior leader in the United States, and Western Europe. You just can't get a job over manager, without any of those skills. It tells you literally nothing about this person, that is meaningful.
I see a lot of these very lightweight personas.
The challenge with personas is they take a long time and are very labor intensive to build. The trap is that you want to do something quick and easy, but it's effectively a useless exercise.
What I do love about them both, is that they both have pictures, and that the not so great one focuses on goals, which we'll talk about a little bit.
One of the things that is a demographic for developer services, is what languages and frameworks do you plan? Your life is completely different as a developer based on which of these communities you're a part of. I don't think about it as work content, I think about it as a demographic. Career history is super important. This is where LinkedIn spelunking becomes beneficial. There's a really different experience between selling to someone who has very little experience within a business, very little experience as a employee, very little experience in the work force, to someone who's been in the work force for 25 years.
The role in the sales process is really about where they fit in, and who they're influencing and how. The typical roles are: you're a buyer, you're an economic buyer, you are an influencer, you're a champion. There's one I like to call the introducer, which is that you lob stuff in. I like those people, but they tend to not last very long in a deal. Exploring how this person is participating in buying, if they're buying by themselves, or if they're buying as part of a group.
That's important to know in a persona. Particularly because it changes the way that you build programs for them. What are their goals? This is, I think, the real heart of it. The goals and the challenges. What are the underlying motivations for this person? Their goals are things like, "I want this to expedite my time to launch my product". That's great goal for a developer service, it's probably the goal for just about everybody buying a developer service.
It's important to understand not just what their goals with the application or their goals for the engagement are, but what are their goals for themselves?
How can we support that? Those are the kinds of business relationships that become very sticky. The other thing is challenges. We always like to say, it's better to be a pill than a vitamin. Getting to the really painful challenges that people experience, that matters a ton. Digging into, what is this guy running into? Both within his the context of his application or his employment, but also, what are his challenges in his job right now?
It could be things that they identify, but in this case, a typical one that we saw at StormPath, was some people aren't super expert in advanced use cases or advanced security topics. So they don't know what's going to burn them, and then you get a lot of value out of working with somebody who does. That's very beneficial to understand, that one of your challenges can be something you don't even see, but that we, as a company, see.
But it could also be a person who lives in a part of the country, or a part of the world, where they're not in San Francisco, surrounded by people in their language community. Providing them opportunities to work together delivers a lot of benefit for them. I don't like to overthink these things too much, because they're already pretty detailed. This would be all about half a page, because it's just a big Word document. Other stuff they care about could be things like they're highly cost sensitive.
This guy is funding it out of his own pocket, and so, knowing that and demonstrating sensitivity to that in the marketing and sales process, is beneficial. Two, he's going to be highly cost sensitive, so it informs your pricing. A second section I look at is, how do we help you?
So who is this person that we're selling to? And how do we tailor our approach? How do we approach you in a way that is helpful? Because this is where we're going to start seeing that relationship and that pick up. It means that the go to market strategy becomes a come to us strategy. The first is, how do we help? It can be really straightforward things, like, the example I use here is that we provide a key application function that this guy needs, but isn't actively available in, or natively available in Angular.
We offer him something he needs, pretty straightforward. We expedite his path to launch, because he wants to build the greatest app, in his spare time, that was ever built and rock the world. Those narrow, and very concrete, but also broader, and more visionary ways that we can be helpful, do inform our programs. Common objections. I love this, because it really helps everyone who is addressing this person, whether they're in UX, or product or whatever.
It really helps them identify the things that matter. This guy needs to have whatever feature on the free tier, so he can evaluate. If we don't have that on the free tier, he isn't going to use our product. He isn't going to evaluate our product. Price sensitivity might go here as well. Common objections might be, you're a tiny company and I've never heard of you, or you're a huge company and I'm terrified by your hugeness. It could be all sorts of different things, but it's going to depend on the person. Our programs can get out ahead of that.
Questions are things like, when you're doing an evaluation, when this person is doing an evaluation, what are they really asking themselves. A lot of times it's how do you compare against something else? What's the benefit relative to using some open source plugin or module? Is there free Slack support? It's things that they're looking for, and understanding what they're looking for, what questions they're asking can really help you get out ahead of those things.
Who are they influenced by? This guy might be influenced by all the people who are on Angular. They could be influenced by their CTO. They could be influenced by all sorts of other people. Catalysts. This is really important, because typically, everybody has a thing, when they're evaluating a product, that tips them over. The catalysts are the thing that tip you over.
In this case, it's like this guy ran into a limitation with whatever class in Angular, or however Angular implements some particular feature, and it limits his development, and that's when he goes looking. Understanding that pivot, that movement, really helps you know how to equip his first introduction to you. Typically, we would have multiple catalysts, multiple objections, multiple questions. These would be big.
The stuff I love is really about how do you dial in, for everyone who's using this, into what the persona really really cares about? This is important because it makes sure that everybody keeps their eye on the prize. This guy, he wants fast onboarding. He wants it to be under an hour to really evaluate the product. He loves the CLI. He wants a particular feature to be super intuitive.
Making sure that those become a little bit sacred, or at least are identified as sacred to that person, helps keep us in check about what are the trade offs, when we start going near them.
If we're not going to support the CLI, or if we have to complicate the implementation of this feature that this persona's super passionate about. This would be another about half a page.
The last one, which is most important, but it's the shortest, is "what do I want or what am I really targeting this person to say about me in a context where I'm not involved?" When somebody's asking me, "oh hey, did you try this service." I called it Super Dove service. So you tried Super Dove service, what do you want them to say? In here are some really important key words that help you understand where the crux of the value is.
How do you whittle a page of details down into the stuff you cannot abandon? The stuff that is really important for this person. It has to show up, either verbatim, or roughly verbatim, in every single thing you do to that person. You write them an email, "hey, I wanted to introduce you to Super Dove service, we're a super easy to use Angular module." Dialing that closely into the language that they used, their key words, their things, it really helps tie this into the programs in a very concise way.
This part of the persona, tends to be some of the most time consuming and contentious, because it's about what is really critical.
Using personas. We don't build these things in bubbles, or for fun. I do, sometimes, for fun. The way I use personas, is that they are all personas, and then I identify the ones that I'm going to ignore, which we'll talk about. Then, identify the ones I want to focus on and sell to. Whoever's left over, is who I cultivate. These are very different activities, with different goals, for each group. It's the most basic segmentation you can do.
Who do you ignore? These are people to sweep from your qualification funnel, really early. Common patterns are geography. There may be places where, because of the cost of development, the local cost of development, your product is not competitive as a service. It could be that the hop from wherever your server is, if you're really early stage, just doesn't make sense, it's too long. It could be a company size, and it could be a company size that's either too small. They can't afford you. Or too big, you can't support them.
A lot of early stage companies fall into the trap of trying to support companies that are bigger than they are capable of doing. Similarly, they try to build a business model on companies that are too small to pay them enough to keep the lights on.
That's a particularly hard one. Programming language. Whatever you're doing may not be compelling to certain parts of, or certain demographics of the developer community. This was certainly true at StormPath. There were languages where we were absolutely gang busters, and other languages where we were less gang busters. Figuring that out took time and was difficult, but it was critical, in terms of how we went to market.
Certain industries are ones you might not want to go after. I remember when the government cut funding for solar panel manufacturing, everybody who was going after that market, stopped going after that market. Wisely so, because those people had no money to spend. Job title. One of the things that can be confusing, are you selling to developer, or are you selling to IT? They're different organizations, and I think that it, and they are approached very differently.
They have different sales motions. Lots of different things. Someone once said to me, "Oh, do we want to talk to this marketer?" and I was like, "No, we don't want to talk to that marketer, ever." The goal with these people, is not to just jettison them from your funnel, not to ignore them. It's you show them the love, but you don't spend a disproportionate number of resources on them.
These people can be tremendous advocates on Twitter, they can be additive and give you feedback on your product and on product releases. They can participate in your community, they can even participate in your support channel. They have quite a lot of benefit, but I would not give a single one of them to the sales person. They should fall out of the sales funnel.
Focus. This is where your personas come into play. You've identified your the key people that you're really going after. There might be like two or three of them. What you're looking for, is they have some qualification. BANT is a very common qualification mechanism, it came out of IBM. Budget Authority Need and Timeline. Typically, that's what a sales person is looking for, but you can see some of that stuff as a marketer.
You should look for it. These are the people to focus on, if they have a need. They have a clear event that just happened. Maybe a great need is when a security issue is identified in a framework. That's a great time to sell to the people using that thing, because they might want to go to a service where it's more incumbent upon the service to maintain security. They might also have timeline. One of the things that I really love, is the aircraft carrier landings. People who are launching products tend to land quickly. They tend to make decisions quickly, they tend to have a lot of people standing behind them breathing fire. And those people go quickly, which is great.
Budget or authority, those two things you can see just from a little bit of marketing data, really. If someone's landing and repeatedly showing up on your website, they may have some urgency to their timeline. Depending on what landing pages they hit, what visits are coming in through Paid Search, or Organic search, you may get a sense of what their need is. Budget, unless you're using more advanced marketing tools, can be hard to identify. It tends to be correlated to company size.
Authority tends to be correlated to a title. You tend to need to have some slightly more in depth analytics, or in depth data around those visits. The other thing here is to look at the goals. What do you want to do with these people? Obviously, score them high to expedite them to your sales funnel. Make sure that you are spending the bulk of your money on them. Make sure that you are spending the bulk of your time on them, which can be hard in an early stage company. Give them a clear way to contact sales. If somebody's coming in with a timeline, giving him a fast way to fix that problem, like a Contact Us page, is very helpful.
Then, whoever's left over. We've gotten rid of all the people we're going to ignore. We've focused our energy on the people who we want to focus on. But then there's this group who may be the largest portion of your list, who don't have an active project. They have no clear need, they have no clear timeline. These people are important to have personas for as well.
They're not your focus personas, you may not want to spend two pages or a page and a half or a page or whatever on it, but you should have a clear definition of who they are, similar to the ignore people. These are people who you want to stay top of mind, you want to be known as a resource for them, so that at some point, when a need or a timeline appears, they think of you first.
So what a lot of companies see is this bimodal distribution of when people show up, or when people take advantage of an offer, they sign up for your API, or they start using your API, or they convert, or they reach out to sales. These are people who are really qualified, and then people who aren't quite qualified yet. Cultivate are people who aren't quite qualified yet.
Because if they're not qualified yet, they're ignored.
Cultivate Personas are important to keep close to your heart, close to your nurture, close to perceiving you as a thought leader, making sure that you're regularly going back to the well to check in on them.
Which could be just a simple message like, "hey, I know you checked out StormPath six months ago. Can you let us know if there's anything you need from us at this point, or anything you're interested in? We've got these updates." They may have also missed something. It could be that they just don't know what they're missing. Sometimes an email about an obscure feature can provoke these people out of the swamp, or a feature release.
It's really important. Your list might be tiny to start with, but your list will get huge, and you'll become overwhelmed with work very quickly. It's important to be ruthless. To know who you are going to ignore, and come to peace with that, and find a way to show them the love that you feel good about, so that you can focus on the people who are going to make you money. Because the key personas are the ones who are going to build your business.
The people to cultivate, while important, and while it's easy to go spelunking in that large pool of data, they are not going to make you money in the near term, or within a timeline that is going to keep your business alive. I think it's hard when you personalize who you're going after and what you're doing with them. To remember that you also need to be incredibly ruthless about where you spend your time and energy and dollars.
Some things to do today. Review and update your personas. This can be hard to do. Identify three obvious ignore qualifiers, everybody's got them. One of the easiest ways to do this, is to ask a sales person, who should I ignore? Who do you want me to ignore? Who do you ignore? We had this amazing moment early on at StormPath, where something like 30 or 40 students from the University of the Andes signed up in one 24 hour period.
We were like, "is this a thing? Are we a service for students? Is this going to be a thing?" We had to be really ruthless, and realize that no, these people were never going to pay us, but they actually asked some great questions and helped us clarify a lot of our onboarding. The other thing is to know your lead time. It's really hard to differentiate how long, or whether someone who's a cultivate, or a focus, if you don't know how long it takes to interact with them.
As an MQL or an SQL, a Marketing Qualified Lead or Sales Qualified Lead. Once you understand those timelines, it's really easy to differentiate those people. Build a cultivation program. A good benchmark for a nurture for someone who isn't qualified, is a 12 week nurture. You can pause it, you can turn it on and off, you can do it in Mail Chimp, it's pretty easy. Expand your personas. If you go back and do a quarterly dumpster dive and say, "hey, we weren't really seeing much in this demographic, but now it's picking up and it's picking up fast, what did we do? What changed here? Those are things you can do today. These slides will be available, in case you want to go through all the things. I'm email@example.com, if you ever have any questions, I'm happy to answer them. I'm obviously deeply passionate about personas, and thanks, everyone, for coming.