June 26, 2014
PagerDuty’s Favorite Dev Ops Blogs
PagerDuty’s Ops Engineering team lead Arup Chakrabarti knows how to scale complex architectures. During his time at PagerDuty, the company...
Thanks so much.
Really excited to be here with all of you.
I appreciate it.
I will say upfront that I've got some bad news to start this. I
normally start with an interpretive dance about customer success,
but as you can tell-- I don't know if folks in the video could see, I just
had knee surgery on Thursday so I won't be moving around as much as I normally do.
But I'll try to bring the same energy and hopefully give you some takeaway value
By the way, since you're going to ask later, it was knee surgery for an
ACL. No, you don't want to do that yourself.
You can watch other people do it, but don't do it.
It's from jumping on a trampoline with my kids
So, we can we can have a follow up discussion about that and why you should never jump on
a trampoline in the future.
But today we're going to really focus on customer success, and I want to say upfront, we're going to focus on the broad definition of customer success. Which is not just about a CSM team. How many of you folks in the audience, raise your hand if you're in a customer success role? Some folks in the room are, and some aren't. One of things we'll talk about is how customer success is really a company-wide need, and there's a customer success team that has an important role in that. We'll talk about some best practices. We'll also talk about what the other groups need to do in a company, and particularly what product needs to do. I talk a lot about that, and we actually recently ourselves got into a second business which is basically helping product managers do a better job instrumenting their products to see what users are doing, and to engage users in the product, and onboard them better and things like that. One big thing we're going to talk about is basically customer success across the company.
But to open up I want to talk a little bit about the back story of customer success, but also why I'm doing this, why I chose to come a few days after surgery and do this talk. Why I'm so excited about customer success and why I do this every week, and there's an interesting little personal narrative that I think ties to what's happened in the industry in general. So, I'll give you the backstory. Backstory starts with me as a kid. This is not actually a photo of me as a kid, but it easily could have been, I was definitely a hardcore nerd programmer and things like that. My dad was in the tech business, he was actually at some old tech companies and then he was the CEO of some small tech companies, and we have this photo of me growing up as a kid that I remember really fondly which is a photo you can see here. That's me on the right with the K-mart three piece suit on there, with a briefcase filled with Legos or something like that. I don't know what's in the briefcase. This is my dad in the 1980s taking me to a Take Your Kid to Work Day. The reason this is a connection to customer success is because I remember my dad, this is a memory from 30 years ago, my dad saying "If you go into business on this Take Your Kid to Work Day, I want you to learn one really important lesson."
He said basically, "In business, you
either do one of two jobs, you're either the
person building the product--" Picture of Bill Gates there
on the left, "Or you're the person selling the product."
Picture of a cheesy salesperson on the right.
Those are the only two jobs that
matter. In the traditional business models, before
cloud and SaaS and developer empowerment,
that was really true. I
businesses it really was about building and selling.
Those are the only two things.
You had engineers and salespeople.
One of these I'm hoping to talk to you about today is how you still need to build
great products, no matter what you're doing, and you still need to be able to sell and
market. That's really hard.
But I think there's a third discipline that's super important to a company, which is
driving your customers to success, which means getting them to
onboard and adopt, use and get value and eventually
That's a third discipline that most companies don't know.
That's why I've done hundreds of these talks, to try to help
explain to companies "What's that third new skill you have to learn?
You know how to build stuff, you know how to sell stuff.
How do you drive success?"
That's what we'll talk about today.
So, I want to put it in context, why did all of this happen?
What disruptive activity caused customer
success to be the third discipline in companies?
I'd argue it's actually really 20 years ago, it's the internet as it got
started, that basically shifted the power.
In older businesses, most people here
luckily never worked in on-premise software, but probably maybe some of you did. In
the old the world of on-premise software where people bought stuff and they installed it
in their own computers, the reality is that the customer didn't have any
power. Because they bought it and they paid for it and they installed it, and at the end
of the day whether they used it or not was their problem.
That's the old world.
But the internet disrupted everything by pushing all the software to the
cloud, pushing everything to a SaaS or subscription business
model, and then giving that customer all of the power.
As things moved to cloud and SaaS it created
this unbelievable empowerment for customers.
Even if you don't know anything about customer success, all of you have experienced that empowerment yourselves if you just think about music as an example. I'm old enough to have stood in line at a Tower Records to buy a Vanilla Ice CD. I'm sharing some stuff about my childhood, and the fact that you had to save up the money and you had to go to the store and you had to see if the CD was in stock. Obviously nowadays kids, with Spotify and Apple Music, you don't even think about getting music. Technology shifted all of the power in the music industry from the producers of music to the consumers of music, and what's happened in all of the businesses that folks in the room are in, whether it's a developer tools business or whether it's a cloud infrastructure business, or something else. The technology has shifted the power from you to your customers, and that's caused you to say "OK we need a third discipline. We need to figure out how we're going to make sure our customers are successful. It's not their job, it's our job." That's what we'll dive into detail on today.
Throughout this we're going to talk about this concept that we're fundamentally in this age of the
customer, where it's not just about sales and not just about
product, it's about understanding that that customer is really at the center of
what you do.
Now that's a platitude that you can say and not make real, so one
of the things I'm really going to talk about is, "What do you do to actually make that real
in your company every day?"
Raise your hand if your company has less than 10 people in it.
A few people are early stage, which is super exciting.
Raise your hand if your company has, let's say, 10 to 100 people in it.
Awesome. Maybe venture-backed, things like that.
Anyone in more than 100 people?
Three people, great.
So I'll try to personalize this, when you're less than 10 people you don't have a lot of different roles, you just need to have the right mindset early on. When you're 10 to 100 you're starting to specialize a little bit, you might have a few CSMs and things like that, and you're 100 plus so you're probably thinking about how you scale. We'll cover all that as we go through the talk.
One of the other things I would say upfront is this idea
of the customer having a lot of power and things like that, it doesn't happen
in every business right away.
You might have a business where you're the only way for your customers solve
Does anyone have a business where you're the only
way, and they just can't help but buy it and they just use it right away and you don't
have to do anything?
I was looking for companies to invest in.
I didn't see any hands raised in the
room, so the reality is all of us
have to work for our customers.
I always joke that not every industry has reached this yet.
There's some industries where you can tell that they haven't yet gotten to the point
where the customers have power. Put
up United Airlines as an example here.
You can pick your favorite consumer brand that you don't like and
say, "In that industry they don't treat me well."
The reality is there's some industries where you don't have choice. Maybe your cable company, maybe your telco provider, maybe your airline. But in industries where the customer has choice, the reality is customer success is not just a "Nice to have," it's really existential. So what we want to talk about is this idea that it's existential and really try to define it precisely. So this equation that I'm putting up here represents two things. Number one, that I'm super nerdy and I love equations. I really do love equations, and I like to simplify things into equations. But number two, it's really helpful to define customer success because a lot of people just say the word and don't really explain what it means. It's pretty hard to understand what another company is doing to think about it. We came up with this very simple framework which says in a business product-- I'm assuming most of you sell to businesses.
When a business buys something, whether it's software or whether it's a device, they're buying it to have an outcome.
That outcome is to ship code faster, be more
secure, generate more leads, generate more sales.
They want to have a goal for what they're trying to do, that's the outcome, and we'll
talk a lot about that.
But they also want have a good experience, they also want to make sure that when
there's an issue you're helping them quickly, that you're nice to them on the phone
and your emails are friendly.
Things like that. They
want a good outcome and they want a great experience.
Now, why do we explain it this way?
Because a lot of people when they hear "Customer success," they think
"Customer support," and when they hear "Customer support--" Because we're all consumers,
they think "Make your customers happy."
Because when we're consumers, at the end of the day what we really want is just to be
happy with the hotel or the airline, or whatever.
But in a business context, this concept of the outcome is so important and so many customer success organizations measure, for example, net promoter score. Which basically is, if you don't know, it's a survey of how likely your customer is to recommend you, and that's great. That tells you about their experience. But what's interesting is many studies have shown that net promoter score in most B2B products is not correlated with retention. It's not actually correlated with the likelihood of somebody to stay with you, because what happens is for a B2B company, just because they like you and just because they're having a great experience with you doesn't mean that they're actually going to stay with you. If you're not delivering the outcome, if the person that you deal with every day on the phone loves what you do and comes to your conferences but their boss isn't seeing the ROI, at the end of the day, especially when that person leaves you're going to totally get a churn from that customer. So what we want to talk about today is, how do you operationalize both the experience which is making them feel good but also the outcome? That's what we'll dive into as we go forward. As we get to the end please think of questions about how this translates to your own business.
Now, one way to think about the framework is to say "OK.
If I think of all my customers, and I think of the ones that are having
great experiences on the vertical axis on the top, and bad
experiences on the vertical axis on the bottom. O
n the horizontal the outcome they're getting, which
is great outcomes on the right and then bad outcomes on the
left. So you can think of your customers, if you have 10 or 100 or
1,000, put them into those four quadrants.
The lower left are the ones that are really tough.
You might call those your red accounts.
Those are the ones that aren't really using your stuff and also
aren't happy about it. That's
where you end up spending a lot of time, naturally.
The upper left are the people-- This is actually really dangerous for startups.
They're the people that are super happy and will speak
on your behalf and will be a reference and talk to a VC for
you, but they're not really using your stuff well.
Anyone want to be vulnerable and say you've got some of those?
Everyone has them, they're on your
website and they're totally a friend of the
company, but if somebody dug in and said "What outcome are you
getting from using this?"
They wouldn't have a great answer.
Then there are the people on the lower right, this is more common for big companies, where that customer is stuck. They're using it, they're getting some value, but they don't like working with you. When you try to introduce something new or get them to upgrade to the new version they won't do it. Then the upper right, which is basically where you want people to be. They're having great outcomes and great experiences. If I took away nothing else from this talk, all the way to the upper right, and to always be thinking about the outcome and the experience. One action item as you leave this talk is going to be "How are you measuring the outcome in your business, and how are you measuring the experience?" Experience could be a net promoter score, lots of different ways to do that. Outcomes can be specific to your business. It could be adoption, it could be the ROI they're getting, it could be the number of security vulnerabilities you're finding, things like that.
Now if you click into that in more detail and say "How do you make that work in a company?" This really applies, especially as you go from 10 to 100, 100 to 1,000 people in the company. The reality is that you've got a customer success team. It sounds like almost everyone here has a customer success team, or is building one. You're going to build a customer success team, but if you think about it, that team doesn't control the whole outcome and experience for the customer. Because the outcome is going to come from everywhere, all the way from marketing, "What was the expectation on the value prop when they were marketed to? What was the ROI that the salesperson claimed they'd get, if there's a salesperson involved? What was the onboarding experience like? How did we do in actually billing them?" Sometimes you can actually really piss off the customer through billing them. That experience in outcome can really range the entire company. So one of the things we like to do is say "It's not just about the CSM. It's about all the different departments in the company."
Fundamentally one of the problems in most companies is that every department is thinking about their own area. Marketing is really focused on generating leads, and sales wants to close the deals. These are all great things. But there's no group that's looking at this from a customer perspective across that whole experience.
This is the fundamental idea behind customer success, is you
flip that concept and you say, "OK.
Let's look at this from the customer's lens.
They don't care whether they're dealing with marketing or sales or support or whoever,
they're trying to get to an outcome with a great experience."
Customer success fundamentally is about flipping this
model to actually look at it horizontally-- I'll
come back to that in a second.
One of the other concepts we want to talk about, is
the idea that the only way you're ever going to coordinate all
those groups to drive towards that outcome and experience,
is to recognize that you're never going to have one group own the whole thing.
So even if you're in customer success, your job isn't to do
everything on the chart.
You can't possibly control the whole sales experience, and the product,
and the marketing, and sales.
You can do your best, but you can't own the whole thing.
So the analogy we like to use is an orchestra. The orchestra concept is that in an orchestra you don't have one group driving the whole orchestra. You have a lot of different musicians, violin and the oboe, the xylophone, all playing their own instrument really well. Then you have sheet music that keeps everyone in rhythm and your conductor that makes sure that all the timing is right, and things like that. So the same concept has to happen in companies. If you're in a customer success role, you've got your own part of the equation, which we'll talk more in detail about. But part of your job is to help create the sheet music for the whole company, to keep everyone aligned around that journey. We'll talk about it in detail, but the best customer success managers, they're not just making sure their customer renews or adopts. They're helping to figure out what's wrong with the product that's causing them to have an adoption process and a problem in the first place. They're helping to improve the support process and help sales sell to the right customers. This concept of an orchestra is really important because it's never going to be one group that solves the whole problem we'll talk about.
So, we wrote a book on customer success, I recommend you read it. It's
become pretty popular, 60,000 copies sold and apparently not
all of them to my mom.
So maybe some other people read it, and actually it's now available in Chinese
and Portuguese and Spanish, and all that. It
just shows you how wide the need is for customer success.
In that book we talked about 10 basic laws, and so these are
10 concepts I'll share with you here.
They're good things you can action inside your company.
The first one is that customer success is a top-down company-wide
commitment. It's pretty obvious, but you won't get to
customer success just by having a few customer success managers
. You'll get there by your CEO, your founder-- If you're a founder in the room.
You yourself, or you convincing your founder to say
"Customer success isn't just a job.
It's like a third discipline in the company, and we need to build stuff, we need to sell
stuff and we need to be able to drive success."
That's one big thing, and we can talk about strategies for that.
Second big thing which some people in the room have control over, others don't,
is really getting the point where you've gotten really good at selling to the right
customer. A big chunk of customer success problems are
people that bought stuff where they're not a good fit.
Raise your hand if you've ever dealt with a customer that is not a good fit.
If you're in CS you're downstream from that problem, but how do
you actually get to the point where the sales team is really using that knowledge to
target the right customers?
Because I think we all know in a SaaS business no matter how much
that sales rep makes on that deal for the company, getting a
customer and then losing them a year later is the worst value prop in the world.
It makes that customer unhappy, they're not a reference, you don't make
money on them because it usually costs a lot of money to get that customer.
So, selling to the wrong customer is like one of the worst things you can do and we can
talk about strategies for that.
Third thing, which is probably one of the hardest realities to accept,
many people in this room are working on early to mid-stage
startups and you're 10 people, 20 people, 100 people.
You are probably actually disrupting somebody that was in a room like
this 5-10 years ago in their early stage startup.
You've got the next best security product, or the next best fintech product.
There's going to be a next best version of you two or three years from
now, so the natural tendency if you don't do anything is your
customer is going to leave you for the next best thing. That's
the thing that's happened, is shifting the power to customers.
So this never ends.
It is like pushing the boulder uphill a little bit.
It never ends and you've got to keep working to make your customer
successful. It doesn't just get easy at some point, it's always going
to be something that'll take work.
Number four, as customers become more
aware of the SaaS model, and especially IT buyers--
How many people, raise your hand if you sell to IT?
How about those who sell to developers, and then
security? OK, so those three groups in particular, and
marketing as well, they bought enough stuff that they've
seen companies do customer success.
Company A comes to them, sells their stuff and then there's
somebody showing up and saying "Here's how you should be using it," or "I notice you're
not using this feature.
Let's do a review of your stuff."
Company B sells this stuff and is like, "Call us if you need help."
Company B is not going to get there, and company B is not going to get
more expansion because there's an expectation set of
what to expect from a vendor in this new world
Number five, one of the things that's really important and actually
historically wasn't well done is getting metrics oriented around
this. Most companies don't do a great job
measuring the customers.
I mentioned the outcome they're getting, the experience they're having.
Customer success is often just throwing people at the problem and
"Let's make them happy."
It's very intangible and that just doesn't work.
Salespeople, one of the great things about them is super measurable.
"What's their bookings, what's their quota?"
Customer success if done right is super measurable.
Number six, a lot of companies-- One of the reasons they don't do this
is they're like "We don't really need this stuff because we've got a great relationship
with Suzanne or Sheila or
Sam at the customer, and those folks at the
customer are going to be with us and we've got that great
relationship so we don't need to worry about it."
But the truth is, if anyone here has been through sponsor change, which is
your customer or the person that leaves you, that personal relationship
is worth nothing once that person leaves you.
The rate of change is so high in jobs that just
building the loyalty through personal relationships doesn't work anymore.
If you talk to salespeople who sold like 20 years ago, things were much
easier, honestly. You sell to somebody, you golf with
them, your families go to the same schools .
Nowadays it's so much more about the outcome and experience and so much
less about relationship, and that's an important message.
, product is the most important thing, so at the end of the day
if you're in customer success, sure you can onboard the customers better and make them
The most important thing is a great product that you can track what
people are doing that's engaging people, onboarding them well, and
actually if you had an extra dollar to spend it's usually better to spend making that
better than almost anything else in the company.
Number eight, I think most people know this, but when you sign
a deal if you have a sales team there's that moment maybe you ring a
gong-- A DocuSign goes through, or whatever people are
celebrating, champagne being popped or whatever.
The customer is not celebrating, they're not like "Oh my God we just DocuSigned
This is amazing."
For them, they've started a stopwatch.
For you, you're like "OK. Time
to go on vacation."
So there's this mismatch, and what people
in this business get really good at is recognizing that time to value
is just so important.
Number nine, obviously in the early days you will have churn. Just don't feel bad about it, it's just what it is. Actually, Jason Lemkin the head of SaaStr talked about it well, which is
"Early churn is totally fine as long as you're learning from it."
So I have some recommendations on a lot of things you can do.
Getting third parties to call your churn customers, obviously
having a root cause analysis of why the churn happened and things like
Then ten, in general just across all of this being more metrics and data driven, and we'll talk through some examples. These will come up as I go through the talk, but this is a good takeaway list that you can remember, it's on our website as well.
One of the things I want to dive into is this idea that product is one of the
most important things. I'm going to share a
story actually from our own experience at Gainsight.
Gainsight, if you don't know, was venture backed, but
6.5 years in and about 650 people in the company
raise $150 million capital and has just as many problems as all of
you, or maybe more. All the stuff just gets more complicated as it
One of the problems we've had is not always
recognizing that some of the things need to be solved between
customer success and product together.
So what we made a big push on this last year and a half, and this is a
little story I'll tell you what really happened in our company.
Allison Pickens is on the left of this slide
and she's our Chief Operating Officer, and she basically is customer success underneath
her. Karl Rumelhart is on the right, he's our Chief Product
Officer. Two critical people on Gainsight.
We had this conversation about a year and a half ago in a management team meeting
where we were talking about product adoption and "How do we drive more adoption of
features?" All the same stuff you talk about. Allison
came in and said, "OK great.
We've got this methodology for tracking adoption of our customers.
We look at the breadth," meaning how much of the product they're using "And the depth,"
meaning how many people are using the product.
A simple way to look at it. Just color
code those, straightforward things. Then Karl said well "That's great.
But actually, I also have two metrics that I look at."
He had a breadth metric and a depth metric that were totally different.
I had in a customer success company, my product team
and my CS team trying to analyze the customers from totally different
One of the big things you can do is get your product and CS team on the same page on how to look at your customers. It's really critical to do that. In our experience, we found that product and CS teams fundamentally want the same things. They want the exact same goals. Just think about some examples, product and CS teams, they both want to scale. Nobody wants this stuff to be super manual. They want to scale the business. As an example, Allison is thinking about "I want to scale. How can I understand adoption more deeply?" And Karl, the product leader, wants to understand how we can drive adoption of new features. Allison wants to do a better job tying that adoption to business outcomes, and Karl wants to understand the outcomes his engineering team is driving. Allison wants to do a better job getting feedback from customers, and Karl wants to do a better job getting feedback from customers in the product, or whatever. Allison wants to do a better job onboarding without having an army of people, and Karl would love to do that as well. Allison would love to do a better job aligning customer needs to the roadmap, and so would Karl.
It's not like product people and CS people want different things. I was a product manager starting my career--Actually, somebody in the audience here was there with me at a company a long time ago and I remember the people in support and sales and stuff, they all wanted the same goals but they just spoke a different language. We'd be thinking about releases and backlogs and they're thinking about individual customers. So one of the biggest things you can do if you're in CS is to try to form a better common language and working relationship with product. As an example, what's not helpful is to go to a product manager and say "Here's my 300 things my customers have asked for." That's the way to guarantee not getting invited to another meeting with the product manager, because they're just like "OK. Great. I'll add it to the 2,000 other things that we have in the backlog, and we will get it to them in the next 35 years."
But what's much more helpful is, "Here's the three use cases that are coming up over and over again. Here's the business outcome that's affecting our customers, here's how much revenue is at risk if we don't solve them."
Really being thoughtful about how you're sharing feedback, I think you can make a huge impact in collaborating. So that's one really practical tip, if you're especially an early stage company.
Now, I wanted to give some tips that I think are maybe a little bit less obvious and a little-- You could go do each of these individually. So there are five tips I've got from a product-CS lens and five from a pure CS lens. I'll run through those as well.
These are very much designed for startups.
Five no-brainer things to do. If
you're not tracking product usage for your customers, if you're some SaaS
application, that's crazy. You absolutely
have to do it, and you should do it right away.
There's a million different ways that we have tried to use, there's a lot of other ones
to use too.
There's some that are free.
It's inexcusable to not understand what your
customers are doing in your product when your product is in the cloud.
It's ridiculous, and you shouldn't be doing it through some really
basic Google analytics or whatever, you should have some sophisticated way to do it.
It should be core to what you're doing in your company.
That's a really simple thing that you can go do.
Second thing you should do very early on, now companies are trying to
do in-app communication.
They're saying, "Even before I hire an army of CS people, let me try to do
as much as possible on the product," meaning onboarding new users
when they log in
again, show them the new features, everyone's doing this.
If you haven't done it you should do it.
There's a bunch of tools, Gainsight has one, there's a few other companies out there that
These things are relatively cheap and easy to do.
So, another important thing to think about that allows you to really
scale from the beginning.
Number three, this is maybe a little more sophisticated.
We created at Gainsight a very organized
and simple process to get feature requests from
the CS team to the product team.
It's called "The product risk process," and there's a bar
which-- It's not just every request anyone's ever asked for.
It's a request, they're called "Risks," that would either block a
sale or block an onboarding, block
major adoption or affect a renewal.
You can define your own criteria, but like a high bar, and
the CS person has to fill out "Why does this matter?
What's the outcome in revenue.
Which customers are affected?"
Then this product team has a SLA on responding to it.
As a measure of success for this process as an example,
you can imagine as a CEO anytime a process is working well you
don't hear about it, and when it's not working well you hear about it all the time
because people complain about it.
This process when we implemented it at Gainsight?
Literally I've heard one thing in my two years about Product C
S alignment on road map.
Which for most companies, that's a really big challenge.
So I would rbuild a form, whatever, a Google form, it doesn't
matter what you do it in.
Build a form, CS people fill it out, it goes to product, there's
an SLA for when they respond, you review it on a weekly basis.
It's a no-brainer to do that and it doesn't cost any money.
Really simple to do it.
Number four, community.
Raise your hand, folks in the room that have a community already, an
online community that you've built?
This is something we did pretty early at Gainsight, and I think another
no-brainer. There's a bunch of cheap tools out there.
You build a place where your customers go,
obviously they can help each other in terms of support issues, but one of the biggest
things you can do is do scalable customer success.
You can launch adoption campaigns, you can get product
feedback. Our community is very vibrant for
Gainsight and all of our feature requests come through
there, directly from the customers.
They upvote things like that, there's tons of good software for that.
Again, it's a relatively cheap investment, and these kinds of things if you do them
well then later on you're not hiring 20
CSMs and tons of people.
It's like you put a little bit of time into it now and it makes a big difference.
So, the last thing on the product side, another little tip that we did, one of things you might want to do if you're in the CS role is get your product team more exposed to customers. That's like a universal need. It will help them build the product, help them understand which customers are the right fit, things like that. For us at least, recording customer conversations is such a home run. Some of you probably do it already. There's a bunch of tools out there, Gong.io is the one that we use. There's a lot of that. I have no horse in that race, I don't really care which one you use, but I highly recommend you record those CS and sales conversations.
Our product managers, what they do is keyword searches, and they'll listen for the customer mentioning their feature, either good or bad. They'll listen to the snippets of those calls, and then they can hear every week what's going on with their feature, what people like and what do they not like. You get that first person view that's very different than an email that's summarizing what the customer said. You hear it in their own words. I think everyone can appreciate that. You hear the frustration or the excitement directly from the customer. Recording customer calls is a very simple way to drive enterprise-wide customer success. Everyone can really understand that voice from the customer. So again, pretty cheap thing to do and I'd recommend it.
Now on the CS side a couple of tips that I see people trip
themselves up on when they try to make a CS team,
one of them is if you're selling to a technical audience-- Which I know some folks in the
Get the profile of that CSM right.
One of the biggest mistakes people make in customer success if they're signed to
technical people like developers or security people is
hiring a bunch of CSMs with no technical experience.
What ends up happening is the CSM reaches out to the customer, and the
customer says "Awesome.
I'd love to talk to you."
They get on the phone and the customer realizes "This person can add no value to me
whatsoever." Technical people often don't love taking meetings just for the heck of
it, so that technical person from then on is like "Can I just talk
to the person that you're going to eventually route me to? Can
I just skip you?"
It's like when you call Comcast or something and you want to just
skip that first person to get to the second person.
How do you get that technical person to be front and
center with the customer if the buyer is a technical buyer?
Most companies that have a technical buyer have a relatively
technical CSM team.
That's an interesting insight.
Second thing on the CS side is, this is a pretty
So while there's a temptation to go hire the
person that's been doing customer success forever, I tell you nobody
really has because it's all pretty new.
So getting experience helps, but
having a beginner's mind and a fresh approach also helps.
Sometimes the best CS people come out of consulting
or technical roles or in your company in product, or whatever.
I'd be open minded to people without experience,
it's a new enough industry that you don't have to get
starry-eyed by the person that their resume says they've done
this for 20 years.
Because, by the way, all they did was change all their job titles on LinkedIn to
customer success for the last 20 years.
So there are experienced people out there, but don't
This is a new industry.
Number three, there's moments in your customer
lifecycle that are just so critical, and the most important one after you
sell and after you onboard is sponsor change.
That's when your key stakeholder leaves.
So, one of the things you can really do to hack the system is be
great at that process.
When a customer success leader buys Gainsight and then they
leave and there's a new person that comes in, there's a gift basket going
right away to the new CS leader.
There's an email from an executive at Gainsight to meet them.
There's a LinkedIn, and they're invited to an event, they get their blog
posts all right away.
By the way, the old customer that went to a new company,
they're being followed by our sales team. Organizing that whole
process doesn't take any money, maybe a couple of gift
baskets here and there. But
it's actually really valuable, it's really understanding your sponsors and when
Another thing is, in the same theme of what I've been talking about, people hire a lot
They will have like 20 CSMs or 30, and they don't hire ops
people. In sales one of things most people know is
you don't hire a lot of salespeople without some ops people to help them figure out what
to do and be efficient and effective.
So if you have four or five CSMs, you should totally have an ops person.
It may feel a little early, but I would take the ops person over another
sales person early because later on that will make you much more scalable.
That's the theme of the talk, "How do you make this thing much more
scalable, so you're not hiring an army of people later on?"
Last tip on the CS side is consider charging for customer success if your customer will pay for it. If you sell to large enterprises, one of the things that's surprising is they're used to paying for stuff like services. Maybe basic customer success is free, and you can have premium customer success, and crazily enough if the customer is paying a million dollars for your software they might pay two hundred thousand dollars for that premium customer success experience. That lets you hire more people and deliver better experience. That doesn't work if you're selling to SMB, but for enterprise that's an interesting way to monetize and scale what you're doing.
So I'll share a little personal story to close out with, because I opened up with that
picture of me and my dad.
I'll share with you another family picture here. This
personal story goes back about a year and a half, and it's a
little bit of a vulnerability.
Some of you have probably seen the Brene Brown famous TED talk on vulnerability.
I'm a huge fan of being vulnerable, even with people I've never met before.
So a year and a half ago I took my mom and dad out to dinner at PF
Chang's, it was a very fancy night out in the Palo
That's my dad and my mom.
For some context,
outside of business I'm a hardcore science nerd.
I love reading science books.
At night I read quantum physics books, "Is time a
reality?"The black hole thing last week was like the greatest thing ever for me.
So, that nerdy person that you may either be or be friends with.
My mom, she grew up in India and she actually did
a masters in physics and wanted to be a scientist.
But India back then, the opportunities for women weren't that great so she never got that
chance. So she always wanted me to go into science, and actually there was some other
life path where I would've gone into science.
So my mom and I-- I'm driving her and my dad to dinner and I'm in the car, and my mom-- Who, by the way, my mom and dad are essentially the Indian version of the Costanzas, if you watched Seinfeld back then. They're super direct, there's no filter, especially my mom. My mom says to me, "You love science. It's so great that you love science." She says to me, "It's too bad you didn't go into science because you really could have done something important with your life." It's one of those moments if your parents aren't radically candid like that, you're like, "OK." I just froze. I was like-- She says it with love, obviously, but she's like "You could've done something important with your life." I'm like, "OK. I don't know how to process that." My dad was a business person, as I said, I mentioned you he has a little more polished. So he's like, my dad's like "No, Mina--" My mom's name is Mina. "Mina, don't worry. He's still 40 years old, he can still do something important with his life." Then my mom without missing a beat says "No, he's too old." But she looks to the future and says, "Maybe his kids can do something." We have three kids and they are going to do much better things than I will.
When you get into business
there's always this question, "Why are you doing what you're doing?"
There's an existential question, all of us have
it, some of us talk about it but all of us have it.
Like, "Why am I doing what we're doing?
Why am I on these crutches here in San Francisco? Why
are we doing it?"
Usually I get it when I'm on some red-eye flight to some random city to go
sell software. When
you're selling software you're not curing cancer, you're not flying to Mars,
yet you have to have some existential purpose.
At least at Gainsight, as we have that question all the time of "What are we
doing with our lives?"
For us, when we thought about this customer success concept, we got one message that really resonated for us and might resonate as you think about customer success. This concept is basically that customer success is about thinking that customer is a human being. That's fundamentally it. At Gainsight we have this concept of "Human-first business." Our purpose at our company is to be living proof that you can win in business while being human first. So we want to always-- No matter what we do, we think of everyone we work with whether it's customers, employees, families, as human beings first. Simple concept, but most companies don't do it and I think customer success fundamentally is about that. It's about really humanizing that customer, having people listen to that phone call with them, really thinking about the outcomes they're looking for. They're not a transaction, they're a human being, and I think that concept is existential to e verything we do. Honestly, it's motivating and it gets you through those red-eye flights and awkward conversations with your parents, and things like that. Hopefully one day we can make all of our parents proud of us. So, if I leave you with nothing else, customer success is a human way to do business and I hope you're all as excited about it as I am.
That's a very astute question, because I think there's two
different parts of it and I want to make sure everyone follows it.
So often in sales you're selling to one person, and
then the people that use your product or service may be
different day to day.
That's part of the spirit of what your question is, right?
There's a lot of pieces to the puzzle, like "What are the things that can
One of them is the champion leaves, and I'll talk about that.
The other one is, by the way, the champion stays
and they lose sight of what's happening and then the users and the
admins or whatever you implemented, you did exactly what those people
told you but they are not talking to their boss well.
Then the champion looks at it and they're like, "What is this?
This isn't what I ordered."
It's like you go a restaurant and you're like, "This isn't what I actually ordered off
So there's two processes that matter a lot.
The first one, before your champion leaves, is what we call stakeholder
alignment. One of the things I didn't mention is
we codified what we think of as the 14 critical
processes in customer success.
It's on our website, it's called Gainsight Elements, and you don't need the Gainsight
product to do this.
It's like best practices on 14 processes.
One of them is called stakeholder alignment.
Stakeholder alignment, so that stakeholder that bought the
product, "How am I staying top of mind with them as they go through the
lifecycle the product?"
This is pretty straightforward stuff.
It's basically a regular cadence of somebody in your company
connecting with that senior person to make sure that we're
aligned on those goals, and we're still delivering toward them.
The best practice process here is in the sales
process, you've understood why they bought, you've captured those goals
and those outcomes they're looking for. In
onboarding you're verifying what that stakeholder, those are still the
outcomes, those are still the problems you're trying to solve.
On a regular basis you're meeting with them, some people call that a quarterly business
review, you're meeting with them and you're validating that.
Then at the renewal it's a non-event because you've got
over and over again feedback from them that you're on track, that the outcomes are
So that process of stakeholder alignment, most people do terribly.
They do really, really badly.
They sell to the stakeholder, they implement with the users and the
admin, they don't talk to the stakeholder.
By the way, the stakeholder is not easy to get to, so it's not an easy process.
We've defined a whole methodology on how you might do that.
The right person from your company to reach out to the stakeholder on a
regular basis, monitoring which stakeholders never responded to
you, because that's a risk.
If they're not engaging with you that's a problem.
Having a really good message when you meet with them, because one of the problems is if
you go try to meet with that stakeholder, and let's say it's 90 days after
You send them an invite and say "Can I spend an hour and a half with you walking through
everything happening with the project?"
They're going to be like, "I'm sorry, I'm really busy right now."
But if you're like, "Can I spend 15 minutes with you?" You
made the message really tight, and the message is "You bought for these three
reasons, and for each of the reasons here's how we're going. Red,
Something very tight like that.
There's a lot of training on how to do that well, and that's called
Then the other one is sponsor change.
So now that stakeholder leaves, and I mentioned before "How do you make sure when the new person comes in the best practice there is 'You're new to this company. We've been working with you for a while. Here's why your predecessor bought. Here's the goals he or she had. We've been working with you for a year, here's what we've delivered. Here's our feedback on what your team could be doing differently, what we should be doing differently. Are these goals the same goals you've got? Because you're a new leader, you might have a different set of goals.'" You may not agree with a person's goal-- By the way, most new leaders tell everyone that the old leader's goals were wrong. That's to some extent part of being a new leader. So you want to get to that new leader and very quickly understand their goals so you can reset and change the shape of that relationship. So, that's an excellent question. One of the most important things you do is stakeholder alignment and sponsor change management.
Yeah, that's a great question. If I did a very simple framework, it's like a sales operations person with a bit more experience on the communication side. Because ideally the CS operations person is doing a few things. They're doing all the systems work, processes and workflow, they're doing the data work like getting all the data together about analytics and health. Then they're also maybe doing some of the communications and enablement work, the emails you're sending out to customers, like the templates. The PowerPoint template, maybe the training for the new CSMs. Sales ops is a pretty good proxy, a lot of people in CS ops honestly came from sales ops. The other place a lot of people in CS ops come from is consulting, because to some extent it's just like consulting. You're going into a client and you're asking what to do, it's a consultative job. So the other place people had a lot of success is getting them from consulting.
That's a great question.
I want to make sure everyone else understood that too, some big
companies have a center of excellence and you're working with them. Then
presumably in your situation there's users, maybe in different business
units or departments, and that center of excellence can actually sometimes
become a blocker.
If you sell to big companies there's this tension, because
that center of excellence could be IT or it could be another group, they want you to
work with them and the users don't really
often like that center of excellence.
If you've worked at a big company most people don't like IT.
There's nothing against IT, just maybe the reality of big company IT.
So the users want to work with you directly. There's
a lot of tips on how to do this, you have to always walk a line, because if
you annoy that center of excellence or your key
person, they can really mess with you in the future.
But on the flip side, if you do everything they say but the users don't engage and like you, you're still going to get a customer churning down the road. So what I've found that's interesting is really when you go to the users, make it not feel at all like selling. Really make it about "We just want to tell you about the new release and some best practices," or "We're going to do a lunch and learn," and maybe you bring the center of excellence person with them. One of the things people will do is they'll arm the center of excellence person with content so they can be the hero, they can actually go do that training for the users. A third thing people are doing more and more now is in product for this reason. Because if the training is in the product, in other words you log in and you're a user and you see "Here's new features," or a little highlight of what's happening. It's hard to blame anyone for that. "No, that was just in the product. That wasn't my fault. That's just what the product people put in there." That's a really good hack to get around quote unquote "Center of Excellence."That's a very good question.
I'll give you a few others.
They're just the ones we waited too long to do, because it is easy, because we're just
like anyone else.
One thing I didn't talk about, which
is methodologies. What do I mean by that?
If your product has some configurability to it or different use
cases, to some extent it's a great strength
because it's very flexible, but it's a bit of a weakness because people could use it
in so many different ways.
One of the jobs of a customer success person is usually to
tell people how they should be using something.
That's a very common job, but one of the problems most companies
have in scaling is that knowledge of "Based on this
use case you should be doing X," is in somebody's head.
So one of the things that a lot of the more sophisticated companies do is
create a methodology.
What I mean by that is like a framework that says "OK, based on these business
problems these are the things you should do with the product, and this is how you should
I talked about our elements framework, that's our
methodology for doing customer success and using it, and that took
tremendous amounts of work because we did it when we were 500
people. By the way, when you're 500 people you've got 500 opinions or maybe more
about what the methodology should be.
But if you have two people on your CS team, you can have a nice little debate, and
pretty quickly figure out the best practice unlike all the
different use cases and things like that.
The way I would do it is on the left column is all the different reasons
people buy, and in the right column is all the different ways you
can use your product-- Try to match.
That's something I would definitely try to do sooner rather than later.
I mentioned sponsor change, obviously that goes without saying.
I'm trying to think of some new ones that haven't talked about.
Very tactically, I alluded to this
before, running strong customer meetings-- I
know it sounds so obvious, but this is really basic.
Show up on time, obviously.
Get your audio/visual stuff, video conferencing or
whatever, tested beforehand.
When you start the meeting the customer is so
impatient usually, especially if they are senior people, so get started right away.
If there's five people on your side and five people on the customer side, don't let
everyone do intros.
Because that takes up so much time in the call,
one of the people on your side can intro everyone and one of the people on
their side can intro everyone.
Don't have seven slides about the company update,
one slide about the company. Go
right into what they care about.
So, optimizing the engagement with the customers is something
that's a really high value thing you could probably do much sooner.
Probably another one that a lot of people do too late, we did too late. Training, like training your team on this stuff, and also investing if your product is a little more configurable, scalable training for your customers. One of the things that you want to do if you have an enterprise product . It's scalable training, online training, things like that. That's something that like CSMs do as a one off, but obviously you could do it a much more standardized way. So, those are a few.
Totally. I wrote a blog post on this on our site, but high level,
this is the most common question in customer success.
So, let me break it down into different pieces.
Number one, "To whom does customer success report?"
There are two options, typically.
It's either part of sales or it's part of
service and support and
operations, and all that.
The majority of people put it as part of service, support and operations,
sometimes people that have a more transactional product put it under sales.
I would say that maybe it's 70-80% put it under service, support and
operations. If it's in sales it can work, the person
that's running all of sales and your CRO or sales leader or
whatever, needs to care about customer success.
It goes without saying that if CS reports into somebody that doesn't care about
that area, that's where it'll break.
That's the main reason it will break.
Now, who owns the dollars? Here's what people typically do. If your product is really complicated and has a lot of learning that you have to do, and education, and evangelism, and configuration, it's hard for the CS person to also do the commercial stuff. If your product is very simple you can actually have the CS person do some commercial stuff, so there's a complexity spectrum. The companies that are very complex large enterprise, typically the CS person owns adoption value but doesn't have any commercial responsibility. The products that are maybe medium complexity, the CS person might do the renewal but not the expansion. And the products that are very transactional, sometimes a CS person can do onboarding, adoption, or renewal expansion and things like that. But product complexity is the driver there. In general, more CS orgs than not don't. Revenue, generally speaking, historically has been in sales and not CS. But there's a trend towards saying, "Can we move some of that renewal responsibility into CS as you get bigger?" I think if you're a small company though I wouldn't worry about that stuff too much, I'd really just That's what's going to make you a successful company.
Absolutely. What do I mean by, "Customers expect success?" In the early days of customer success and SaaS, a customer was buying SaaS for the first time and their experience was just like buying old software, so they didn't expect anything. The bar was very low. But now most customers have bought Okta or Workday or Salesforce, or something else. In your field it could be different, but they bought something, and good chances are some of the things they have bought have had this very proactive approach. So because of that, when they buy they don't plan as much. They don't plan how we're going to adopt it, how we're going to get value. They're expecting you to do that, and so because of that they've assumed, generally speaking, it's your job, it's not their job. Therefore it's just part of the expectation.
Totally. We have a blog post on this on our site, so
if you send me a LinkedIn or something I'll send you more details, but
high level is when a customer success person is working with a
customer and there's a set of conditions that would cause a risk to be
open. I alluded to it before, but just to refresh--
Something that's a blocker to a sale, like a
feature gap that prevents a sale or prevents
onboarding, prevents major adoption or prevents the renewal.
So you say, "OK.
That feature--" Now, that's a high bar. It's not
like they don't like the color of the button, that's the trivial example.
It's something significant, so then they fill out a form, and in
our case they do it actually in our Gainsight software, because Gainsight can do that.
But you could do it in a Google form, and they basically put in
What's the ARR?
When's the renewal date?"
Then, "What's the request?" With a lot of detail, hopefully on the use case, and "Why is that a blocker for that customer? What's the reason it's a blocker?" That goes into a workflow then, and you have a regular cadence every week, our product management team has to review that. It goes to the right product manager and then they've got an SLA to initially respond. Initially they may not be able respond with an ETA, like an estimated time, but they can at least respond and say "I'm going to go look at this." So their response could immediately be, "We're doing this already," or "We're going to do this. It's quick." But most likely the response is, "We're going to go scope this." Then they've got a responsibility every week to update that, so "OK. I just scoped it, we think it's going to be about two months," or "I think it's going to be in this release," or whatever. Then let's say it's going to slip, because that happens everywhere, a feature slips. They're going to update it again, and then once it's done they're going to close it, and then obviously we can tell the customer that it's done. It's not complicated at all, but it's just amazing that people don't do it. It creates so much frustration for all sides, the customer, the engineer and the salesperson and the CSM when people don't do it.
The most common mistake is probably literally not knowing the right person that should be onboarded. Going back to one of the questions earlier, you sell to person X and person Y is the person that is going to do the onboarding, and you're sending a bunch of e-mails to person X but should be Y. That's a trivial but important one, and a second mistake is if there's a sales process-- I'm assuming you're not self-service and you have a sales process. There's usually information captured in that sales process about why that customer bought, and for most companies, it's weird. You spend all this time telling a salesperson why you're buying this software, all this time all these demos, you're explaining, hours-- Then the onboarding starts, and "Tell me, why did you buy this product or service? Let's just start over." The real big issue is not handing off the customer's desired outcome. Going back to that word, "Outcome," from before, to the onboarding team. I think the third one is in the onboarding process it's not-- You would love to have it be like you go this many days and it's done, but the reality is it's you want to get to a certain milestone of adoption or outcomes. So, most people are trying to move onboarding to say, "It's not about getting to 30 days. It's about getting to this many active users, or this milestone of adoption, or whatever." Having onboarding not end until there's some real meaningful usage, that's the third one I'd add. Good question.
Totally, yeah. A simple framework to think about metrics-- The
question was just for the video, to talk about metrics of customer success.
Simple framework for metrics is you've got lagging
indicators, like when I say lagging, you can imagine there's leading
indicators and the things that are going to tell you what's going to happen.
Lagging is the things that actually happen.
We all are in business from a financial perspective, to drive lagging
indicators that are things like "How much upsell did you do?"
Or "What were your renewals?"
Pretty much in SaaS, there's standard nomenclature, if you
don't know it it's easy to find.
But something called gross renewal rate or gross retention rate, which is
basically all your customers and how much of that you're retaining, but without
including offsetting dollars from upsell.
So the maximum you can get on that is 100%, and then net retention rate
which includes the ability to expand those customers, that can be more than a 100%.
Those are your lagging indicators, but most people then
manage to leading indicators, and those leading indicators of customer success
tied to those two concepts I talked about of outcomes and experiences.
Experience could be a measure of net promoter score,
something called C-SAT which is after somebody calls for support, what's
their satisfaction? Things like that.
Then for the outcome, usually people start with measuring adoption, so they
get an adoption score and put some analytics in their product to
measure percentage daily active users or something like that.
But then eventually they get a little more sophisticated, for example, they
might say "It's not just about how many people are using our product, but how well
they're using it.
How many of the features they're using."
Some of the bigger companies will create a customer sophistication
score. "How sophisticated is this customer in using our
features, are they using the really basic stuff they can get anywhere?
Or are they using the really advanced stuff that we only do, nobody else does
it?" Or you measure the outcome, like I mentioned, of
literally "How many dollars are we saving them?
How much revenue are we driving for them?"
If you can do that, sometimes you can't measure that.
Then you've got your lagging indicators, you've got your leading indicators, and then you do measure your activities too. "How many meetings did you have? How many quarterly business reviews?" Etc. But it's pointless to measure the activities unless you're measuring those other two, so lagging indicators and leading indicators, and activities. Thanks so much everyone.