April 27, 2017
Ep. #12, FaaS and the Benefits of Serverless
In the latest episode of JAMstack Radio, Brian invites Ryan Scott Brown and David Wells to examine the rise of Functions as a Service (FaaS)...
Before we get started, how many folks know Trello? Full house, good. How many of you guys pay for Trello? Oh, all right I've got my job cut out for me still. I've got to sell you guys. All right, cool. So that's good. So you guys know Trello. I won't spend too much time talking about that then. I'll talk a little bit more about my favorite subject, which is myself. So I'll tell you guys a little bit about me and some of the roles I've had before this.
My first management role for a sales team was at a management securities services company. I ran an SDR team, and then I eventually ended up running a new business team there as well. And it was a very traditional sales cycle. So we were selling managed security services to banks, and there were no leads for us outside of things we'd generated for ourselves. We had a big SDR team. Went through a very traditional cycle.
You brought on your regional vice president to do all the hand shaking and do the sale, but then that person couldn't actually talk about the product at all, so you had to bring in a sales engineer to actually do all the hard work. And then those guys couldn't implement it, so you had to bring in TAM, and then you had to bring in an account manager. So, you know, as traditional as you can get from a sales cycle perspective.
I left that organization, and I went to Fog Creek, so if you guys are familiar with Trello, you might be familiar with Fog Creek as well. Same co-founders, Joel Spolsky, Michael Pryor, and we were focused on selling developer tools into those groups. And that started to be kind of the first place where I was in a less traditional sales cycle. Fog Creek had Fog Bugs, which had a really great word of mouth. People who followed Joel knew it. Engineers bought it on their own.
They didn't really want to talk to sales people, so it was an easy process for us, and it really was something where people felt like they could just try it, buy it on their own, and only really needed to talk to sales if they had specific questions. So I was one of the first people they brought on to run a sales team for them. It was really part of experiment, and
It was the first place where I saw this model of a lot of automation and not a lot of sales-heavy focus.
When I left Fog Creek, I went to another company that was doing data security again, and that was another company that actually looks kind of similar to Trello in the sense that it had a business to business sales process. But it also had a consumer sales process, and that was where that whole business just ran completely on its own. Those people bought, renewed, everything without really interacting with us, and then we had this enterprise side of the business, which was the only place that we were talking to people.
So all that kind of came together to be a little bit of what we ended up building at Trello, which is very similar to that, which is that Trello gets adopted on its own. Obviously, most of you guys if not all of you put up your hands that you know it or use it, and for us, we're pretty hands off on that. We let the product drive that. We let marketing drive that, and our sales team really only comes in where it makes sense to have a human touch at. And so I'll talk to you guys a little bit about how we got to that, and how we've kind of combined all of those models together.
So today, we're going to talk about kind of the main three areas of your sales process, right, which is the evaluation stage. How do people end up buying your software? The purchasing process itself, and then the renewal process. And I'll tell you some of the stuff that I probably didn't do soon enough, and if I could go back in time, I would do differently.
So from an evaluation standpoint, there's a few things when you first go to market with your software. And one of the really traditional ways to think about going to market is to gate your demos and your trials. It's not let somebody download something if you have SaaS software, not create an account, whatever it might be without having to have a conversation with a sales person first.
I think you need to kind of ask yourself how easy is it for someone to get up and running and actually start testing and playing around with your stuff? Why are you gating your demo?
Is there actually a good reason to gate it? Is it something that you really feel like if people just try it on their own they would completely fail at or is it something where you're just trying to manage that conversation because you think they should talk to you before you do that. And for me I think unless it's an extremely technical solution there's really not a great reason to gate that right at that point. People should be able to download and try it right away.
I think you should let the software drive itself. Obviously Trello is a little bit unique. I don't know if anybody has freemium solutions in the room, but the nice thing with freemium in this example is, I was saying this to a couple people downstage earlier. Which is that we come into the sales process and somebody's already selected us. We don't really get into the conversation like "well, we're comparing three or four pieces of software, which one's the right one for us?"
We come in and we just have a conversation about the value of Trello at your organization. And if you should be paying for it or not. And so letting the product drive that model. Well, it's scary at first to kind of let people test and do whatever they want with your software without you getting involved right from day one. It can really drive a lot of that. You have to have adoption between the product and the sales team to make that work.
So this is a more traditional trial model. Somewhere on your website, someone fills out a request to do a trial. And they get a nice little response that says someone will be contacting you shortly. And then from that you get that lead, maybe it's going into Salesforce or whatever you're using for that management. And then hopefully, you or your account executive or someone on your team is making lots of phone calls and emails to try and get in touch with that person to let them try out your software and learn all the interesting things about it.
If they don't get in touch with those people, usually you close that lead up and you say, "okay, it was a no good lead." Maybe, ideally you get in touch with those people, then you're scheduling up a call to walk through the trial before you let them test it. Then they're going through an evaluation process. And normally in this stage you're setting up three or four meetings, you're doing a lot of emails back and forth to make sure the conversation is going.
And then somewhere in that process, someone's deciding whether they want to buy or not. And that's going to involve a lot of email, a lot of phone calls out to those people to make sure you're staying on top of it.
If you think about a slightly more hybrid model, which is, this is more like what I've done at the companies before Trello, which is we let you download and start that trial. You clicked you wanted to do it, you went for it, either you downloaded it or created an account for you right away. And then from that state we basically did some lead scoring against you. And so we knew what our criteria was from an ideal customer profile. So we were looking at things like what kind of industry are you. For us we wanted to talk to somebody who was in an industry that had regulations against it, so banking, health care, whatever it is.
And then we also wanted to look at the size of the organization so we were using tools like Clearbit to supplement this data and to do some lead scoring. And then from there basically we would decide if you've scored at a certain criteria or not. Okay, you're going to go into a drip campaign. And no one's going to talk to you unless you reach out and you really need to speak to somebody.
But otherwise you're going to have this email drip campaign that you go into, you get six to eight emails based off of whatever interval it is and hopefully that's going to guide you by providing you pre-recorded resources to watch, materials to read, whatever it might be. And then ideally from there it's going to guide you through the purchase process. Otherwise, if you scored, you're going to go into a human campaign.
Which means that you've round-robined into a sales rep's queue where you're territory based. You've gone into that person's queue and they're now going to reach out to you and try and get in touch with you. Ideally here you also have those people in a drip campaign, so at Trello we use Yesware. There are a bunch of good tools for this that we use Yesware. And what's nice about that is that the rep doesn't need to make a reminder to send out an email.
They can cue all that up, they can have it customized to their accounts and if somebody responds to those emails it stops that campaign and it lets you take it from there. So ideally you've kind of cued up that campaign so someone's not manually doing that. And then you have someone calling against all of that.
And then here, if someone's deciding not to purchase or they go silent, you wait some period of time and then again ideally you can use Hubspot or something like this where you're going to score those guys back up and you're going to put them probably in a re-engage campaign depending on what your software is. For us we would wait six months. And then we'd say, hey do you remember us? Come back, take a look at this again. And that would all be an automated model that would then put you back in to this, depending on how you responded to that.
The Trello model is slightly different. So this is a freemium model for us. So the unique thing with Trello and freemium is that for us, Trello never breaks.
So if you want to go out and you want to use Trello, and you want to make 8000 boards and add however many people you want to add to those boards, you can keep doing that. There's never a little popup that says "hey, time to upgrade, pay now."
So for us, that kind of trial stage can last a very long time. It could last forever in theory. So how we supplement this is that we're doing that lead scoring concept again. Which is that we're saying once you grow to a certain size in Trello, and you kind of look like these things. So again, are you in an industry that's secure? So if you're a .edu and you have 7000 users, it's probably a lot of students. Like you're not a great candidate for this.
If you're at a bank or something like that and you have a couple hundred, you're probably a fairly good candidate. So again we're using Salesforce, we're using Clearbit, we're supplementing that information to say, "okay this is a good client profile." And then from there if someone reaches out to us, we're getting you into that model of saying, "okay, how is that, how was that sourced," now you're going to have a human interact with you or now you're going to have an email campaign.
The caveat here is that scoring can only get you so far. So if somebody writes in and they don't match that criteria necessarily, but they're asking those triggers about tell me more about your data security, " tell me more about your hosting, where are your servers at." We know that that's probably something where we should be having an enterprise or a more intense conversation with you.
And so then that human kind of gets it out of the queue even though it wasn't scored perfectly, and then routes it into the model where someone's reaching out and talking to them. So we've kind of made a little bit of an in between model there for us. So, assuming everything goes well, hopefully during your evaluation stage, ideally the customer's going to decide that they want to buy from you and you want to start thinking about how easy is it for someone to buy your software.
And I've seen this go wrong with a lot of good companies, that then when it goes to the point where someone has to actually buy it's very difficult and they make you talk to somebody and they make you interact with somebody for absolutely no reason.
There are some industries and some roles where people just don't want to talk to you. And you should embrace that. Ask yourself "how easy is it for someone to buy your product?"
You know, for us, one of the reasons that Trello was able to be so successful and have such a lean sales team is that people could just go in, pop a credit card in, have their options, how they wanted to buy and really didn't have to interact with us at all if they didn't want to. So you know, depending on the software you're talking about, like can people just buy it? Can they buy it in their own currency? Can they buy it monthly? Can they buy it annually? What are those options? And how user friendly have you made it for them?
So you might be sitting in the room and you are saying, you know we're super expensive! Like we're charging $300 grand annually for this, and no one's just going to pop a credit card on and buy that, and you're right. They're probably not. I've very rarely seen that. And so then the question becomes, once they're in that buy process, how easy is it to add on modules to that without interacting with you ideally? How can they renew? Can this be an automated process? Do you have to dedicate an account manager resource to it or not?
So for us, when we picked up the phone, you know Trello adds 100,000 plus people every week. We are not calling all those users manually. And certainly a ton of those people write into us all the time and we're not getting on the phone to answer those questions. So for us, we picked up the pone if, one it was an enterprise sale fit, which means that you kind of lead sourced your way so you looked like an enterprise fit.
You were the kind of company that was going to be buying enterprise or you were a sensitive enough industry where we thought we could have that conversation and it would be valuable for us. The caveat here when I first started at Trello was that Trello had been a product for a while, and then the company spun off. And people would write in and say "hey, I want to buy like couple hundred licenses, I need someone to fill out the security form for me. I need somebody to talk to me."
And we didn't have a sales team in so the support team was like "no, sorry, we can just work in the queue" and they just couldn't afford to take the time to get out of that queue. And they were really working like a machine there. The problem is that when somebody wants to talk to you and they can't that's extremely frustrating for them.
Just think about how many times you've been on a queue for something and you've just smashed zero until somebody picks up and talks to you. Nothing is a worse customer experience than that.
So you want to balance that "we don't get on the phone, with we don't want to tick you off because we're not getting on the phone" and you have to figure out where that is. So for us, we would always answer the questions, but we really did say you had to be interested in enterprise for us to give you a demo, for us to do markups on your contract, for us to do a security review. Otherwise we very politely pointed more towards mass resources that those people could access.
So, in terms of when we picked up the phone and when we didn't, we really treated our sale like an expansion sale. We don't have a big team that's doing nothing but landing. We let the product and the automation and the marketing land, meaning that those folks ideally go in and accept terms and conditions and purchase on their own. And for us it's very hard, I think it's hard for anyone to kind of fake that relationship at the beginning. And to kind of pressure people in to buying your product. It's an unpleasant customer experience usually at that point.
But what is the opposite of that is that once somebody is a customer and you give them really good customer service and you understand their business and you help them grow with that, those relationships work really well.
So while we didn't focus a lot on building the team that lands those accounts, we did focus on having humans that were around to have that conversation from a relationship stand point.
And so we're totally comfortable with taking a lot of little yeses.
We're totally comfortable with landing a lot of small deals because we know that if we can get in and have a relationship, understand that business, that we'll have a good bit of expansion off of those people. So when you look at our sales team, we try to automate as much as possible at the initial sale. And we really focused all kind of, the human relationship aspect of it into the expansion and the renewal of that account.
So in terms of renewals, we're annual. I've worked in businesses where we did three year contracts, where we've done monthly contracts, I've seen it all. The traditional renewal process usually is pretty manual, depends on where your business is at. But yeah I've seen very large companies that still have almost 100% manual renewal process meaning that somewhere in Salesforce, there's a reminder that your account's about to renew.
And you send them an email maybe three months out, you call them, and you set yourself another reminder and you call them again in six weeks out and you keep kind of manually calling and emailing this person until either they renew or they don't.
So for us, from a more automated standpoint, we have a lot of customers that just wouldn't scale for us. So we do in-product notifications. Taco pops up, if anybody knows who Taco is, it's the little dog inside of Trello.
He pops up he says, "Ru, you're about to expire." And um, because by the way huskies don't bark, they say "Ru", which is a fun fact for all of you to know now today. So Taco pops up, he says Ru and then you can go in and you can renew right on your own if you want. Other than that, you're in a drip campaign. We're sending out an email to you. You have different trigger points. Depending on where you're at we pop you out of that.
If you've responded or if some other trigger event occurs. And basically we don't talk to you at all through your renewal process. We're just letting the product and the automation and the emails take care of that communication. Now, the difference here is that at a certain point, maybe it's a couple of days before the renewal, maybe it's after the renewal is up, depending on how you scored from a lead score perspective, where we're saying this customer is a certain criteria or not, then someone's going to reach out and have that conversation with you.
Otherwise you just go back into a drip campaign. Now it's a little different for us. We are a freemium model so for us if you stop paying for it, you don't necessarily also stop using it. So we kind of have a little bit of flexibility of we can retain a customer. But I think that this method works best. Because most people kind of know what they're doing with your product, most people want to keep it up. And as long as you have good buy signs from them for the rest of the year, meaning their activity is not dropping off, things like that, you don't need to spend a dedicated resource where someone is making phone calls and manual emails about a renewal.
You should be able to look at product adoption throughout the course of the year and understand really if it's a sticky or not. Now if you're seeing your MOW drop in that account, or whatever it might be, you know, that's where you might want to break this. That's where you might want to have a manual process come in and ideally you probably have seen that sooner than three months out anyway. So you've already got a customer success manager or somebody jumping in there. And having those conversations before that point.
So for us from an automation standpoint, the lowest hanging fruit to automate is email, right? And I think especially now, salespeople don't want to get on the phone as much. They rely really heavily on email. And so I'd be talking to my team and I'd be like, "well how are we doing? What are you guys doing? How many phone calls have you gotten through?" And they'd be like, "I got through like 200 email."
I was like, "okay, how many phone calls have you gotten through?" And it was always a conversation about how many emails they were getting through. And my statement is like "listen, if you can be replaced by like Yesware, then you're not really providing the most valuable experience for the company." Like if you can be replaced by a machine that can just send out emails, we're misusing you in this process.
And so email was something that we immediately said, listen you want to have good, thoughtful email templates, you want to have things that connect with people. But in the end a lot of email is just reminding people that you exist. And like do you need to be the person who manually reminds somebody that you exist? Probably not. Let's use you on phone calls. Let's have you have higher level conversations.
And so for us email got automated pretty much right away. And then in terms of also like lead scoring, making sure we knew who we were going after, categorizing those accounts, triaging things in our support system, round-robining. That all became manual. And it really was something where we would round-robin, and you still have that human element, right, someone still looks at that, like quality control and says "oh this shouldn't have gone to me. Like this is a better lead."
And we still have that manual process there. But you can eliminate 90% at least of what you're working on by just kind of scoring it right, automating it out, using tools that support that. So like I said, Hubspot, Yesware, Salesforce, all of those are things that we used in this process.
So, what I would do differently, we didn't really understand Sales Ops very well at the time, I think we kind of felt like it was something that was a little bit of a luxury for us to invest in early on. It felt like something I could easily just kind of score up our stuff and it was going to be fine. In hindsight, I wish I had brought in Sales Ops much much sooner if I could DeLorean it back in time, that would be the thing I would do.
I think you need to have somebody who thinks about nothing but the analytics of your sales.
So as a sales manager I was worried about staffing my team, I was worried about growing those people, I was worried about making sure I was sitting on calls, I was coaching that. And then this very small portion of my brain was spent thinking about the analytics of our business. And in reality someone should have been a partner with me in that process who was saying nothing but " are you optimizing this process the best way you can? Have we AB tested that? Have you set a group of customers aside and tried to automate the whole thing? And then have you set a different group that looks exactly like that right aside and have done exactly what you were doing? Did those things change?"
And for me I was just so focused on making sure the team produced like they should be producing, I just didn't have the luxury of thinking about that kind of stuff. And so I would definitely say if you're early on in that stage and you haven't brought in Sales Ops yet, I think a lot of early stage companies kind of lean on a VP of sales to figure that out. And it's a really unique VP of sales that can be a good coach, a good mentor, a good trainer, a good recruiter, a good business sense person, and be a good analyst.
They're out there, they exist, but it's a hard group of things that you're asking for. So I would definitely say just bring in an analyst and let them go. I would have done a lot more experiments. I think that's the other thing that kind of comes with the luxury of looking back. But I think teams like to have clarity, and we wanted to stay clear on that. We wanted to give you really defined roles. But I think one of the nice things about being a startup is you can play around with things a little bit more.
And I think there was definitely places where we should have kind of had two or three people who were really experimenters for us. And we'd say, "here's the team that's kind of chugging along on the business, let them keep going. And here are three people who are smart and get things done, or entrepreneurial, and I have some projects I want you to run. And kind of compare that against what the rest of the business is doing and really see, were those guys making a difference or not."
And it's okay to fail on that stuff. And that was one of the things too that we did a lot of in the beginning and that we learned the most from was those experiments that totally failed. And I wish we had done more of them. The big point that a lot of people take away with automation is that you think you don't need people. You swing to that side where I was talking about Trello was when I first got there where we wouldn't talk to anybody.
And automating as much of this stuff as possible doesn't mean you don't need people. You still need that quality control person who's looking and seeing if that stuff looks right. You still need that person to talk to the customers who want to talk to you. It's just a matter of applying those people wisely. And for us, every time we peeled off that grunt stuff the team got happier.
Every time someone didn't have to set a Salesforce reminder to email somebody in six weeks, that team was like I am much happier in my job. I'm doing higher quality things. I'm using my brain at a higher level. And so they'll resist it at first, because they think that the value they bring is that active automation on your email. But when you can really free them up and say "no, you're a higher caliber person, we need you doing higher caliber things for us, let a tool do that other stuff for us," there's a quality there, but they, people just naturally resist that. They're afraid to be replaced by those things.
The other thing to think about is to fix it once. If you find yourself doing a super rote thing, we used to do this horrible process where to understand if you were a good fit when you wrote in, we had to look up your company and all this stuff manually. And it took like, I think we timed it out, it took like a minute for every lead, which doesn't sound like a lot, but it totally adds up. And the team absolutely hated it.
And someone said to me, "can't we not be doing this?" And I was like "yeah, I'm sure there's a way to not be doing this, but its going to take an engineer." And it was one of those things where I had to go and I had to say, "I know we don't have someone for internal tools, but this process is absolutely killing us, like this is a totally broken process." We made the business case and we got an engineer to come in and help us and that all became an automated process where it looks up your company.
And then eventually we bought Clearbit and we integrated that at an even higher level. So sometimes you have to spend a little money on those tools, but it makes your team so much more productive. So I ask my team all the time, our standard one on one conversation is "what's broken for you? What are you doing that feels like it's just stupid?" And there's been good feedback, like someone was like "it's really hard to find phone numbers I have to manually do that." I was like, "there's a tool, let's buy Zoom, it lets us look up phone numbers easier."
So if you're not doing that with your staff now, just ask what feels stupid and see if you can fix it for them. And then finally this kind of reinforces that stuff, but really invest in those resources to help you get there. We used a terrible video conferencing tool, which I won't name for a really long time. And it was super painful because every time we did a call with a customer it was two minutes of like they couldn't hear us, we couldn't hear them. But it was cheaper and we shared the login, which we probably shouldn't say.
But then we were like, "this is really, this is a painful process for us and we bought everybody zoom.us logins." And it's been amazing because they can have the tool they need and it works perfectly for them. And it's just one of those things where it's a little efficiency that we gain every day. So that's it, that's me. You can email me. You can Twitter me. I think that's what they call it these days. But yeah, reach out to me, I'm always happy to chat. I'm in New York so I'm not out here very often, but always happy to talk to you guys in email or online if that's helpful. So that's it for me. Thanks.
I think you invest in a good email system. We used Mailchimp for a long time, or whatever it is. But I think you need something. Especially if you're going to be a small team like that. And you need something that's qualifying those ops for you.
If 15 people write in, all 15 are not equal. And Sales need to have an easy way to see what that is.
And so putting that person in some kind of drip, letting them have those responses coming in, letting those people kind of self select, at least initially, I think that's an easy win right at the very beginning. It's just to have that system in place. And you'll use it for other stuff, or you'll use it to drip out to your existing customers, you'll use if for renewals. It's just a good solid first investment.
I've seen it and I think it makes sense. It's a question too of are some of them probably never going to convert for you? Or could you probably guide them enough without there being an interaction? So if you're having an SMB or somebody tested out, you could gate that, you don't need to gate that maybe necessarily, but you should have a really prompt, good communication that goes out to them guiding them through that.
And then I think if a Fortune 500s coming in and you know that that's going to be an experience that you want to make sure that they are having more tailored to, I still don't know if you need to gate it as much as I would have it be something where someone's immediately reaching out.
People don't want to be lost in something, so they'll let you know that they're having difficulty for the most part trialing your stuff. But I think there's nothing, even with a very technical solution, there's nothing more frustrating especially then if you think you're about to get access to something and then something pops up and it's like we'll contact you in 24 to 48 hours to let you have your login.
It's just that it's a rough first customer experience. I think you can still let people have access and give them some of those materials and then have an account executive or somebody contact them within even a couple of hours of that lead coming in. So I think there's a hybrid there. I think gating is still an experience for most stuff that people feel like has to be there, but if they actually would open it up a little bit they'd realize that they'd been wasting a lot of time qualifying kind of junk leads and spinning circles on things.
The average person is usually slightly more sophisticated at the level when they're trying it that they can get through it enough that they'd rather at least have a go at it before they have to talk to somebody.
So that's a great question. That's actually one of the experiments we did run. So when I came on, we didn't have an enterprise product and we kind of figured out what it was going to look like. And then we had that conversation about if we should put the pricing on. And I was like absolutely not, you never put enterprise pricing on a website. The people'll get sticker shock, they're not going to talk to you, don't do it!
And everyone's like, "oh yeah Kirsten you're the expert we'll listen to you, and I was pretty sure I was absolutely right on that."
And then we thought well, let's see what happens if we put the enterprise pricing on. And absolutely nothing changed. We didn't lose any leads.
Nothing changed in terms of a drop. And I thought for sure all the sudden you would see things plummet. So then we took it off again, just to see. And again, nothing changed.
And then when Atlassian had acquired us, I don't know if anyone's familiar with their model, but they're very transparent on all these things. And one of the great things about how they've done it is that they don't negotiate pricing. And what's nice is that we were spending a lot of cycles, because people just thought that that was the natural engagement they needed to have with us.
And so we were marking a thing up to be able to mark it down, it was like a real back and forth. And it was making our sales cycle longer. So when Atlassian did the acquisition for us, we put the pricing on the website. And at that point it really enabled us to say " listen, we don't negotiate. That's the price, I don't care who you are, how you come in, you don't have to be an expert negotiator to get a better price out of us. If you catch me on a good day, you're not going to get a better price than if you caught me on a bad day. You're not going to have to do that whole song and dance where it's like, well, if I can sign by the end of the month, well it may get my regional VP to come in and have that conversation with you."
So for us, because we tested it and we saw that it didn't affect the leads that were coming in, we went ahead and put that pricing on. And then when we did stop the conversation about negotiations and we pointed that out to the website, we saw our sales cycles start to decrease because it was just, it just was what it was. Like it's the price. Go back, figure out if it's in your budget or not. And so it took that 30 days of us going back and forth with people on pricing and it eliminated that. So not only did it not hurt us, but actually tightened up our sales cycles for us.
So I think the key was really doing the industry research. And this is where Ops and those guys come in and are very handy to say, okay this is what people are buying out in the industry, this is what those prices look like. You can roughly assume they're discounting x amount off of that. People aren't usually giving 70% discounts. They're like kind of discounting in the 30%-ish range or something along those lines. And then you just price yourself fairly.
And it's one of those things where it might feel like a better deal from the other guys, but in reality it kind of all shakes out with the wash. And I think you do have some people who love the negotiation, right? And you will disappoint those handful of procurement people who really love to do that. But most end users, most business units that are buying it, they don't want to do that.
And like for us we were dealing with dev teams, we were dealing with business teams. Like these guys were folks who wanted it just to be an easy process, they didn't want to have to do that big rigamarole with us. So for us it was a matter of pricing fairly and then having it be a more pleasant experience. Which is you don't have to negotiate with us for this.
Initially we kind of when out to the market and we were like talking to our customers and we would say "hey, we have this enterprise thing, here's what we're going to offer. And we'll get you guys on it for the same price as the business class products, but in exchange, you have to give you the feedback of what you wanted." And pretty much consistently across the board in those early days, every single one of those customers was like, "do you not integrate with SSO?" And we're like, "oh yeah, if you're managing a couple hundred people, you probably want integration with SSO".
And so those initial features came out from those guys just saying, honestly saying like this is painful for me to do and can you help us fix that? And so a lot of that low hanging fruit, feature-wise that I think were the right decisions really just came from having those customers give us that feedback. Our job is to funnel that feedback back to the product team and those guys are making that decision.
So one of the things that was really clear from the very beginning and I think this comes to some extent from having a developer focused co-founder and the origin story of the company is that at some companies Sales dictates that stuff, but not at Trello. We were feedback vessels to those guys, but products still had a vision of what Trello was going to be and they made decisions on that.
So there were features we said we were hearing a lot from customers and then product would say that doesn't fit in with the vision of what we're doing. And that was okay. And then sometimes there were things where we said, " listen, you're leaving a lot of money on the table. Like, I know it's not beautiful to put out SSO, but it's just expected and needed." So it was a balancing act, and I'm sure we're not done with it. It's a continuous process.
But I don't know if there's anything glaring that I would have put in that we didn't get in. I think the things that we've fought for, we made a good business case, and made it make sense and we got those things and there was probably some stuff that we asked for that was probably knee-jerk and I think product probably gated us smartly on those things.
Yeah, I probably could have done an hour, right? And thrown marketing into that, Marketing is a huge component for us. And so, there's no sales notion where we call up people who don't have Trello accounts and try and convince them that Trello is a good fit for them. All of that is marketing. All of that is word of mouth marketing or efforts that our team is doing around community and things along those lines.
Where marketing has been involved with us before as Hubspot is a marketing tool that we use. So they're looking at things like what are you visiting? Have you come and read our white papers? Have you attended webinars? And that adjusts your score accordingly. So I think marketing's a huge component of when I talk about where that lead score is at, where those guys fall. I think that's all kind of generally owned by the marketing team.
And so we work really closely with those guys for that. I think it's probably just where we fell, like I think for Trello a lot of our growth was focused on just get them in the door and have them using Trello. And that's what marketing's job has historically been for us, and then it was our job kind of to square that stuff up. But we were on startup mode too. As marketing evolved and we evolved, it became more of a co-ownership of those things.
I think that's probably pretty much owned by marketing at that point. So we are doing that, we are saying " if you're a business class account and there are certain actions that are taking place those guys are getting into the queue." And then there is lead scoring that's going off and then coming in to Salesforce for us to follow up with on that. And then we put an additional kind of human eye against that because marketing scoring is not perfect either.
So sometimes you'll score really high and there's like a weirdness about why that happened or it's an org that even though they're having a lot of growth, they're a non-profit so we're going to be tapped out on what we can actually do. We can't go after these guys at a high level from an enterprise or it's just not a thing that's going to be of interest to them, because they don't care about that level of security.
But yeah, I think that play is mostly marketing at that point. And then that's them feeding those leads into us in Salesforce and us taking it from there. I think relying on marketing for that stuff, at least at the initial level where you're figuring out if someone's a good candidate or an expansion candidate, cuing it up, getting them to have that level of interest generated, there's no reason that shouldn't primarily be a marketing play.
And sales should really be coming in later on, if needed at all, honestly. If you're a customer that's sticking and you're growing really rapidly like that, you might just get a self-serve upgrade or whatever it is without even having to have a person go after that.
I said that one of our first hires that I missed was Sales Ops, because I had control over that. But I think
One of the hires we missed as a company earlier on was an internal tools developer.
I think especially when you have a lot of developers, we were primarily developers, right? There's this like, we can hack together anything, it's fine. And they will hack together some solution for you, and it is generally fine, but they're also focused on making sure new features are going out to the customers. And those guys are having a great experience and in the end, those two poles will almost always pull towards the fun stuff of working on their product and making sure their customer is happy.
And so one of the best things we did was we got just an amazing internal tools person who is like, "you are my customer, I want to make you happy." And the work that that guy banged out for us, it made all of our lives easier. So yeah, if I could have had control over that, that wasn't really my hire to go after. I would say that was probably the second thing I would go back and do differently.
I would have gotten him in differently. Because the concept of data existing in those different silos that's not unique, right? Everyone experiences that. And you need to have someone who does nothing but think about you as the sales team or the marketing team as the customer. It makes your life easier and pull those things together for you and I think internal tools is a really good role for that and it's probably how I would suggest someone resolve that if they could. Thank you.