Gurkan Erdogdu, CTO and Co-Founder of ManageCat, shared with us how to install the ManageCat addon on Jelastic Cloud. Cloud computing is currently everywhere and lots of companies are trying to move their Java EE applications to a cloud infrastructure to get agility with scalability, availability and security of their...
The post Integrate ManageCat with Jelastic to provide Cloud Management of Apache Tomcat and TomEE Servers appeared first on Jelastic Blog.
Missed the 2015 SaaStr Annual? Weâ€™ve got you covered! Weâ€™re publishing the full series of both the videos and transcripts from all of the 2015 sessions. Today weâ€™ve got David Sacks, Co-founder of PayPal and Yammer, now COO at Zenefits, and Stewart Butterfield, President/Co-founder of Slack, where we had a wide-ranging, 40 minute discussion on how Slack, Yammer and Zenefits scaled to achieve Unicorn status.
Some of the most interesting takeaways from this session are delving into how itâ€™s possible to achieve $100mm ARR without a sales team; how product virality contributes to enterprise level inbound sales; how charging only for active users can propel growth; the viability of Freemium as a growth strategy; and how much attention you should be paying to the competition (not as much as you might think!)
Check out the full transcript below!
As youâ€™re reading, take a second andÂ sign up for the 2016 SaaStr Annual. Â Weâ€™ve expanded to 3 full days of incredible speakers and unparalleled networking with 5,000 of your SaaS BFFs. You definitely want to be there.Â Â Â
Jason Lemkin: Â A lot of stuff, a lot of fun and mostly we want to talk about tactical stuff here. Itâ€™s about scale and revenue. David, I want to start with you to get some of your thoughts. When was Yammer started, a long time ago in internetÂ time, â€™08 or something like that?
David Sacks: Â Weâ€™ve launched about September of â€™08.
Jason: Â A lot has changed since then. A lot of things I want to chat about. At a high level, why are things scaling faster than ever from your perspective?
David: Â One of the big changes obviously is mobile. It was around in â€™08, but it was much more of an afterthought. Yammer was really launched entirely as a web page product.
If I had it to do all over again, I would have made mobile the primary way that the product worked.
The specific way mobile translates into faster growth is that you can slurp the address book. We could
have been way more viral had we done mobile from the beginning.
In the original Yammer sign up flow, we literally prompted you to type in the user names of the email addresses of coworkers that you can remember. Itâ€™s like an incredible amount of friction.
There was no way actually to get the work email address book. On the consumer side, that had already been figured out a long time ago, going back to like Plaxo and Facebook and all that. I donâ€™t know if people can remember that stuff, but no one had ever really figured out on the enterprise side, because everything was in Outlook. You had to upload a CSV file, which was never going to happen.
Mobile has completely solved that problem. Actually, Iâ€™m not even sure why there are not more explosively viral enterprise apps now that you can slurp the address book. Thatâ€™s a huge difference.
Jason: Â That would have been an accelerator. We can talk a little bit about that. We donâ€™t have to spend hours today talking about Zenefits, but it is very interesting. Zenefits is growing far faster than Yammer did and Yammer was an epic success.
Obviously, itâ€™s a great product and great market, but things are different. Things are scaling faster. Yammer certainly scaled faster than I scaled back in the day with EchoSign or Aaron when he was here, but the pace of Slack. Any particular insights on why that is? It got you back in the game this hyper growth.
David: Â Zenefits is scaling faster than about anything Iâ€™ve ever seen. Itâ€™s hard to know whether thatâ€™s some sort of broad based trend or whether itâ€™s very specific to the mechanics of that product.
In the case of Zenefits, itâ€™s a product that every small business needs and weâ€™re selling it for free. That combination is explosive.
Jason: Â Stewart, picking up on that, I didnâ€™t become a Slack groupie until about three months ago when I was building some product for SaaStr, and then I got it, when I actually had to build software.
Weâ€™re groupies. Iâ€™m a fanboy. I love that itâ€™s everywhere on all my Macs, on my mobile. I love search.
The growthâ€™s been epic this year. In some of the press, Iâ€™ve read youâ€™ve said you donâ€™t know all the reasons why. Itâ€™s taken off. There are some other things I want to talk about, but do you think if we had the same Slack three years ago, would it be growing as quickly? Is it a function of the times?
Stewart Butterfield: Â The times play into it, for sure. I said when I was asked, â€śI have no f***ing idea,â€ť because that got a headline.
Jason: Â It was a good one.
Stewart: Â If I had given the 45 minute explanation of why I really think that, it would not have gotten any headlines. There are a couple things. Mobile is one of them, for sure. Mobileâ€¦ itâ€™s not any particular feature of it. Itâ€™s the fact that everyone now has this fantastic internet connection and a amazing computer in their pocket all the time.
Our relationship to technology has changed a lot in the last little while in a way that I donâ€™t think weâ€™ve really noticed yet, like the frog in the pot of water. It will probably obvious in 10 years. In retrospect, it will be apparent what the specific things were.
I rememberâ€¦Iâ€™m pretty old, so 20 years ago I walked into my dadâ€™s office and heâ€™s a real estate developer, and for the first time ever in my life I noticed that everyone at the office had a computer on their desk. Iâ€™m old enough that I remember when people had desks that looked like this, or what you sometimes see in a movie, where they had a pad of paper and a penâ€¦
Jason: Â Something like this?
Stewart: Â Like a Rolodex and a telephone. There was suddenly this transition to where the receptionist has a computer, the accountant has a computer, the architects have a computer, and the planners have computers.
In a kind of analogous way, everyone uses the Internet now, which wasnâ€™t true certainly, when we started Flickr, in late 2003. We started development in early 2004. I donâ€™t think that even 50 percent of US homes had Internet at home at that point, and certainly, a tiny minority of them, probably under 20 percent, had high speed Internet at home.
We forget how when you started Yammer, you said it would have been better if youâ€™d paid more attention, but you would have had to have been almost impossibly prescient to know that at the time. It was early 2007 the iPhone was announced. It wasnâ€™t actually available until later.
The first version of the iPhone, people forget there was no app store. That didnâ€™t come until another eight months later.
It was also, as revolutionary as it was, compared to what we have today it wasnâ€™t nearly as big. Android followed after, and that really, late 2009, early 2010, that suddenly changed.
I know, because we were working on a game, we started our company in the beginning of 2009, and we completely missed the boat. Even in early 2009, we didnâ€™t realize the extent to which peopleâ€™s discretionary computing time would shift to mobile. Itâ€™s maybe two percent laptop and desktop now, and 98 percent mobile.
Jason: Â A lot of things I want to chat about with limited time, but I want to talk about business models, because weâ€™re here about scaling revenue. I went, the day before yesterday whatâ€™s the new Slack program? Slack for Business? Slack forâ€¦whatâ€™s the name?
Stewart: Â We introduced a new tier, so we started with freemium model with a single tier.
Jason: Â Chat about that. Whatâ€™s the new one called, though? Itâ€™s Slackâ€¦
Stewart: Â The new one is called â€śPlus,â€ť and thereâ€™sâ€¦
Jason: Â I go to Slack Plus yesterday to buy it. Iâ€™m like, â€śWhere can I talk to a sales rep?â€ť [laughs] Iâ€™m clicking around the â€śContact Me,â€ť â€śTalk to Me,â€ť so obviously weâ€™ve gotten from nominal to double to eight figures in revenue without a sales rep. We all sort of think we understand the Atlassian model sort of, and we all get the Yammer model wrong, which I want to ask David about. Can you get to $100 million without a sales team, do you think?
Stewart: Â We donâ€™t have anyone who has a title thatâ€™s â€śsales,â€ť and then some people would say, â€śOh, you
really have a sales team,â€ť and itâ€™s kind of ambiguous.â€ť What we have, we call them account managers,
and thereâ€™s no outbound.
Almost exclusively the people they talk to are people whoâ€™ve already made the decision that they want to purchase, but they work at a big enough company that purchasing is not something that you can just do. Thereâ€™s a vendor review approval process, and a security policy analyst, and their in house counsel wants to mark upÂ the TOS.
Itâ€™s almost like midwifing the sale, as opposed to being a salesperson, so we canâ€¦
Jason: Â No oneâ€™s on commission today.
Stewart: Â No, and Iâ€¦
Jason: Â Iâ€™ll say you have no sales team.
Stewart: Â I believe we can have no commissions forever.
Jason: Â Forever.
Stewart: Â We can have no outbound sales forever.
Jason: Â When Iâ€™m fromâ€¦ pick your company, Dow, and I want to buy 1,000 or 2,000 seats of Slack next year, Iâ€™m not going to talk to a sales rep, and Iâ€™m going to pay a list price?
Stewart: Â Youâ€™ll talk to one of these account managers. The difference is, we wonâ€™t be approaching you, and the reason youâ€™ll want to talk to us is because like most large companies today, even now, thereâ€™s already multiple teams using Slack at the company.
We originally wanted to copy Yammer in the sense that anyone can create it. Then thereâ€™s enough people using it that the administrators want to pay to have control over it if we didnâ€™t quite follow the pattern properly, and that kind ofâ€¦ It didnâ€™t work for us, anyway.
We now end up, a company like Adobe, has fourteen different paid instances of Slack, so we have a different problem, which is not a whole bunch of individuals using it. Itâ€™s a whole bunch of teams using it, and thatâ€™s what they want to consolidate and have administrative control over.
Jason: Â David, contrast that a little. My understanding of Yammer, and maybe itâ€™s a little bit apocryphal, is that everyone looked at Yammer as this amazing freemium product that went crazy in the day.
As Iâ€™ve met more folks that actually worked on your sales team, [laughs] I think my learning, and correct me if Iâ€™m wrong, is that certainly Yammer exploded as a product through viral usage, at least within enterprises. When you got bigger, when you approached eight figures in revenue or more, you sold to the CIO, literally targeted the CIO and went to close real deals with the CIO, for a variety of reasons.
Some of your original vision was all freemium, very much akin to Stewart.
David: Â It was sort of a sale to IT plus usually some business sponsor, because IT primarily buys products that make their jobs more secure, like preventing security breaches or something like that.
They have responsibility for enterprise collaboration often, but thatâ€™s not exactly where their incentives are aligned. You normally need somebody in the business to push them, saying, â€śHey, we really like this Yammer thing,â€ť but their budget would usually come out of the IT budget. It was actually a complicated sale in that respect.
Certainly my belief going in was a little bit like Stewartâ€™s is now, that we hoped never to have to do sales, and our hope was that employees would pull out their company credit card and buy it.
We could never quite make that work. It turned out that the virality was a tremendous lead gen engine, and we never had to spend a dime on marketing, but then it really took a salesperson to get the sale over the top. Maybe itâ€™s because of the thing that Stewart mentioned, which is we were selling all or nothing company networks, and because of that, you had to make this enterprise wide decision, and that kind of pushed you into this enterprise IT realm.
Looking back, if we could have done it, Slackâ€™s done a whole bunch of things better than we did, and one of them might be around this idea of letting groups buy it, because that can be a much more self service sale.
Stewart: Â If we had started in 2008, your question before, we would not have taken off like this, for a number of other reasons. It was just harder.
There were fewer people, especially for Slack in particular, with its feature set, fewer people messaged back then. You had a BlackBerry in 2002. Iâ€™m sure, probably half the people here did.
Then there were Finnish teenagers you could use T9 to text, but it was not the normal case that people used messaging products at all.
Even at the height of popularity of Yahoo Messenger, AIM, MSM, and stuff like that, it was still a relatively small percentage of a much, much, much smaller user base, 150 million people using the Internet worldwide.
Jason: Â We can talk about otherÂ things, but to dig in on that, because Davidâ€™s point is, and I love the science of sales and the art of sales, but, as founders, weâ€™d all love to have no sales team. No matter how much we love sales, because, of course, itâ€™s the dream to spend all your time making the product great, and sales, even in the best case, is a huge tax on your time, even with the greatest team.
What I find that people sometimes get confused on the Yammer and now Yammer/Slack model is understanding what verticals that can work and what canâ€™t.
Stewart, totally disabuse me if I get it wrong, but notwithstanding all the blue chip logos you have, youâ€™ve got a very core software development, project management core.
Actually, Yammer came out of the same place, because you and Adam had the same idea. I donâ€™t think that Yammer had that same software developer core that could get accelerated.
Stewart: Â Youâ€™re right. The reason for that, I donâ€™t think is because the product is especially difficult to use or requires a lot of technical expertise, but because itâ€™s intended for internal communication.
If you take the whole US workforce, weâ€™re not going to work for retail, food service, health careâ€¦70 percent of the US workforce itâ€™s not going to work for at all.
Of the 30 percent that remains, the knowledge workers, at one end of the continuum, you haveâ€¦My favorite example is real estate agents, who almost never talk to anyone who works in their same office. Maybe three percent of all their communications and the other 97 percent are external, while software developers donâ€™t talk to anyone except for other software developers.
Maybe thereâ€™s a UX researcher and thereâ€™s a marketing person who is bringing them news from the outside world, but day to day, they are only talking to each other, so Slacks works perfectly for them.
Weâ€™ve done well in tech, weâ€™ve done well in media, and weâ€™ve done well in creative industries, like ad agencies and design, and stuff like that. We do in this case in all kinds of other places.
First of all, weird companies, industrial paint and solvents, or shipping and logistics, and then some groups that donâ€™t belong to those categories: some sales, some customer support, those kinds of things, but youâ€™re right. Itâ€™s concentrated there.
Jason: Â The interesting thing about recurring revenueâ€¦ we chatted about it earlier with Aaron Levie and Bob Tinkerâ€¦ in my experience, if you squint, you can barely see 10x on the horizon, what the company looks like.
You donâ€™t exactly know how to do it, but you know what it feels like. What youâ€™re saying is the Slack customer base that we have today can get you to 10x is what you feel in the customer base.
Some people get overly attached to freemium as founders, and I want your thoughts on this statement. Sometimes, itâ€™s a disservice, because they think freemium can take them further than it can, and sometimes they underinvest.
Stewart: Â Iâ€™m totally open to the possibility that Iâ€™m being hopelessly naive.
Jason: Â In your gut, you can agree, David. You can feel 10x.
David: Â Freemium worked really well for us. It certainly has its limitations, but I generally like it as strategy. Itâ€™s probably not right in all contexts, but it definitely worked for us, seems to be working for Slack.
I guess a good question like, â€śWho it wouldnâ€™t work for?â€ť but I tend to be a fan.
Jason: Â Let me flip around, since youâ€™re a fan. Itâ€™s good to dig into the Yammer story, because people get it wrong, at least a little bit wrong.
Running Yammer, when do you realize youâ€™re going to need to hire a sales team, and how did you approach that and scale it? I met you probably before that, you come out of PayPal and all this, and itâ€™s a change. Itâ€™s like, â€śNow Iâ€™m building an enterprise software company.â€ť [laughs] Youâ€™re very quantitative, so how did you make that decision and execute it?
David: Â It was probably about six to nine months in, we realized that sales was going to be an important thing to us. The backstory to it was we were all consumer Internet folks who were tackling this enterprise problem.
We knew that we were doing business software and enterprise software, but we thought we could do it in this completely consumer way. That meant viral, freemium, using A/B testingâ€¦
We generally thought that there was, and I actually still think this is true, that thereâ€™s this over compartmentalization of knowledge where, if youâ€™re a consumer company, you use this set of techniques, and if youâ€™re an enterprise company, you use this set of techniques, and both sidesâ€¦
Consumer companies should be much more open to sales, and enterprise companies should be much more open to growth hacking.
In any event, in our case, the growth hacking worked from a top of the funnel perspective, but it was not closing deals. We realized six months in that we had to do something to close deals, and thatâ€™s when we started building a sales team.
Then, from there, we essentially reverse engineered all the classic functions of an enterprise software company. Once you have a sales team, you start realizing youâ€™re in competitive deal situations, so then we need a marketing team that can enable the sales team to create the right materials.
Then we needed a customer success team to make sure that customers were successful with the product and therefore renewed, so we ended up rebuilding all these functions out of Salesforce.
Jason: Â In the out of Salesforce driven culture and team. Thereâ€™s a question after that. You reverse engineered Salesforce, and now you have a chance to take these learnings and practice them again at Zenefits.
Now itâ€™s different, but it isâ€¦ Sam and Parker here, thereâ€™s a hundred sales reps, 60 SDRs. [laughs] Itâ€™s a sales drivenâ€¦ What are you going to do better this time? What are your learnings that you want from this process?
David: Â Every one of these companies is really different. Thereâ€™s actually more similarities to PayPal with Zenefits than there are to Yammer.
Itâ€™s a more operationally intensive company the way that PayPal was. The transactions are a little bit more like PayPal. Yammer was this collaboration tool and revolved around, obviously, people posting messages and content.
Thatâ€™s important, but if you lost one message, itâ€™s not the end of the world. With PayPal, when you deal with peopleâ€™s money, or Zenefits, when youâ€™re dealing with HR information, mistakes are much more painful. You have to get it right. and out of that flows a whole bunch of operational requirements. Zenefits, itâ€™s actually a lot more like the PayPal story.
Jason: Â Got it. So the learnings from all these experiencesâ€¦ Â On the revenue side, youâ€™re less focused on bringing the Yammer sales learning to Zenefitsâ€¦
David: Â As youâ€™ve probably heard from Parker and Sam earlier today, theyâ€™ve really figured out the sales model. That thing is cranking. The other areas are what need more attention.
The fact that, actually, PayPal is a better reference point for Zenefits than Yammer was. It kind of goes to my point about over compartmentalization that there is now this body of knowledge, itâ€™s from consumer enterprise that can beâ€¦ It needs to be selectively used in the right places. These domains are all collapsing.
Jason: Â I want to talk about collapsing. Stewart, let me ask you a related question. You hired as your CMO or VP of Marketing someone out of Salesforce. You went and you added your enterprise DNA.
Stewart: Â Ex Salesforce and, most recently, the CMO of Zendesk.
Jason: Â Tell us a little bit about that decision to bring that DNA in. How thatâ€™s driven the company, and what youâ€™ve learned from that.
Stewart: Â When I was speaking before about the difference between account management and sales, it totally depends on your perspective. We donâ€™t feel like sales is an icky thing. We also donâ€™t feel like marketing is an icky thing, like some people with software backgrounds might.
To us, marketing is critical. Itâ€™s everything we do Â the copy that we write in the UI, even things like the technical operations team, and the speed at which pages load, and all those things. I know that it sounds crazy to think of that as an aspect of marketing, but the biggest thing would be what we call customer experience.
Instead of making customers support a cost that we have to try to manage and thinking of that role as an entry level job and kind of low paid, we sought out former engineers, or technical writers, people with a higher level of expertise.
We combined customer support with functional testing inside of QA so that people could write docs, and they would know what features were coming. The support is phenomenal, even for the free users. Thatâ€™s, in my mind, a marketing cost.
The thing that we were missing was any kind of knowledge of how things actually get marketed in terms of paid advertising, in terms of lead gen on the one hand and branded advertising on the other, because we thought we wanted to do both. We needed someone to be responsible for it, but there was no one at the company that had any real expertise.
Bill was actually unique because heâ€™d led online marketing for Salesforce, and he was a CMO at Zendesk. He also had background doing consumer stuff in the games world. Thereâ€™s like a little bit of eachâ€¦
Jason: Â Got it. Itâ€™s the convergence that David talked about.
Stewart: Â The thing that really made the difference there was it was that expertise, that background on the enterprise side combined with the idea, which is rings exactly true to us, that every interaction that anyone has with any experience they have with Slack. Whether itâ€™s the customer experience, whether itâ€™s the initial user experience when you first run the mobile app, whether itâ€™s a bit of copy, any of those things is a marketing opportunity.
How people think of us is the sum of all of their experiences. Thatâ€™s critical.
Jason: Â In that model for Slackâ€™s goals for 2015, do you do one on ones or anything like that with your management team? One on something?
Stewart: Â Theoretically, I do one on ones.
Jason: Â Theoretically? At least itâ€™s not an alien concept.
Stewart: Â with the Team.
Jason: Â When Parker and Sam were here, the idea of one on one seemed alien. When you meet with Bill and you talk about marketing, in a not a sales driven, but you invest into marketing, what are the KPIs? What are the goals? What is the quantitative side of things that youâ€™re looking at, if there is one?
Stewart: Â We were lucky enough to have one metric that we really pay attention to. Thatâ€™s daily active users, because itâ€™s the kind of product that if youâ€™re not using it every day, youâ€™re not really using it. We can think aboutâ€¦
Jason: Â Thatâ€™s higher than ARR or MRR.
Stewart: Â I guess theyâ€™re almost one and the same.
Jason: Â Marketing can drive daily active users as their core metric too. Marketing doesnâ€™t have a covert goal, a lead goal, a revenue goal. Itâ€™s a daily active user goal.
Stewart: Â We grew 28 percent in January. Thatâ€™s on top of a great year. Actually, we have started a very small number of tests of paid advertising. Thatâ€™s not the result of a marketing effort yet, again, other than making sure that people have a good experience. A hundred percent of our growth, historically, has come from people recommending us, specifically on Twitter.
Thatâ€™s been an under recognized aspect of this, because people wonâ€™t post on Facebook, because it is not appropriate to say, â€śHey, we switched tools at work, and I really enjoy this new software.â€ť They have no compunction doing that on Twitter. They might have 400 followers, or they might have 8,000 followers. They might have 140,000 followers.
That really adds up. We have a custom timeline of, at this point, probably 56,000 tweets of people saying, â€śJust switched to Slack. Love it.â€ť Every one of those, it might only be seen by 20 people, but might be seen by tens of thousands.
Jason: Â Thatâ€™s interesting. Letâ€™s step back, because one of the things when weâ€™re building recurrently, we think a lot about lead sources. Where do they come from?
You know I love the Twitter, but literally, Twitter, as best you can measure it, is your number one source of paying customers and users. Are people on Twitter saying, â€śSlack rocksâ€ť? You track that from Twitter?
Stewart: Â Twitter and press and PR. If we didnâ€™t like sales or have had the kind of person that says, â€śI donâ€™t know. Sales is gross,â€ť then weâ€™d say, â€śNo. We donâ€™t have sales. We have account management.â€ť We could say that we donâ€™t do any marketing at all. We didnâ€™t do any paid marketing, but we did lot of PR. We invested a lot in that. Theyâ€™re all different ways of looking at the same thing.
Jason: Â Is that pretty evolved, from your experience, David? Is it consistent? â€śTwitter is the number one leads source,â€ť is a new one for me. Itâ€™s a cool learning for me from this session.
David: Â Twitter is the place that people get their news from. PR is a great form of marketing. Twitter and PR sort of go hand in hand as a strategy. I really like that as a lead source. There are more specific things you can do, depending on the market youâ€™re in and things like verticals that your verticals and channels can go after.
Jason: Â Weâ€™re back to your question on people. You talked a little bit about Bill. Obviously, 2014 was a great year. It sounds like January was a solid month. [laughs]
Beyond that, any key critical hire that helped you inflect the revenue side, that made a difference?
I know you have the worldâ€™s greatest team, as we all do. Was there something, especially in the revenue side, that made a difference and was a new experience for you?
Stewart: Â There are a lot of little techniques, but if I had to choose one thing that might be interesting to a broader audienceâ€¦
Jason: Â Please.
Stewart: Â Maybe two days, 48 hours before we officially launched, which was February 12 of last year, so weâ€™re coming up to our first anniversary, we had thought through the billing flow, and how people sign up, and how they pay. We charge per user per month, like a lot of services.
We were thinking about our biggest single customer at that time, which was WalmartLabs. When they had set it up, they had set it up so that anyone with a Walmart.com email address could sign up, that was the bit that we ripped off from Yammer.
A little bit incorrectly applied in our case, because itâ€™s a product for teams as opposed to the whole company, so youâ€™d have people who worked in other parts of Walmart showing up here, like some guy who works inâ€¦ who stocks the film in Bangalore or something like that, showing up at an user experience project for new concept stores or something like that.
We were asking to make the purchase decision, â€śDo you want to buy this product?â€ť They had 800 people or something like that on their team.
About 400 of them were actually using it. The other 400 were people who had either signed up because they saw it somewhere, or because they tried their Walmart email address, or they hadnâ€™t spread to their team yet or whatever. We had to make him decide, â€śDo you wanna spend an extra 400 times 80 bucks for the year for all of these people, or do you wanna go and manually deactivate 400 accounts?â€ť
Jason: Â [laughs]
Stewart: Â Thatâ€™s a terrible choice. People are going to think about that a little bit too long and then maybe not buy at all. We instituted what we called the fair billing policy, which is we can easily detect accounts that are inactive, once an account is inactive. Every night weâ€™ll go through every single account and scan them.
If theyâ€™re currently active, and theyâ€™re no longer using it, weâ€™ll mark them as inactive. If they were inactive, but they started using it again, weâ€™ll mark them as active again. You only pay for the ones who are active.
Jason: Â Letâ€™s chat about that for a minute, because thatâ€™s interesting. You believe that was a significant accelerator to making less friction on the revenue side of the business. This is what youâ€™re saying, right?
Stewart: Â Iâ€™m a huge believer in less friction. Of all the places you can remove the friction, the purchase decision would be a big one.
Jason: Â I want to hear all that in comparison to Yammer, because if you talk to most folks selling SaaS to the enterprise, youâ€™ll hear a story that, â€śI sold a thousand seats, 10 deployed the first quarter, 500 by the end of the year. And then Iâ€™m hoping that the renewal comes [laughs] in.â€ť No one ever uses all the seats they buy.
As founders, we love our product. We love this idea of deprovisioning seats, because itâ€™s what weâ€™d like. For many SaaS companies, it might destroy their economics. I love Salesforce to death, but having worked with a lot of Salesforce customers, a lot of the times, they donâ€™t even fully deploy until the second year.
If you delimited, or whatever the term was, those seats, the whole first year ACV. Thatâ€™s generated enough. Itâ€™s taken enough friction out of the sales process that they led to a revenue boost.
Stewart: Â Some people didnâ€™t even notice. Our average paid team size is 20 people. A lot of them donâ€™t know everyone who is on the team. It will be eight people or something like that, but then, one person goes on vacation for a couple of weeks, or one person gets fired, or one person quits.
Jason: Â The fired one.
Stewart: Â They forget that they didnâ€™t deactivate the account.
Jason: Â They always do.
Stewart: Â Then, 10 days later, they get this email that says, â€śHey, we noticed that so and so isnâ€™t using this anymore. Hereâ€™s a prorated refund for all the money that you paid for this person.â€ť
Jason: Â Thatâ€™s cool.
Stewart: Â That, obviously, didnâ€™t get us the sale in the first place, but it gets us a, â€śWow, thatâ€™s amazing.â€ť Theyâ€™ll tweet about it, tell their friends about it. Theyâ€™ll be very happy with us. Theyâ€™ll be much more likely to renew. They have a positive impression. That positive impression, obviously, makes a huge difference.
Jason: Â David, youâ€™re hyper quantitative and thoughtful on this. How did you think about that at Yammer, because Iâ€™m sure at Yammer you wanted to get all the seats you could get too, right?
David: Â Right.
Jason: Â [laughs]
David: Â I love that idea that Stewart implemented, because it removes any perverse incentive that the admin would have to remove people from the network because they donâ€™t want to pay for these extra seats. If you have a network product where the value of the product increases with the number of people that are contributing content, then you want the network to be as big as possible.
The way we did that at Yammer is that it would either be a contract negotiation, and they would pay for a certain number of seats. Then we would typically waive. They knew that the network would keep growing. We didnâ€™t want there to be any deterrent to letting the network continue to go viral. We would frequently waive the overages for the rest of the year.
Then, when the renewal came around, then the ASP would go up. We kind of saw that as like a win win. Now virtually all products have networked elements to them where you have users within a company interacting with other users. This is actually a great idea. Most companies would be afraid to only charge for their active users becauseâ€¦
Stewart: Â They are afraid.
Jason: Â [laughs] Theyâ€™re afraid.
David: Â â€¦Theyâ€™re afraid that they have too many inactive users, and they love the idea of selling shelfware. If you want to keep yourself honest, actually, I really like the idea of only charging for active users.
Jason: Â Iâ€™m not sure most sales teams would love it. They want a consistent ACV to get their quota.
Stewart: Â Much of that has brought applicability, because weâ€™re a messaging product. People use it all day, every day.
Jason: Â It does, but a great user experience has broad applicability. Trying to quantify the value of that versus the money youâ€™re losing by that is sort of founder angst sometimes, right?
David, I want to ask you a follow up question to Stewart about inflection points and teams. You join Zenefits now. What are the resources people that you brought in? Because youâ€™ve had a chance to think about this, Iâ€™m sure you really want to twist the knobs and the dials. Itâ€™s a great rocket ship, but your real goal is to even help it accelerate further.
Whom did you bring in? Who have you brought in? What type of teams, people? If itâ€™s too early, tell me itâ€™s too early, but Iâ€™m guessing youâ€™ve thought through this pretty carefully.
David: Â Obviously, before joining the company, I had a series of conversations with Parker and one of the things that came up quickly is that the company didnâ€™t have a product management team. Thatâ€™s because Parker and one other person were essentially doing it all himself.
Jason: Â CEO is a product guy. Weâ€™ve all lived it.
David: Â Which is great, because Parker did a great job, but at some point with a 600 employee company, at some point you might decide you want a product management team.
The first week that I was there, we started building a product management team. That led to product marketing and other functions along the way.
Jason: Â Whenever someone comes in the kind of role that you have, oftentimes, you come in thinking youâ€™re going to do a certain thing. Then, no matter how much diligence you do, you get in and you realize thereâ€™s a different area where you can add the most force multiplier.
Has that changed at all? Do you always think that professionalizing product was job number one?
David: Â Iâ€™ve only been there six weeks, so my views have definitelyâ€¦
Jason: Â Itâ€™s a fire hose, though, isnâ€™t it?
David: Â Yeah, for sure. What I would say is I didnâ€™t have to know that much about the company to know that you want to have a product management team. Itâ€™s one of those core functions that a company of any size will eventually want to have.
That was kind of like a no brainer. There are some more subtle things that I picked up along the way over the last six weeks, but itâ€™s probably a little more involved.
Jason: Â One thing, curious to get both your thoughts at different levels. One thing about founders, sometimes we worry too much about, because we canâ€™t help it, is competition. Itâ€™s so visceral and thereâ€™s the tweets and the press releases.
Do you think at the growth rate you have, do you think about competition at all?
Stewart: Â No.
Jason: Â Is it irrelevant?
Stewart: Â To us itâ€™s irrelevant.
Jason: Â What about at more a Zen level. Letâ€™s say I went from Hipchat to Slack. It is possible if I went from HipChat to Slack, I could go to Zack someday.
Stewart: Â Absolutely.
Jason: Â Thereâ€™s competition today, which at this growth rate is probably irrelevant. Then thereâ€™s sort of the Zen competition. You spend much time with that, with this growth rate, can you even think about it?
Stewart: Â 70 and maybe as high as 80 percent of our customers if we ask them, â€śWhat did you use before Slack?â€ť will say, â€śNothing.â€ť
Theyâ€™ll shrug their shoulders. If we say, â€śOK, well, letâ€™s say you had a change in your benefits plan and you need to send an announcement to everyone in the company, what would you do?â€ť They say, â€śOh, weâ€™ll send an email.â€ť
â€śOK. What if you are going to be ten minutes late for the meeting and you want to tell someone so they can apologize?â€ť They say, â€śOh, well, then Iâ€™d send a Hangout message.â€ť
Itâ€™s usually not that they use nothing. They used this heterogeneous mix of different unconnected messaging systems.
Jason: Â Eighty percent of Slackâ€¦
Stewart: Â Eighty percent of Slackâ€™s customers.
Jason: Â The cool kids, the hipsters, didnâ€™t use anything before they used Slack?
Stewart: Â Well, they say they used nothing.
Jason: Â Oh, I see.
Stewart: Â Itâ€™s because they donâ€™t imagine internal communication tools as a category.
If youâ€™re going to start a sales team today, the first thing you would do is choose a CRM. You wouldnâ€™t get started. If you were going to have engineers writing code, you wouldnâ€™t get started until you had chosen a source control system and so on.
People think of those things as distinct categories. If you start a team for some reason, you donâ€™t think, â€śOK, what are we going to use for internal communication?â€ť
You start going. You set up Google Apps if youâ€™re a new startup. You assume that youâ€™re going to get peoplesâ€™ individual phone numbers and send them SMS or whatever. Itâ€™s not a decision that you consciously make. Given that thatâ€™s the case, the challenge for us is to establish internal communication tools as a product category.
There are now in the Bay Area today, 50,000 people were using Slack or something like that. Many of those people are going to go on to join new companies or start new companies or join new teams, found new teams and they will make the decision.
â€śWhat are we going to use, because we used to use Slack.â€ť Itâ€™ll be true too if they use HipChat as well, but thatâ€™s the real challenge for us is to establish that as a category.
Jason: Â Now, I want to get back to David and then weâ€™ll probably only have time for one or two more questions.
Squinting out there and seeing 100 million, 200 million in revenue, what does Slack want to own? Itâ€™s a communication tool. Itâ€™s sort of an ad hoc search/project management tool. What kind of company are you at 200 million?
Pick whatever aspirational number you want, but itâ€™s not the same Slack today. What space do youâ€¦and then competition changes, right?
Stewart: Â A platform for internal communication as â€śblahâ€ť as that sounds as a category, the name is probably right.
Jason: Â Whoâ€™s an incumbent you displace in your mind?
Stewart: Â I donâ€™t think there is one yet, or itâ€™s Microsoft, but Microsoft in a super broad sense that includes Lync and Skype and Yammer and SharePoint and Exchange and Active Directory and a bunch of other stuff.
The strategic position that we occupy when weâ€™re making $200 million a year is the bottom layer of the business technology stack, because when you use Slack, you plug in Twitter if youâ€™re interacting with your customers on Twitter. You plug in Zendesk if you use it for customer support tickets. You plug in GitHub if youâ€™re using it for source control, on and on for like marketing, analytics, BI, CRM, whatever, continuous integration testing, application performance monitoring.
The 30 different services you use all plug into Slack in a way that if you plug in 30 things into Slack, you couldnâ€™t have plugged all 30 of those things into one of the other ones.
Jason: Â Oh, for sure. It is great.
Stewart: Â In other words, you couldnâ€™t plug everything into Microsoft.
Jason: Â Maybe in 2016, theyâ€™ll get there. A good engineer can build in integration these days.
Stewart: Â You built N squared integrations. For every service you add, youâ€™re going to have to integrate it with the other 30 or 31 or 32, whereas everything will integrate with Slack. Having one search box that searches across all of your systems is incredibly valuable.
Jason: Â Itâ€™s invaluable to me. Iâ€™m curious how you think about it, because the thing thatâ€™s innovative today becomes table stakes a couple years down the road. The Slack, the integrations, my CircleCI, GitHub, that getsâ€¦ itâ€™s beautiful, but given enough time and any team of engineers can build something.
Stewart: Â Iâ€™d say we donâ€™t pay attention to competition and this might sound a little bit arrogant in the sense that if you were a musician and you were recording a new album, you donâ€™t go and look at all the other albums that are out today and see how you can make yours better relative to that. You have some songs that you wrote. We have some software that we wrote.
Jason: Â [laughs]
Stewart: Â At the same time, we are paranoid that someone will come up with a way to deliver 80 percent of the value of Slack with 20 percent of the complexity. That someone will find like some simplifying essence, and that will be easier and that will take the core idea and the value that weâ€™re getting, and be able to deliver it.
Jason: Â Thatâ€™s what you obsess about, the order of magnitude easier? The disruption you canâ€™t see or feel.
Stewart: Â Yeah.
Jason: Â David, quickly. I know weâ€™ll run out of time. I remember when Yammer launched. You formally launched at TechCrunch Disrupt or something and then this product that sounded kind of similar called â€śChatterâ€ť launched, right? I donâ€™t know if you remember that. It sounded similar.
What are your Zen learnings on competition? Zenefits we talked a little bit about earlier. Itâ€™s interesting. Huge amount of indirect competition, maybe bounded direct competition today, but what are your learnings through all these experiences?
David: Â I would love to have no competition, but unfortunately, with PayPal and Yammer, we ended up having a lot. I agree, the question is, â€śWell, how much should you focus on them?â€ť There is a danger in becoming obsessed with them. I would see people whoâ€™d come obsessed with us and start copying all
these features that didnâ€™t work.
Jason: Â Thatâ€™s a danger.
David: Â Weâ€™ve got plenty of them. Theyâ€™re doing all the wrong stuff. Itâ€™s like funny.
On the other hand, you canâ€™t completely ignore the situation either. Obviously, with PayPal, we were competing against eBay and they own the platform where 70 percent of our business came from.
Jason: Â Brutal.
David: Â There was no way to ignore that. With Yammer, Chatter was sort of the threatening one. It wasnâ€™t threatening because it was a better product or anything. It was because it was bundled with Salesforce. It took them a couple of years, but they finally realized they should give it away free.
That was the point at which weâ€™re like, we think weâ€™re better, but on the other hand, now we have to compete with â€śfree.â€ť
Jason: Â Free is scary. Itâ€™s scary.
David: Â Thatâ€™s a dangerous place to be. Frankly, thatâ€™s probably why I havenâ€™t created a company that hasnâ€™t sold itself.
Jason: Â A little bit of fear, competitive fear. PayPal, it was fear, but maybe even at Yammer, though.
David: Â Honestly, both of those dynamics really played into those decisions. I donâ€™t know if it was the right thing or not to be so worried about it.
Jason: Â Those are good. Thatâ€™s good. I feel like we got five minutes, but weâ€™re out of time. We got to get to the Crunchies. We got to get to an internal team company meeting with 2,000 employees, so weâ€™re out of time. Letâ€™s thank David and Stewart. This was epic. Thank you very much.
Jason: Â If you can hear me, everyone go downstairs. Weâ€™ll have drinks downstairs. Theyâ€™re going to break this room back up into a party. Thank you, everyone. I really appreciate all the time.
Transcription by CastingWords
(Cross-posted @ SaaStr)
Amazon CloudWatch monitors your cloud resources and applications, including Amazon Elastic Compute Cloud (EC2) instances. You can track cloud, system, and application metrics, see them in graphical form, and arrange to be notified (via a CloudWatch alarm) if they cross a threshold value that you specify. You can also stop, terminate, or recover an EC2 instance when an alarm is triggered (see my blog post, Amazon CloudWatch â€“ Alarm Actions for more information on alarm actions).
New Action â€“ Reboot Instance
Today we are giving you a fourth action. You can now arrange to reboot an EC2 instance when a CloudWatch alarm is triggered. Because you can track and alarm on cloud, system, and application metrics, this new action gives you a lot of flexibility.
You could reboot an instance if an instance status check fails repeatedly. Perhaps the instance has run out of memory due to a runaway application or service that is leaking memory. Rebooting the instance is a quick and easy way to remedy this situation; you can easily set this up using the new alarm action. In contrast to the existing recovery action which is specific to a handful of EBS-backed instance types and is applicable only when the instance state is considered impaired, this action is available on all instance types and is effective regardless of the instance state.
If you are using the CloudWatch API or the AWS Command Line Interface (CLI) to track application metrics, you can reboot an instance if the application repeatedly fails to respond as expected. Perhaps a process has gotten stuck or an application server has lost its way. In many cases, hitting the (virtual) reset switch is a clean and simple way to get things back on track.
Creating an Alarm
Letâ€™s walk through the process of creating an alarm that will reboot one of my instances if the CPU Utilization remains above 90% for an extended period of time. I simply locate the instance in the AWS Management Console, focus my attention on the Alarm Status column, and click on the icon:
Then I click on Take the action, choose Reboot this instance, and set the parameters (90% or more CPU Utilization for 15 minutes in this example):
If necessary, the console will ask me to confirm the creation of an IAM role as part of this step (this is a new feature):
The role will have permission to call the â€śDescribeâ€ť functions in the CloudWatch and EC2 APIs. It also has permission to reboot, stop, and terminate instances
I click on Create Alarm and I am all set!
This feature is available now and you can start using it today in all public AWS regions.
Enterprises are suffering a scarcity of insight. Think about it. There is an unprecedented amount of data in the world, and it is growing exponentially by the day. In fact, according to IBM, 2.5 billion gigabytes of data are created each day. This data is being generated by everything from smartphones to enterprise databases; scientific, market, and medical research, as well as vehicle telemetry data, and all of that information pouring in from the burgeoning Internet of Things. Everything that can have an IP address gets an IP address, and it is generating data.
There is so much data being generated that it is placing extraordinary pressures on enterprises to properly capture, store, analyze, and then present in a comprehensive, insightful way. In the years ahead, it will be the most successful enterprises that are able to turn all of that data into insight and identify the important changes and trends to best respond to customer and market conditions, such as tightening commodity availability or surges in customer demand. Much of the insight needed is there to be found, but it must be unlocked.
To be certain, to get the most out of this data, enterprises must not only be able to collect it, but they must also be able to harvest it and create meaningful insight. This is a challenge facing not just consumer and IoT businesses â€“ but every business.Getting at the data that matters
It starts with data collection. If data isnâ€™t collected, it may as well not have ever existed in the first place. Whether that data is coming from far away sensors, a smart-phone or -watch, or applications and databases that reside on-premises, cloud or end-user end points â€“ it needs to be collected and stored. Increasingly data, whether user generated (health monitors etc.) or user created (notes taken from observations) is coming from mobile devices: phones, tablets, smart watches, etc. Where does the data go when it is collected? Too often it goes to some siloed database, or an isolated cloud server, or sits on the device never to be utilized again. What a waste.
That brings up an important point, when it comes to accelerating innovation with data: the data must be accessible. The more centralized the data can be stored, the easier it can be accessed. This design lends itself to cloud architectures. Fortunately, data analytics tools are finally being developed as cloud native. This makes the data accessible to everyone.
Often, when data is centrally stored, itâ€™s still not accessible in a readily or informative way to the bulk of enterprise users. Legacy data warehouses, for instance, donâ€™t provide businesses the ability to capture all of the information they need from where itâ€™s created and then provide that information to the users that need it to more effectively and successfully do their jobs. And when data is available to query, itâ€™s often incomprehensive gibberish delivered in multiple reports that even data scientists would struggle to assemble into something coherent.Real-world innovation at the speed of information
Big Data is a hot space right now. Too often though we forget the data is only as good as how you use it: What insights can you gain from it? How easily it is to analysis, and how quickly can you get these insights? At Salesforce, we’ve see customers, who were previously struggling with data sprawls and poor account visibilityÂ transform dozens of disparate reportsÂ into a single dashboard, with that one dashboard providing more functionality than those previous reports combinedÂ provided. The result is improved agility and theÂ ability to share and collaborate all of that information with other business teams, executives, and others. This is how data analytics can speed innovation in organizations.
Another example is a beverage manufacturing businesses that had a manually generated scorecard of various key performance indicators and other metrics. They would process scads of data within Excel, format it, convert it into a PDF and share it via email with their management teams, who would then print out these emails and share it at meetings. Not efficient.
This manual process would take about three weeks. The executives at this organization were persistently making decisions based on data that was roughly three weeks old. By implementing data analytics they were able to create a real-time dashboard that is able to access the data where it sits. That means all of the scorecards, KPIs, and other metrics generated can be seen in near real-time. Thatâ€™s powerful. It’s what we often call a Customer Success Company – one where the customer, and the interactions, come first.
In todayâ€™s competitive market, insights like that canâ€™t go hidden. When competitors have access to insight to changing market conditions â€“ they can rapidly improve and out compete and out-innovate those that canâ€™t.