Previous Lecture Complete and continue  

  Reading - Marc Andreessen On the Future of Enterprise

Marc Andreessen On The Future Of Enterprise:
By Alexia Tsosis

In doing research for a post on “The Enterprise Cool Kids” at the tail end of last year, I interviewed Silicon Valley veteran Marc Andreessen about where he thought the enterprise was headed.

While excerpts of that interview made it into the post, the transcript of the entire interview was so good it deserved to be published in its entirety.


Alexia Tsotsis: Since people like me (millennials) are putting pressure on our IT departments to buy products that we can actually use and aren’t blinded by, what do you think the enterprise space will look like in the next five years?

Marc Andreessen: Yeah. So let me maybe start with sort of – top-down and bottoms-up is how we think about it, because both are important — so let me start with historical context and then maybe go to the stuff happening right now. Is that all set?

Alexia Tsotsis: Yeah, it’s perfect.

Marc Andreessen: So the computer industry started in 1950 and basically ran for 50 years with the same model, which was a model where all of the new computers, all the new technology, all the new software started out being sold for the highest prices to the biggest organizations.

So originally the customer was the Department of Defense. It was the first customer for the computer. In fact, one of the big first computers was called SAGE, which was a missile defense, the first missile-defense computer, which was like one of the first computers in the history of the world which got sold to the Department of Defense for, I don’t know, tens and tens of millions of dollars at the time. Maybe hundreds of millions of dollars in current dollars.

And then five years later computers became — they dropped half in price and then the big insurance companies could buy them, and that’s when Thomas Watson, who ran IBM at the time, was quoted as saying, “There’s only a market need in the world for five computers.”

The reason that wasn’t crazy when he said it is because there were only five organizations that were big enough to buy a computer. So that’s how it started. And then IBM came along and productized the mainframe, and then all of a sudden big normal companies — manufacturing companies and banks — could start to buy computers. And then DEC came along and came out with the minicomputer, and then all of a sudden smaller companies could start to buy computers. And then the PC came out and then all of a sudden individuals could start to buy computers. But the PC only ever got to hundreds of millions of people. It never got to billions of people.

Now, the smartphone has come out and it can get to billions of people.

And so it has always been this kind of trickle-down model for 50 years. We think that basically about 10 years ago the model flipped. And so we think that the model flipped to a model where, today, where the most interesting and advanced new technology now comes out for the consumer first. And then small businesses start to use it. And then medium-size businesses start to use it, and then large businesses start to use it, and then eventually the government starts to use it. But this is a complete change from the way it has always worked.

Alexia Tsotsis: It’s grassroots versus trickle-down.

Marc Andreessen: Versus trickle-down. And the reason is because – the reason fundamentally is because now that you have got these things, you have — now that you have a computer in everybody’s hand, all of a sudden all these barriers — it used to be these barriers to market entry were so big, it used to be there just weren’t that many early adopters in the world. To bring out a new technology for consumers first, you just had a very long road to go down to try to find people who actually would pay money for something.

And now all of a sudden you have got this global market of all these early adopters that have smartphones connected to the Internet, and they can just pick up their things and run with them.

And of course consumers can make buying decisions much more quickly than businesses can, because for the consumer, they either like it or they don’t, whereas businesses have to go through these long and involved processes.

So that’s the big, big, big change that’s happened. And that’s been reflected in the entrepreneurial community, where entrepreneurs, especially between 2000 and 2008, entrepreneurs really only wanted to do — for the most part wanted to do consumer software, because that’s the only software that they could actually get anybody to adopt. It became very hard to get businesses to adopt new stuff.

In the last five years, there’s been this sort of acknowledgment of the consumerization of the enterprise, which is consumer product development, design methods applied to business software, of which SaaS and cloud and all these things are examples., Evernote is an example. So now you have got the rise of this new set of companies that are sort of consumerized technology for businesses.

Then from a bottoms-up standpoint what you said is exactly right, I think, which is that the new generation of employees grew up on smartphones and tablets and touch and everything, social networking and Twitter and everything else. And so if you take a typical mainframe or, even these days, PC-based system and you give it to a 22-year-old college graduate, it’s like beaming in products out of the Stone Age. Why would you do that? Why would you force people to use all this old stuff?

And then that leads to the big thing that’s starting to happen right now, which is this “Bring Your Own Device” movement, where more and more companies are saying, well, basically, if I have to support smartphones and tablets anyway, and my CFO is probably carrying around an iPad and all my new employees are coming in with iPhones, so I have already got to support this stuff, so then I might as well encourage it.

And I might as well basically have a model where instead of issuing a company laptop to everybody or even a company phone, why don’t I just let people bring in whatever device they want and just plug-in and access it.

And then they get all excited, because then they say, well, not only are my employees going to be happier and more productive, but then I don’t have to buy them hardware anymore, so I can cut my budget. So that’s the big thing that’s starting to happen right now.

Alexia Tsotsis: So how does it affect the way people are building? I have about five companies that have made this list. Some of them are yours — Okta, which is big on the Bring Your Own Device thing, because you are logging in through Okta. Okta, Cloudera, Box, GitHub, Zendesk, and Asana. Are there any that I have missed?

Marc Andreessen: We just invested in this company called ItsOn, that we announced yesterday.

Alexia Tsotsis: Oh, the mobile –

Marc Andreessen: The mobile billing. The advantage — the thing that that’s going to be able to do is do split billing in a new way, between the business and the consumer. So on a single device you will be able to cleanly build data usage by application. So your employer can pay for your and your Workday data usage on your phone and you get to pay for your Facebook and your Hulu usage. So that will be another enabler for a lot more of the Bring Your Own Device stuff.

Marc Andreessen: We have a bunch of stealth investments. I mean, this is a big, big thing — big change for us, so we have a bunch of stealth investments — I mean, companies that haven’t talked about what they are doing yet.

Who else? I mean, there is a bunch, who else should we have –

Alexia Tsotsis: Platfora.

Marc Andreessen: What’s that, Platfora? Yeah, Platfora is the actual user interface layer on top of Hadoop. So sort of Platfora and Cloudera kind of go hand in hand.

Actually — we have another one actually that is — well, it sounds esoteric, but it actually is very relevant. We have a company called Tidemark, which is in a category. It’s called Enterprise Performance Management, which is kind of a weird term. It’s basically large-scale financial planning and analysis for big companies.

The significance is it has a — I believe it only — I don’t know if it only or primarily, but I think it only has an iPad UI. So it’s the first complex financial system for big companies, where the assumption is that the user is on an iPad.

That’s a really big deal, because that category of software, line managers and businesses have never actually used that software themselves. Instead they employ analysts to use the software who become highly trained on the software.

By putting the iPad UI all of a sudden you can have anybody in the business have access to all the financial analysis and planning. Even in a very deep sort of sector of enterprise software where most people would never see it, this change is having a big impact.

What else? Asana you mentioned.

Alexia Tsotsis: Asana, Box, Zendesk, these are the companies that I am assuming I will be using four years from now to run my business.

Marc Andreessen: Yeah, exactly! Exactly right! Then of course Workday, of course,, of course NetSuite, 37signals. We probably have three or four others in our portfolio that I am blanking on, but yeah, this is sort of the — and then by the way, the corresponding thing is that a lot of this is on how you run a business and then how you do marketing, of course; AdWords and Facebook and Twitter, all these systems and then all the enabling systems for that; so HootSuite and Marketo and –

Alexia Tsotsis: GoodData.

Marc Andreessen: Oh, another one, GoodData. So GoodData is at the intersection of kind of marketing and business. So GoodData is an actual easy-to-use analytics package. It’s sort of like a supercharged version of Excel that lets you suck in data, you can suck in all your Facebook advertising campaigns, you can suck in all your data, and you can run — you can actually, yourself, as a small business person, actually analyze and find friends and data.

Alexia Tsotsis: I have heard good things about them and they just sent us a guest post.

Marc Andreessen: They are very good. So then you add up all these companies and you are like, “Well, okay, so number one, they are all basically new companies. I think who is not on that list are all the existing companies that sell business software.”

Alexia Tsotsis: SAP, Oracle … I mean I wrote a post about this that was supremely misunderstood and then today SAP came out with SAP Jam, which is a competitor to Yammer and to Salesforce, but it’s their own socialized CRM, like HR management software. I worry about this because it’s not going to work. You can’t fight the future.

Marc Andreessen: Oh, right, right, right. I mean, the joke about SAP has always been, it’s making 50s German manufacturing methodology, implemented in 1960s software technology, delivered to 1970-style manufacturing organizations, like it’s really — yeah, the incumbency — they are still the lingering hangover from the dot-com crash.

So a lot of incumbent business software companies did what a lot of big companies actually did and other industries, media companies after the dot-com crash, which is they said, “Oh, thank God we don’t have to worry about this Internet thing. It’s over. Stick a fork in it. It’s not going to be a big deal.” And then it turned out that it actually wasn’t over, and they still haven’t adjusted.

Alexia Tsotsis: Yes, and we are watching that now. And so the other reason that I am very interested in delving deep into this space is that it seems like IPOs like Workday, Palo Alto Networks are sort of — they have metrics and analytics that Wall Street understands, more so than a Facebook; like “We are going to sell X number of this in the next year.” So it would seem like they are an antidote to, or at least less offensive than, social/consumer Internet companies are to the public markets.

Marc Andreessen: For now. The whole market goes back and forth in whether they prefer enterprise businesses or consumer businesses. The argument in favor of consumer businesses is you don’t have these crazy end of quarters like when the IT purchasing manager doesn’t buy the product and the company misses the whole quarter.

The advantage of the consumer businesses is they tend to be much broader-based, much larger number of customers, that tend to over time be a lot more predictable. The advantage of the enterprise companies is they are not as subject to consumer trend, fad, behavior.

But I would say the market is schizophrenic. So right now we are in an era where the market wants enterprise companies. I am just saying like wait a year, that will flip again; wait another year after that, that will flip again.

It’s sort of the picks and shovels thing. Like everybody — it’s like the consumer businesses get really hot and then everybody realizes that there is lots of competition and that those models have — they are complicated businesses and they have their issues, and then everybody gets all excited about picks and shovels.

And everybody rediscovers the picks and shovels analogy and says, ” Oh, the Gold Rush in California, the people that made all the money were the guys who were selling picks and shovels to the prospectors.” And then people realize the picks and shovels business is really hard, and then everybody says, “Oh, we should invest in the consumer company because they” — so it’s just –

Alexia Tsotsis: It’s cyclical.

Marc Andreessen: It’s cyclical. It’s deeply cyclical. But we are in an environment right now, to your point, where there has been huge rotation out of the consumer companies into the enterprise companies.

Alexia Tsotsis: It seems like the consumer market is starting to cool — I mean, not starting, but the signaling is there.

Marc Andreessen: Yeah. It’s unpredictable. All you need is for one of the new enterprise companies to completely whiff a quarter and their stock will collapse and then everybody will get all freaked out. I mean, it’s just a continuous — the reality is every single business is hard.

Alexia Tsotsis: I love this.

Marc Andreessen: There are no easy businesses in the world other than maybe Google, but other than that, there is no easy business anywhere in the world. So what happens is Wall Street gets enamored by the businesses that look like they are easy, until it turns out that they are not, and then Wall Street gets disillusioned and freaked out, and then rotates into the businesses that they think are going to be easy, and then they get endless disappointment. It’s like a seventh or eighth marriage at some point.

At some point the problem isn’t with your seventh wife. At some point the problem is with you.

Alexia Tsotsis: Is the solution “keep calm and carry on,” or what is the solution to this?

Marc Andreessen: There is no solution; it’s a permanent state of affairs. So this is a big part of what actually we do. A big part of why venture capital actually is important and enduring is because the public market is flighty and late-stage investors are flighty, and customers for that matter are flighty, and so you can’t — if you are running one of these companies you can’t — you just can’t rely on people being balanced. They are just not going to be.

And so you have to have a level of determination to just stick through the good times and the bad times. And you need to have investors at the core of your company who are going to support you through that.

The big advantage that we have as a venture capital firm over a hedge fund or a mutual fund is we have a 13-year lockup on our money. And so enterprise can go in and out of fashion four different times, and we can go and invest in one of these companies, and it’s okay, because we can stay the course.

And then what happens is everything tends to get better, all the products tend to get better, all the companies tend to get better over time if they are working hard at it. So we are fine. Like if everything we are investing in goes out of fashion, we are not going to change anything we do, because we can’t change anything. We are already invested in these companies; we can’t sell our stock. We don’t have to sell our stock. So we just say, we will go back to work. And then at some point it really gets exciting again.

Alexia Tsotsis: I guess the trick is to be hyper-aware.

Marc Andreessen: So the big thing we try to do is be aware of the difference between the reality and the psychology, and the reality tends to progress in a certain way and then psychology tends to whip all over the place.

It was very educational for a lot of us to go through the dot-com crash, because you remember, in 2002, like there were a number of universal truths asserted in 2002; the Internet didn’t matter, consumer Internet business was dead. Larry Ellison in 2002 came out and gave a speech and said the correct model for enterprise software, enterprise computing, will last for 1,000 years.

He said all these kids that were trying all this new stuff and it didn’t work, and now we know it didn’t work, and so the model is going to be the existing IBM and Oracle for the next 1,000 years. And everybody kind of said, hmm, you know, that makes a lot of sense, like all that innovation stuff didn’t work, and so –

Alexia Tsotsis: That’s what David Sacks said.

Marc Andreessen: Exactly, this is the fact. People reach a point where they start to get a little bit too rich, maybe a little bit too old, and they start to say these things.

And then so here we sit 10 years later and we are in the middle of a complete reinvention of everything in enterprise computing, and it’s like, okay, like that’s the reality. People happen to be excited about it again at the moment. That’s great. I am happy for that. But wait two years and they will be depressed about it again, but that won’t keep it from happening. It will still happen.

Alexia Tsotsis: It’s just like the fashion industry. So because it’s heavily fashionable now, do you see it being over in a year?

Marc Andreessen: No, I don’t mean to make a specific prediction. I don’t know if it’s a year, two years, four years. Look, all of the products are going to keep getting better. All of the trends that we are talking about are going to keep continuing. Nothing is going to stop consumerization of the enterprise. Nothing is going to stop Bring Your Own Device. Nothing is going to stop Software-as-a-Service. Nothing is going to stop cloud. All those things are just going to keep going.

I am just saying people are going to be — they are all excited about them now. At some point again they will be unexcited about them and then at some point after that they will be excited about them again. So it’s hard to draw conclusions about the importance of the trends or the progress of the trends by the current level of press coverage, the current level of Wall Street enthusiasm.

Alexia Tsotsis: So beyond the press coverage, beyond the fickleness of trends, beyond the application layers — because most of those companies are just apps — what are the real opportunities you see in the enterprise stack as it stands right now?

Marc Andreessen: Well, there is a whole bunch. So there is a big thing — there are a couple of big things that are happening. So one of the really big things that’s happening is, historically the best enterprise technology was only — it’s a trickle-down thing — the best business technology was only ever available to the biggest companies.

And so if you were a Fortune 500 company with a big IT department, you had a huge advantage over a small business that was trying to compete with you, because you just had so much more budget and staff and professionals and expertise and access to all these big vendors and you could spend tens and millions of dollars on all this stuff.

So it was very easy for — in the old world it was very easy for big companies to use IT as a weapon against small companies.

The classic was Walmart versus local retailer, right? Walmart’s advantage in logistics and in pricing and in data analytics was just so great that they could kill small retailers at will.

Today all the consumerized enterprise stuff is as easily usable by the small business as it is by the large business. In fact, it’s probably more easily usable by the small business than it is by the large business, because with a small business it’s like you can just use it, like you don’t have to go through a long process, you don’t have to have a lot of meetings, you don’t have to have committees, you don’t have to have all this stuff, you can just start picking up and using it.

So the best technology for inventory management and for financial planning and for sales-force management and for online marketing can now be used just as easily or more easily by a small business. There is an opportunity here for a shift of the balance of power for big businesses to small businesses.

And then for vendors, the companies we fund, there’s an opportunity to really dramatically expand the market, because a company like Oracle, as successful as it is, it only really has about 5,000 customers that really matter worldwide. Whereas, a company like Box or a company like GitHub could have 500,000 customers or 5 million customers that really matter, and that’s a huge change.

So market expansion, small business versus big business, what else? Oh, the shift, the other big one, the shift from CAPEX to OPEX. So the shift from buying a lot of servers and databases and software licenses and networking equipment, the shift instead to just renting it all. So the shift towards cloud services.

So we don’t have — no company that we invest in anymore actually ever buys any hardware. I mean, they buy their laptops and that’s basically it. And increasingly they might not buy their laptops, because their employees will just bring their own devices. But they don’t buy servers. They don’t buy storage devices. They don’t buy any of this stuff, they just rent on AWS. And they don’t buy sales-force automation software, they rent on

And so having sort of a much lighter-touch way for businesses to be able to get funded, you just need a much smaller budget. And that’s why you see these — you see it in the startup world, you see three or four kids with laptops who are able to go do amazing things on a global scale for no money. And I think businesses are going to figure out more and more how to do that as well.

Alexia Tsotsis: Do you think that the biggest inefficiencies are at the network layer, the database layer, or the storage layer currently?

Marc Andreessen: All the above. They are all changing. I think they are all changing.

Alexia Tsotsis: What do you think about the interplay between the enterprise market becoming more efficient and the explosion of the consumer market because you don’t have to pay for something like storage?

Marc Andreessen: I don’t know, it’s sort of all intertwined. I mean it’s all — because a lot of what businesses do is then offer consumer services based on all these changes. So it’s kind of all — that’s why I say it’s kind of all happening at the same time, a lot of the same stuff.

I would say the consumer Internet companies — in a lot of ways if you go inside the consumer Internet companies and you see how they run, it’s how all their businesses are going to run. They are going to be doing all of the same kinds of things. The big businesses are just in the process of trying to figure out how to catch up.

So everything, Hadoop and scale-out architectures and cloud services, and the whole thing it’s all — and use of new technologies like Box and GitHub, the consumer Internet companies all are just built this way. And then if you go inside a big consumer product’s company or a big manufacturing company, they are all trying to figure out how to make the jump. But it’s all kind of the same stuff.

Alexia Tsotsis: So which of the big incumbents do you think are most likely to get disrupted by this new wave of the enterprise cool kids?

Marc Andreessen: Yeah, this is the part where I get into the most trouble.

Alexia Tsotsis: That’s why we save it for the end.

Marc Andreessen: Yeah, exactly! I don’t know if I am going to — let’s see, I am going to try and figure out if I am even going to answer the question.

So I would say for sure — like the systems companies, like the companies that provide hardware, the server companies and networking companies, the bad news for them is the end customers are not going to buy as much stuff; the good news is the cloud companies are buying a lot of stuff.

Link to original article here