The five-minute CIO: Ross Piper, Dropbox

11 Mar 2016336 Shares

Share on FacebookTweet about this on TwitterShare on LinkedInShare on Google+Pin on PinterestShare on RedditEmail this to someone

Ross Piper, VP of Enterprise at Dropbox

Share on FacebookTweet about this on TwitterShare on LinkedInShare on Google+Pin on PinterestShare on RedditEmail this to someone

“One of the most exciting parts of our growth isn’t just the 500m users, but it’s the 2.8bn connections that have been created by users,” says Ross Piper, vice-president of enterprise at Dropbox.

Ross Piper is responsible for driving Dropbox’s growth in the enterprise market, building on its presence in 97pc of Fortune 500 companies, and spearheading the company’s efforts to make IT administrators’ lives easier and more productive.

Earlier this week, Dropbox celebrated reaching the 500m-user milestone.

Prior to Dropbox, Piper served as senior vice-president of enterprise strategy and alliances at Salesforce.com.  While there, he was responsible for the company’s engagement with the enterprise IT community, including strategic alliances, product development, and customer innovation programmes.

Piper has led the strategy to grow Dropbox’s customer base from 100,000 paying businesses just a year ago to 150,000 today. New customers include NewsCorp, Expedia, Zillow, and Meredith Corporation.

Dropbox used to be fundamentally about storage but now it is changing how we view productivity and collaboration. How did this come about?

Dropbox started with end users, no question, and the consumerisation of IT tag is never one I’ve been fond of because it has been overused by the industry.

But the substance behind it is true – it’s not about consumers it’s about end users and by Dropbox focusing on being a great experience and providing real value to them that’s how it grew to 500m users.

That’s also how 500m users brought Dropbox into 8m businesses, including 99pc of the Fortune 500, and that same trajectory plays itself out in every region.

The thing that’s been coolest about that evolution is that it has gone from the first core use of Dropbox was access, making sure you get your files on every device and every application, which was a big part of the early growth. But the part that has changed in that time has been the importance of collaboration and the importance of connections.

Dropbox, through its shared folder capability, enables people to create connections that are persistent; not just sending a link or an attachment, you are creating a shared space for you and another person, where you and hundreds and, in some cases, in the largest scale, of one person sharing with millions through web-based links.

It is really powerful connections that it has enabled and, for us, one of the most exciting parts of our growth isn’t just the 500m users, but it’s the 2.8bn connections that have been created by users.

To me, that volume of connections is actually the most powerful part. When you think about it those are persistent connections between individuals that represent the network by which people actually get work done.

The other cool thing is that growth of connections is growing faster, even as fast as user growth. We announced 2.8bn connections by the end of last year and when we announced it the previous year it was 1.3bn.

So, it is still growing at over 100pc a year and it’s an extraordinarily large number and no competitor has suggested an alternative number, so we don’t really know how to benchmark it. Competitors numbers are smaller or they don’t even draw importance to it as we do. But to us that is the life’s blood of Dropbox, it’s those connections that users make.

That’s one of the big evolutions in terms of what people are doing when they have a platform like ours, going from access into connections and a huge collaboration network.

How has Dropbox changed how people work and collaborate? And how do you work with the new generation of productivity apps?

The second big evolution has been the evolution from being an end-user tool into an enterprise tool. The end user has already brought it in and I think there is a dynamic that’s occurred within companies where, for decades, we have seen lots of different tools migrate to being end-user focused.

The one that has taken the longest period of time is what we call workflow – how people get stuff done, how they create a document, how they change it, how they evolve it, how they make it anew, how they share it with others and how they complete it into a finished document.

A document could be anything from a word document to a powerpoint to a huge product catalogue in AutoCAD or CAD design tools or videos or audio; it doesn’t really matter what the type of content is.

The way people actually create things has changed. For decades, companies have tried to create top-down hierarchical workflow tools and content management systems and all this stuff to try to guide people on how to work.

In the meantime, through platforms like Dropbox and tools like the enormous number of productivity apps that are now out there, workers are redefining this themselves.

They are empowered to figure out their own workflows and they are no longer dependent on another system to integrate with other applications to go across functional barriers within their company, barriers between companies, geographic barriers, etc. They are really empowered with platforms like Dropbox to redefine their work in incredibly innovative and productive ways that are driving huge innovation.

The two keys to that from our perspective have been that collaboration network, the ability to share with anyone and knowing there are people on the other side who can work on the same platform.

The next piece is with all that rise in the innovation of productivity tools – some driven by big platforms like Office, Adobe and Google, but a lot of it is driven by hundreds of thousands of special purpose applications like FileDraft, which is the de facto standard for screenwriters in the media industry, which drives huge value for people to create things.

The other part about Dropbox is that with 300,000 apps built off the API we are the predominant platform underpinning any of those productivity tools. There are more than 140bn documents in Dropbox. We are the largest repository of Office documents, we are the largest repository of Adobe documents. Dropbox is the platform for more applications than any other platform

And through this integration you have integration to companies like Slack, Evernote and FinalDraft and a host of special purpose apps that mean that you can use those apps and switch between them and be collaborating with other people seamlessly and not worry about what apps they are using or whether their apps work with your platform; it just makes it ubiquitous and easy to cross all of the divides.

Some of these companies are really cool and some of them are building big product sets and others like FinalDraft are transforming a specific profession. It ranges and it is about giving people the ability to tap into all of that.

This collaboration and recognition of Dropbox as a work tool, did this lead to Dropbox for Business?

For sure. All of that excitement and energy is what led Dropbox to launch Dropbox for Business three years ago in February 2013. Because we had been embraced by businesses they were saying ‘give me the tools that allow me to manage it.’ They were saying: ‘Give me the tools that allow me to take advantage of when there are lots of Dropbox users in my company’.

What has been really exciting because of the pent-up demand is that in the three years since we launched Dropbox for Business we now have over 150,000 paying companies using Dropbox for Business.

I was at Salesforce for several years before I came to Dropbox and it took us 10 years to get to 100,000 companies. So Dropbox did it in less than two years and has gotten to 150,000 paying businesses, so, to me, it has been an extraordinary growth.

That includes some of the largest companies in the world. Obviously, the 150,000 includes businesses from small, medium to large but we have also seen companies like News Corp who saw that usage explode.

It did a pilot with us out of New York that went really well and now it has rolled it out to all 23,000 employees.

News Corp is a big, very successful media company with 10 different major business channels and they range from Dow Jones and Wall Street Journal to News Corp UK, which has a number of major newspapers in the UK, to Harper Collins, to a host of different other businesses that are all related but have a fair amount of independence. What they have done with Dropbox is use it as a platform to bring all of those 10 businesses together to enable cross-functional collaboration across those organisational divides in a lightweight but very powerful way that none of the other tools it used had been able to accomplish.

It is easier to use Dropbox to drive unification across all the countries it operates in. It is able to drive the modernisation of its overall IT infrastructure.

In terms of IT strategy, how do you manage the complexity of your infrastructure and all the data you have to store? 

There are a couple of things about our infrastructure that are unique. One of them is the fundamental architecture. We don’t necessarily store files as a file.

If you take a 400Mb video as an example. What we do when a user saves that into Dropbox is we actually break that into a hundred 4Mb file blocks. We then encrypt all of those individual blocks in transit or at the data level through SSL tunnels to the storage service and kept at 256-bit highest level industry standard encryption and then they are stored in the storage service. All the metadata about those file blocks, also through encrypted tunnels, are a completely distinct service.

The reason we did that was for performance but also for security benefits.

For performance reasons, if I am working on a 400Mb video and using it with Dropbox and sharing with 50 other people, each of those 50 people may have two or three devices. So there are around 100 different places where that video exists. If I want to make a change and want everyone to see that change, with a lot of traditional tools every single one of those 100 versions would have to reload in a new 400Mb version.

For us, we have already broken it into 100 4Mb file blocks and without even reading the file we know which of those file blocks changed.

And then we sync the changes to the just the one 4Mb file block that changes and fit it in seconds to all 100 versions of that 400Mb video.

You can imagine the performance that this requires when people are working on something together. And a lot of users think ‘yeah, I love Dropbox because it just works’, but there is a lot of sophistication around making that work.

The other cool part of that is that is how we handle version history. So, for us, when someone makes that change to the 4Mb block we now have the original 400Mb collection of blocks and one additional 400Mb file block. If there were 10 more versions, each of those versions is just another 4Mb file block, so we can revert to previous versions really easily and allow users to swap it.

When a person reverts to the previous version, instead of having to reload the previous file, we just swap the file blocks out to the version they are looking for.

This was one of the founding architectural principles in the very early days of Dropbox.

The whole service as it has scaled has been built on top of that. A lot of competitors are now trying to match it. We call it Delta Sync and they are all trying to create variants but they have to rebuild their core to make it work across everything whereas it has been part of our core from the very beginning.

Security in the cloud is paramount. How do you protect 140bn documents?

Each of those file blocks is individually encrypted. They are stored at random within the storage service with 1bn new files per day. Imagine 1bn files – that’s well more than 10bn file blocks.

If a person wanted to find that file and hack into service they would first have to try and get through the encryption tunnels, find the right 100 4Mb blocks amongst 10s of millions of file blocks that were saved that day let alone the whole history of file blocks and they would have to find those 100 and individually unencrypt each one of them. They would have to do the same hacking into the encryption and tunnels in order to get to the metadata about those files because it is computationally and massively difficult for somebody to have 100Mb random file blocks and piece together without the metadata.

It is a fundamentally secure architecture.

Editor John Kennedy is an award-winning technology journalist.

editorial@siliconrepublic.com