Robert, can you tell us something about yourself and your background before founding Tableau, LLC?
I had the good fortune to grow up in the San Francisco Bay Area – the place known as Silicon Valley – and it’s only now that I live in a different area that I’m starting to realize how unique the Bay Area environment is and how the environment there breeds a certain kind of can-do attitude.
Back in the early 1980’s – at the ripe old age of 16 – I landed my first job as a computer programmer. I was already a “seasoned” programmer, having started programming when I was about 11. Fortunately, the president of that first company took me under his wing and showed me the difference between “coding” and “engineering”. But he didn’t stop there. He also gave me the guidance and opportunity to learn how to run a small, high-tech company, teaching me how to negotiate with vendors and customers, how to recruit and develop employees, how to consider the financial decisions inherent in any business…And on and on…Ultimately, he promoted me to be the VP of Engineering, we grew the company together, and a few years later he and I negotiated the sale of that company to a much larger public company.Looking back 25 years I can see just how unusual that opportunity was. Part of what makes the story unusual is that my mentor in that first company didn’t look first at my age or experience. Instead, he looked at me and my capabilities, gave me guidance, and then gave me an amazing amount of latitude to sink or swim. Many people think that the Internet is the technology/inflection point that created opportunities for 20 year old CEOs. I disagree. I think the Bay Area – and more broadly the high-tech industry – has long had a kind of de facto meritocracy where proving you can get the job done – really delivering – trumps many other value systems.
Anyway, I’ve kept busy since that professional start 25 years ago. Tableau is actually the fourth company of which I’ve been an owner, founder, or both. After we sold that first company some colleagues and I created a software venture which we sold to Microsoft in the mid 1990’s, and Tableau has been keeping me busy since 2003.
And the lessons of those earlier ventures still resonate. Hire great people. Make it clear what the expectations are. Guide them. Then give them a lot of room to run!
Tableau was founded in 2003 and is now a major player in the computer forensics hardware marketplace – how do you account for that success?
By many objective measures it’s fair to say Tableau is a “success.” However, I’m always reluctant to use that term. “Success” is very often confused with a conclusion, but a thriving business never “concludes.” Instead, I like to remind people – our employees and even our business partners – that whatever Tableau is today is nothing more than the starting point for the rest of our business’s history.
Another reason I don’t like to use the term “success” is because it can easily breed complacency. Being a market leader is never a “right”, it’s a position which must be re-earned each day.
What does your role as President involve? What is a typical day like?
Every day starts with breakfast with my children and I’m home every evening for dinner with my family. The only exceptions are those days when I’m travelling, and I try to keep travel to a minimum.
Why, you ask, did I start with the part of my day which doesn’t involve work? The answer goes to the way we try to run the business. Many companies run day to day based on the financials, based on a review of the profit and loss statements and the balance sheet. We don’t run Tableau that way. Instead, we try to run Tableau based on a set of values.
For example, we value the idea of building the best products that meet real customer needs and selling those products at a fair price. If we have competition in a given product, our product simply needs to be better than the competitor’s, or sold at a better price, or both – or we shouldn’t build the product! Nothing profound in those ideas, but it’s amazing how often simple ideas like that are lost when companies are making far-ranging executive decisions.
Honest labor is another simple value at Tableau. In between some of my startups I used to do consulting work where I was paid by the hour. I saw many billing abuses – outright lapses in ethics – among my contracting peers, and I see a broad lack of self-honesty among many full-time employees who grossly overestimate the amount of real work they are actually putting into a job (there was an interesting article to this effect in the Wall Street Journal recently). So at Tableau the standard is simple. Each employee should give the company 40 honest hours of work each week, no more, no less. I know from my contracting days how demanding – draining, really – an honest 8-hour day of engineering work can be. We figure if we get that honest amount of work from each employee then we’re doing just fine as a company. And in the bargain each employee gets to go home to their families, hobbies and other personal pursuits, letting them return to the office refreshed rather than resentful.
Don’t get me wrong. My job isn’t a bunch of “values” fluff! We do look at financial results to ensure we’re running a viable business, and I spend a lot of my own time at the office dealing with very concrete, tactical issues. Creating a good presentation for a trade show can take several man-days of my time. Making sure operations is running smoothly takes time. Making sure customer support issues are being dealt with effectively takes time. Making sure we’re listening to what marketing & sales are hearing from customers takes time. I spend a lot of time with the engineering team working on the architecture and implementation of new product. And, yes, I still do hands-on engineering work!
So it’s not simple to describe a “typical day”. But, it is pretty easy to describe the values that guide each day.
Can you give us a brief overview of the current product line from Tableau – and perhaps a sneak peek at what to expect in the future?
Today Tableau has three main product lines: forensic bridges (a.k.a. write blockers), forensic duplicators, and hardware accelerators for password recovery. The bridges and duplicators are both acquisition technologies – devices which capture the data at the front-end of the digital evidence lifecycle. Hardware accelerators fit in the analysis phase of the digital evidence lifecycle, providing a tool for unlocking the increasingly large quantities of encrypted data captured by forensic practitioners.
We prefer not to pre-announce products. It’s a running joke in the tech industry the way some companies will announce “products” that may never see the light of day, and we’d like to avoid being lumped together with such companies. However, I can give you a flavor of things to come.
In acquisition technologies there are two primary issues: compatibility and speed. You’ll see us continue to grow the range of device interfaces with which we have compatible products, and you’ll continue to see us improve compatibility with specific devices. Most users have probably already guessed that SAS will be the next major write-blocker we’ll be adding to our lineup. In fact, we’re doing more ground-up engineering on high-performance SAS technology than we’ve done on any previous write-blocker.
You’ll also see us continue to push the performance of our other products to higher and higher levels. Just as with our T35es, we’ve added a host-side SATA port to our newest “OEM” products, the T3458is and T34589is, so you can now use either FireWire800 or SATA to connect to the host computer. In fact, our new OEM products let you connect both FireWire800 and SATA at the same time and we use some new patent-pending technology to pick the best available interface for each acquisition.
In the course of those projects we also studied and fixed the issues which were holding back the performance of SATA/eSATA-based acquisition. And the T3458is and T34589is both incorporate a re-designed internal data path and that has let us boost the performance of existing functions like USB write blocking. Our new T9 FireWire write blocker uses some other new patent-pending technology, and we’ll be reusing that technology in a next-generation USB write blocker.
I’ll also share with you that we see a large and growing need for reasonably priced, scalable storage solutions. The current storage products on the market are ok, but we think it’s time to introduce some new classes of storage solutions into the forensics community. And that’s as close as I’m going to get to pre-announcing anything!
What trends do you see in the forensic hardware business and what new challenges do you envisage in the future?
By far we think the biggest challenge is the sheer volume of data. The existing acquisition methodologies and software tools will not be able to scale as quickly as the datasets which are being acquired.
Unfortunately, there are no easy solutions to these problems. The companies that introduce new paradigms for analyzing and managing these huge datasets will be heroes in the forensics community. I hope we’ll be one of those companies.
Can you tell us something about the development process at Tableau? How do you decide which products/features should be added or improved?
There are really two parts to your question. The first part asks how we decide what products to build. The second part asks how we go about building those products.
The first part is easy. We listen to our customers and resellers. If you’re listening, your customers will always tell you what they think they need. The subtlety lies in translating those needs into viable products. Customer A may wish for features X, Y, and Z, while customer B may wish for features X, Y, and T. If a lot of customers want X and Y, and if we think we have the technology to build X and Y cost effectively, then that’s a pretty good candidate for a new product. Features Z and T may also be good candidates, and frequently we can add those – often at no additional charge – through firmware updates.
But there’s a pitfall here. Once you’ve entered a market and you’ve proven you build good products, then customers regularly come to you with many, many product ideas and requests. One of our biggest challenges is learning how to say “yes” to the product ideas where we can really excel and “no” to the ideas where we’re just not positioned well as a company. Cell phone forensics is a great example of a “no” for us. It’s a large and growing field, but unfortunately our core strengths as a company – in this case the technologies we’re very familiar with – don’t position us all that well to become a great cell phone forensics company. So, as painful as it is, we’ve left that market to others. On the other hand, our vast experience in I/O technologies used in storage applications positions us very well to enter other markets with compelling products.
To the second part of your question, how do we go about developing the products? Most importantly, we develop all core technologies and products in house, and about 40-45% of our staff is in engineering. That gives us great control over the features and quality of our products. We also believe strongly in nurturing our engineers through regular coaching, mentoring, and design reviews.
There’s an interesting tie-in with the values-based management philosophy I talked about a few questions earlier. In our engineering meetings we spend a lot of time focusing on the “why” behind good engineering practices, and we use every design review as an opportunity to share and teach sound engineering practices. Over time that raises the capability level of our team, and today I think our engineering output capacity is probably 3x that of just two years ago. What that means for customers is that our rate of technology development and product introductions is going to continue to accelerate, hopefully solving more and more of our customers’ problems with each passing year.
How does Tableau aim to differentiate itself from other hardware vendors?
That’s easy: quality at every level. We put more effort into conceiving, designing, and manufacturing our products than any of our competitors. Then we back that with a consistent focus on supporting those products once in the field.
The I/O technologies used in storage – where a lot of our products play – are notorious for incompatibilities. What a lot of users may not realize is that bad engineering is really quite prevalent in high tech products, and many of the compatibility issues we face are the result of other devices which violate specs and just simply have bugs.
So, when you consider the huge number of different storage devices in the industry you realize it’s just about impossible to guarantee compatibility with every product. But, we can guarantee that we will analyze every compatibility problem reported by your users and then issue firmware updates to address those problems. We have a policy of 100% compatibility. That means that we will chase down every compatibility issue and get to the root cause. Unless there is a fundamental electrical incompatibility – which is pretty rare – we’ll find a way to modify the firmware in our products to resolve the compatibility issue.
We also have a policy of layering new features into a product over that product’s lifetime. Most features can be added into units already in the field through our firmware update utility, and we generally don’t charge anything for those new features. I’ll give you a couple of examples.
Recently we posted a new firmware update for our model T35es SATA/IDE Forensic Bridge which boosts eSATA write blocking performance by up to 34%. That’s a pretty big deal for users trying to image large hard disks. And the update is completely free.
As a different example, over the next few months we’ll be releasing four important new features for our TD1 Forensic Duplicator: USB keyboard support, FAT32 long filename support, large file support, and the best of all, direct support for Encase .e01 files with hardware-based, real-time compression! Again, those features will be available as free updates to all existing TD1 customers.
So here’s the important message I’d like to get out to your readers: Check our web site for firmware updates! We’re continually amazed at the large number of our end users who don’t know about the after-sale support we provide for our products. The firmware updates we issue make our products better – offering improved compatibility, performance, and new features – and the firmware updates don’t cost the users anything except the time needed to run the update.
Validation of a particular methodology or tool is of paramount importance when presenting evidence in court. How do approach this issue as a developer/manufacturer of forensic hardware?
We consistently make the point to our end users that they, not we, are the ones who will be testifying on the stand. So, the onus is on the users to validate their tools and to be able to speak intelligently as to the function of those tools.
In that context we have a role as an information provider. We openly and transparently explain the function of our tools whenever a user has a bona fide question. We use speaking opportunities at trade shows and conferences to provide our users with genuine education about the behavior and usage of forensic technologies – I personally abhor presentations which are nothing more than thinly veiled product pitches!
We also support the work of agencies like N.I.S.T. (the U.S. National Institute of Standards and Technology) to help them define requirements for products like ours and then to support their efforts in testing those products.
And we’ve recently rolled out another free software product, TabTest. The development of TabTest was in direct response to user requests for a tool which could be used to validate and re-validate our equipment. It’s a nice little Windows application that graphically guides the user through the a pre-defined test suite for each of our products. Users can then take the TabTest reports and incorporate them directly in their lab/equipment documentation.
What is the most common question you're asked about the Tableau product line or forensic hardware in general?
How do I make it go faster?
Again, the rapid growth in datasets is increasing the time required to acquire and then analyze the evidence. It’s also making it quite cumbersome and time-consuming to manage the large amounts of data.
What is the most rewarding part of your job? What aspect of your job do you find most challenging?
There are two aspects which I find equally rewarding: 1) building and developing the team and then watching the team perform, and 2) conceiving of a product idea or vision and then seeing that idea realized as a tangible product which does well in the market.
The most challenging aspect has to do with complexity management. Running a business involves the interaction of hundreds or even thousands of variables, and it doesn’t get any easier as the business gets larger. To deal with that you try to be very clear in your strategy and decisions, you hire and nurture great people, and you decompose the complexity so each part of your team can deal with a manageable chunk. It’s easier to write it down than to make it happen day in and day out!
What do you do to relax when you're not working? What are your plans for the future?
I spend time with my wife and children. Turn the cell phone off. Refrain from checking e-mail. Tableau is my fourth company – and it may not be my last. But, I get one precious chance to be a good father and husband, so I’m going to make the best of it!
As for the future, who knows? Eight years ago I never thought I’d live in the U.S. Midwest. Now we’ve lived here for six years and love it. Life can be somewhat capricious, and we’re happy to focus on the here-and-now.