Q&A: Former Autodesk CEO Carol Bartz on the past, present and future of construction tech
Construction technology is taking off as developers large and small seek digital solutions for challenges across the job site, and investors respond in kind. The rise of CAD (and, later, BIM) in the AEC industry saw a similar trend, and the factors that drove its growth — such as the need to customize platforms for different user groups and new hardware options — are mirrored in the uptick in third-party apps and the growth of mobile and the cloud.
To put the current construction-tech trend in context, Construction Dive spoke with former Autodesk CEO Carol Bartz, who led the company from 1992 to 2006 as it grew from one of many digifab software options to the undisputed leader of the pack in the U.S. Today, Bartz is the lead director on the board of Cisco Systems and a member of the board of San Francisco–based AEC tech startup PlanGrid, whose mobile app digitizes project documentation.
Construction Dive talked with Bartz about how Autodesk handled segmentation, the current landscape for construction tech and the potential for mobile.
This interview has been edited and condensed.
Autodesk grew during your time there into the market leader for design and construction software. To set up the conversation about these new, smaller software providers, what were some of the factors to which you attribute that growth?
BARTZ: We really worked hard to stay on the technology curve. We had open APIs, which was not considered a standard concept back in the 1990s and early 2000s. We had thousands of independent software developers providing software that enhanced what AutoCAD and some of the other products did. That was unique. And we developed a strong channel. We went out and found architects and people in civil [engineering] and construction to represent and support our product to their [peers]. That was not a standard IT practice in the ’90s and early ’00s.
You want to say it's because we had the best technology. Not true. You have to have the entire ecosystem package, which is how do you partner with people, how do you actually get your product to market, and how do you support the product once it's in the market? It's changed today because you get your product to market online, you get support online, your channel, everything [online]. It doesn't matter what the period is, you still have the same basic questions.
You led Autodesk under the 'fail fast forward' mantra. What effect did that have?
BARTZ: In the late ’80s and [early] ’90s, Autodesk was essentially a one-product company: AutoCAD. [Whether you] were a mechanical engineer or a civil engineer or an architect or a construction company, you figured out how to make AutoCAD work for what you needed because we didn't do that for you.
The internet came in the late ’90s, and everybody froze and said, 'Well our engineers, our architects, would never use the internet because it's not secure, not safe. They don't need it.' That's when we had to kick into motion the fact that we have to try some things and we have try them fast and it's OK to fail. Try to figure out quickly that whatever you're trying isn't going to work and hopefully it moves you forward. Try it again. Try something else.
The idea was to start looking at segments and figure out how to enhance the product so that the vertical markets felt the product was specialized toward their needs. Back then we worried about what the box looked like, for heaven’s sake. 'Fail fast forward' was getting people away from the idea that the only thing they had to do was add 20 new features for every AutoCAD release.
What about the cloud and the rise of more portable and, eventually, fully mobile devices? Did that make a difference?
BARTZ: It’s the same concept. Any time you dramatically change either how you get the product or how you support or service the product — for example, you go from a subscription service to a cloud-based delivery service, or you go from a box to downloading the software [from the internet] — you have to get out of your comfort zone.
Is there a relationship between when Autodesk started tailoring AutoCAD for different segments in the ’90s and what’s going on in the construction-tech space today around the emergence and development of smaller, third-party software providers?
BARTZ: There's a total relationship. You could look at a database company, for example, and say, 'They're missing A, B and C. I could do A, B and C, and people will really want that.' That's all true, but sometimes it doesn’t make it a whole company. This will go on and on and on. [But] not one company is interested in every tiny little aspect of what their customers do.
In the early days, Autodesk paid no attention to landscaping. Now, why would you? Well, every time a building is put on this earth, something happens around it called landscaping. We were happy as heck that there were over 50 companies that provided landscaping [design software] that coordinated well with AutoCAD. We didn't want that business but the customer needed a solution, and small developers out there could make money by providing it. Voila, it all works.
How can those third-party software companies today best determine whether they’ll fill a gap or be redundant?
BARTZ: Look at the size of the market and the time that's wasted, the real money that's wasted. Any time you have that kind of a problem, you need answers. To me, getting my laundry delivered is not a real problem. Companies like PlanGrid are looking at the construction process and seeing where the holes are and are coming up with mobile solutions. Fix these [kinds of problems] in an integrated, seamless way that everybody trusts and could use — you've got a huge opportunity.
With Autodesk, and other players, so prevalent in the space, what advice do you have for startups looking to provide these specific solutions?
BARTZ: When you have a behemoth like Autodesk, you have to pay attention to where that giant puts its foot next. You have to integrate with them, hopefully you integrate with their blessing, and if not, you do as good a job as possible. Then you figure out later on whether you are truly going to be able to be a standalone public company or you're acquired or you merge with a couple others your size — there are so many options, but they are real options. Everybody doesn't have to bang a gong at the New York Stock Exchange. That's its own set of problems. If you can gainfully employ people that are making good money and coming to work satisfied every day, working on a real-world problem in conjunction with Autodesk and other companies, that's wonderful.
When you started at Autodesk, what did you have to do to get yourself up to speed on the needs of architects, contractors and engineers that were using the software?
BARTZ: I did a lot of listening. More important, I had a great group of people working for me that knew what they were doing and could teach this new, inexperienced-in-the-field CEO what the hell to talk about. I give them all the credit. If you don't go and find out how your customers are truly using the product, and listen very carefully to what they like and don't like — and they are very happy if you sit and shut your mouth for a while and not appear defensive — you learn so much.
What would be an example of that?
BARTZ: I wanted to visit all the main verticals. So, the first time I sat down with an architect, I asked him what he thought about CAD and Autodesk and the normal stuff and he goes, 'Well, I use AutoCAD, but I don't really like it because it's for the engineer.' Then I went and talked to an engineer. What did I hear? 'I use AutoCAD, but I don't really like it because it's designed for the architect.' I thought, well who the heck thinks this product is for them? It had become so generalized.
They all [indicated] it was important because they were frustrated that it wasn't specialized enough for what they really wanted and they didn't feel that they were the main reason the product existed. That's one trip and you already have three years' worth of action to do.
To your earlier point about solving specialized problems, PlanGrid, as one example, addresses the issue of paper project documentation with a mobile digital solution. What are some other areas in the space today where smaller software providers can help provide a solution?
BARTZ: We have only begun to understand what mobility means. There still aren’t enough mobile devices out in the field. Job site phones were a big deal. GPS came along and you could [more easily] figure out if you should dig a hole or if you'd run into cables. Mobility in the field is in its infancy as far as what it's going to be able to do, and so you don't have to go much further than that. Then you have to say: What kind of mobility? Who needs it? How do you make [using mobile technology] reasonably efficient, so that instead of adding to users’ workloads it actually enhances their work? That still has to happen.
Follow Hallie Busta on Twitter