We All Start Somewhere

I began my career as many do in technology, as a developer. I delivered SaaS solutions to simplify how our customers used our products at CIGNA and to increase the effectiveness of automotive dealership digital marketing and operations at Dealer.com.

As I began to lead teams, my foundation as a software engineer helped keep me grounded in delivery and focused on ways to optimize delivery for our teams. By keeping my perspective of a software engineer I was able to focus on making our engineers—our most valuable company resources—as productive as possible. It’s no doubt that this perspective and approach directed our journey.

In March of 2014, the private company Dealer.com (where I was CTO) was acquired by the public company Dealertrack Technologies. A year and a half later, now as CTO of Dealertrack Technologies, we were again acquired, this time by Cox Automotive. We quickly went from a private company operating in a portion of the automotive industry to the full-blown leader with solutions across nearly every category.

Cox Automotive acquired Dealertrack Technologies for over $4 billion, the largest acquisition by Cox Enterprises[1] to date. It was a big move for Cox Automotive as they set out to change the way the world buys, sells, and owns cars. The company knew, as did the industry, that combining solutions from over 20 brands and over 40 acquisitions—including Autotrader.com, kbb.com, Manheim Auctions, and the largest dealership website provider and credit application network—was sure to create the much-needed change the industry yearned for. Our customers and consumers demanded a simplified car buying process, increased transparency, and a more collaborative, enjoyable experience. Changes only a provider like Cox Automotive could deliver.

At Cox Automotive, our mission was clear. We had to integrate our products to simplify dealership operations and the consumer experience while unifying our companies, people, and software. We had to transform ourselves while we worked to transform the industry. This change had to be done correctly and it had to stay focused on the right outcomes. For technology, this meant we had to create an environment where our scrum teams could be their best. We had to provide unifying technology, process, tools, and skills while removing as many delivery and operational barriers as possible. We were like many enterprises facing digital transformation: change or die. We chose change and off we went.

We began by tackling our delivery approach, creating a technology strategy, and embarking on a journey to transform our culture. It was a tall order with a lot of moving parts, and it required us to be decisive and move swiftly. Time was of the essence!

Enter AWS

There are so many complexities involved in transforming a technology organization. While you can certainly exchange your datacenter for a cloud provider, if that’s the only change you implement you’re missing out on so much…and so are your customers.

I’m sure you’ve heard comparisons of the evolution of the cloud to that of the energy industry. Just as electricity is available when you need it and you only pay for what you use, so are cloud services. However, I have noticed that many people gloss over the revolution embedded in that simple comparison. As a result, I spent time at Cox Automotive talking about the history of electricity, from water wheels to private power plants to today’s public utilities. Think about all that we now take for granted in our everyday experiences. How much of what we accept as commonplace would exist without the mass adoption of public utilities and electricity? Think about all the jobs, businesses, industries that are only possible because of the availability of low-cost electricity. In the culinary industry, we can thank electricity for enabling the creation and mass adoption of appliances, restaurants as commonplace, culinary schools teaching a wide variety of techniques, and the reimagination of the home kitchen. Now, think about what the cloud, offering the same availability and consumption model (in many cases better consumption models), will create.

With that comparison in mind, the cloud is clearly much more than just a datacenter replacement. In order to free yourself of the past and create the future, you must change how you build software. We quickly learned that if you don’t: (1) automate how you deploy, test, and rollback, (2) automate your infrastructure and security, and (3) take advantage of the potential to significantly increase your speed to market, then “the cloud” can only do so much for your business.

At Cox Automotive, we looked to not only change how we prioritized and measured our work and where we ran our software but how we approached software development on the whole. We implemented our version of Scaled Agile Framework across the enterprise, aligning on delivery cadence, transparency, and collaboration. We shifted our organization to better support a “You Build It, You Run It” approach to delivery to emphasize automation, telemetry, tooling, and, most importantly, ownership.

The cloud for us was way more than avoiding having to run datacenters and provision servers. We saw big value in AWS services like Amazon Aurora, AWS Lambda, and Amazon SageMaker, and declared that we were focused on:

1. Avoiding non-differentiated work, favoring fully-managed services.
2. Delivering value quickly, favoring speed over insurance policies, like avoiding lock-in by not fully embracing the services that existed. No, we will not solve this with another layer of abstraction, and Yes, you can use AWS Lambda and other proprietary AWS services.
3. Unifying our teams through a common platform, common vernacular, and common skillset. Multi-cloud was off the table but left as a two-way-door [2] in case certain workloads required different solutions in the future.

At the end of the day, our business required speed not insurances. Our industry demanded innovation, and they needed it yesterday. We had to move quickly. With eyes wide open, we chose to embrace services that would make us faster versus worrying too much about how we’d migrate off of a service down the road. I’ve watched many teams and businesses invest a lot of money to create the illusion that changing the technology down the road would be easier. Frankly, the majority of time the change never happened, was never needed, or was never that simple. Be thoughtful about how much you plan for “what if” versus delivering for “what is.”

Be Smart, But Be Fast

We know from history that businesses come and go. Some transform and last a long time while others fail to adjust. My advice is to find the right path for your organization and run along it. Those slow to change will be left behind and those too risk averse may not make it to the finish line. Remember, our businesses were started by taking major risks, over and over. Transformation requires fortitude and big bets. Be smart with the bets you make but you must make a big bet to move quickly.

The good news is that you can start by simply getting started. You can make a big bet shortly thereafter. We see that adoption is often a slow ramp that eventually becomes exponential. Find ways to “hockey stick” your adoption, there are people around you who can help make that happen.

Bryan Landerman
Enterprise Strategist & Evangelist @ AWS
@bryanlanderman
brylando@amazon.com

[1] Cox Enterprises is the parent company of Cox Automotive. It’s owned by the Cox Family in Atlanta, Georgia, and Cox Automotive, Cox Communications, and Cox Media are the divisions owned and operated by Cox Enterprises.

[2] At Amazon, we talk about decisions in terms of one-way and two-way doors. If it’s a decision that is reversible, i.e. you can go back through the door from which you came, then we can make decisions more quickly and be more accepting of failing fast. If it’s a one-way door, we spend more time upfront ensuring that it’s the right direction and decision.