Magazine Article | October 1, 2018

The Perfect Equation To Find The Right Developers

Source: Software Executive magazine

By Abby Sorensen, Chief Editor

PrecisionLender grew to 16 employees in the first three years after it was founded; then, a revamped hiring process scaled headcount to nearly 9x that number in the next three years as its cofounder learned how to find the right kind of developers.

Carl Ryden
In 2012 Carl Ryden had what he calls “a brush with greatness”: Alexis Ohanian was then a speaker at theBusiness of Software conference Ryden regularly attends. Seven years later, the company Ohanian cofounded, Reddit, was valued at $1.8 billion and he was married to tennis star Serena Williams. Ryden, the co-founder and CEO of PrecisionLender, has done well for himself, too. PrecisionLender, a sales and negotiation solution that empowers bankers with actionable, in-the-moment insights and coaching works with more than 200 banks globally, recently expanded internationally and is hiring rapidly. Being at the same place at the same time as Ohanian is really just a fun story to tell; his presentation in 2012 didn’t drastically transform Ryden’s company. Instead, another speaker did. Mikey Trafton’s talk, “How to Build a World Class Culture in 3 Easy Steps,” was “life changing” for Ryden.

So life changing, in fact, that the learnings took Ryden’s company from a slow-growth startup to an international, high-growth banking software powerhouse. Case in point: After being founded by a team of five in 2009 the company had grown to only 16 people by the end of 2013. Then, in roughly the same amount of time between 2014 and 2018, PrecisionLender’s employee count grew to around 140 (and counting — Ryden projects it will reach around 180 by the end of this year). And it wasn’t because of a sudden injection of venture capital. The growth was thanks to a recommitment to company values and a reevaluation of its hiring process, especially the members of its development team.

Ryden (pronounced rye-deen) possesses that special combination of engineering chops and a deep-seated desire to fine-tune his business acumen. He remembers sitting in a closet building the software, knowing he had just a few months to ship it. And to this day he still codes, a fact that is stated right in PrecisionLender’s engineering job description: “The CEO and VP of Software Engineering develop software when their other responsibilities allow.” Those “other responsibilities” include a roll-up-your-sleeves approach to fixing the company’s hiring process, one that took many years and several missteps to refine. Along the way, Ryden has adapted his own philosophy about what makes a good developer.

HIRING MISTAKE #1: GET WHOEVER YOU CAN

From 2009 until 2011, PrecisionLender was far less strict with its hiring process. The co-founders had credibility from prior business ventures, but the company was so new that it had no real office and no real software, which meant it had no way to attract the software engineers it needed to sign on its first customer by January 2010. At this stage, the candidates were good and talented people but not what the company needed to scale and grow. In fact, “They were the only people we could hire at the time,” Ryden says.

HIRING MISTAKE #2: DO WHAT THE WORLD TELLS YOU

By 2012, PrecisionLender had a real office and just shy of 20 employees. That was the point when Ryden emphatically stated, “With God as my witness, from this day forward, I will hire only adults.” So that’s what he did, starting with an adult who was tasked with rebuilding the company’s development team. The next three developers hired are still with the company today.

Next up was to hire a transactional sales team to build on PrecisionLender’s few million in recurring revenue, because as Ryden puts it, “That’s what the world told us we should do.” By 2014, the company fired its entire commission-based sales team, saw its sales double and then double again, and went to work retooling its hiring strategy with more adults.

Taking a page from the Jim Collins’ Good To Great playbook, the company focused on getting the right people on the bus. This included two hires who showed up on day one without strictly defined job descriptions. One now runs the talent and revenue side of the company and the other oversees operations. They’ve spent time in marketing, client development, engineering, and IT. Both have proved capable of hiring people who can perform better than they can in a given role. Both are still with the company.

"Everything is about empathy for solving the problem. ... If you don’t understand the problem you’re solving, you have no way of judging a good architecture except for what’s good and easy for a developer to maintain."

HIRING MISTAKE #3: GOOD CONTRIBUTORS DON’T ALWAYS MAKE GOOD MANAGERS

By the start of 2014, with the transactional sales team gone, PrecisionLender had its core employees in place. Its headcount crept up to about 30. Ryden and his management team took their annual trip to the Business of Software conference and listened to Mikey Trafton’s talk on culture (this one titled “Recruiting A Bad-Ass Team”). They wrote down values, including a list of the only things PrecisionLender would ever fire someone for. They set out to hire “A” players, avoiding what Steve Jobs coined as a “bozo explosion.” They started to hire to “elevate,” not to “delegate.”

“We either needed to invest a lot in hiring and spend time and energy on that, or we were going to end up spending and investing a lot of time and energy in firing or in failing,” says Ryden. “Out of the three, I’ll pick the first one.”

Another important realization for Ryden during this phase was that not all great individual contributors can make great managers. He saw that play out first-hand with the stellar hire they brought on to build the development team. “All the things that made him great as a developer did not serve him well as a manager,” says Ryden.

During this inflection point phase between 2014 and 2017, PrecisionLender grew from 31 to 90 employees with revenues doubling every year. But Ryden was still viewing the company as he did when he helped run a few-million-dollar startup — he didn’t think it was capable of attracting the true top-tier talent. “Companies grow up like kids do,” he says. “But you’ll still want to see them as the little companies that they were. That was a good lesson, to hire leaders worthy of the team.”

Ryden’s company had grown up. And the kind of developer he needed to keep growing had also changed.

RETHINKING THE 10X DEVELOPER THEORY

At this point, you should stop reading Ryden’s story unless you’ve read Joel Spolsky’s blog post “Hitting The High Notes” (and if you’ve already read that, it’s worth another pass). Spolsky’s concept explains the variations between the very best programmers and the mediocre ones. His thesis is based on research from a computer science class at Yale: The best of the best developers are 10 times more productive than the average ones — they’re faster and have fewer bugs and make better software. Spolsky writes, “The real trouble with using a lot of mediocre programmers instead of a couple of good ones is that no matter how long they work, they never produce something as good as what the great programmers can produce ... the ‘average productive’ developer never hits the high notes that make great software.”

Ryden agrees with most aspects of that 10x developer theory. Especially early on in a company’s life, a 10x developer can quite literally pour the foundation for a successful business. But what Ryden needs now is a “10x + 10y” developer, a “10 more” developer. By that he means someone his team could work with 10 more of — developers with a “y dimension” of empathy and humility. “A lot of times you can only have one 10x developer,” says Ryden. Think of those like beta fish — those fish are beautiful, but they can’t swim in the same tank. “Two of them are corrosive and collide with one another. Some of them are what I call a 10x minus ‘y,’ meaning yes, he’s 10x, but not only would I never hire another one, I wish I didn’t have the one I have right now.” And he also knows a 10x developer can also make for a 0.1x manager.

To Ryden, the bread and butter on a development team are the 2 or 5x developers who make others around them better and who can scale with the company. Ryden learned the importance of empathy and humility before PrecisionLender had ever shipped a line of code. Back when it was just him, he hired a software architecture consultant, the kind that commands several hundred dollars per hour. He then sat down to explain what his customers — relationship managers at banks — face in their day-to-day jobs. The response he got was, “I don’t really need to know any of that. I can abstract all that away.” Ryden fired him after a week. Why? “Everything is about empathy for solving the problem. No one cares about the architecture except to the degree that it helps them solve the problem,” Ryden says. “If you don’t understand the problem you’re solving, you have no way of judging a good architecture except for what’s good and easy for a developer to maintain.”

The true unicorns are the developers who have both dimensions. They are faster, make fewer mistakes, and are more creative, and they are empathetic, humble team players. He says, “You can make the compromise early on and forego the “y” dimension when it’s just you writing software in a closet. But as you get bigger, you really need the “y” dimension. I think a lot of the Silicon Valley companies miss this.”

"With God as my witness, from this day forward, I will hire only adults."

HIRING THE 10X + 10Y DEVELOPER

All candidates answer three questions during PrecisionLender’s hiring process:

  1. Tell us your story — where have you been, and where are you going?
  2. Can you describe a time where you received “wow” customer service?
  3. What would your last two bosses say about you?

The first question obviously tests communication skills. The third question is borrowed from Topgrading (How To Hire, Coach and Keep A Players) by Brad and Geoff Smart. The second question trips up the most candidates because they’re wired to talk about themselves and give an example of a time they delivered “wow” customer service (not a time they received it). And even when they talk about delivering exceptional service, oftentimes it’s just an example of showing up and doing a job. Ryden says the company has regretted every single time it has made an exception for a talented candidate with a stellar resume who has failed the second question.

After the three-question screen, there is a phone screen, a casual “coffee screen,” and then a take-home assignment. Although there’s been pushback in the software world about not scaring away coveted developers with an overly time-consuming interview process, Ryden stands firm with this homework. “If you’re hiring a dancer, you have them dance. If you’re hiring a developer, you have them write software.” The company used to ask candidates to write a piece of software in the office, but a take-home assignment gives developers a chance to shine without having to feel uptight during the pressure of an in-person interview.

The hiring process sounds intense, and it’s designed to be that way. A “gauntlet” is what Ryden calls it, another term borrowed from Mikey Trafton’s “Business of Software” talk. Rock stars want to work with other rock stars. “The really, truly outstanding people who go through the interview process are not only judging the company by the computer monitors and the break room,” he says. “They’re looking at the process they go through and say, ‘This is the process that’s going to hire all the coworkers that I’ll have here.’” The “gauntlet” produces a very different type of hire from the scratch and dent bin type of hire from Precision- Lender’s early days.

If you think PrecisionLender’s process sounds too tough, or if you think Ryden’s wish for those 10x (or at least 2 to 5x) plus 10y developers is too idealistic, then just know this: It works. The company benchmarks itself against other high-growth tech companies and knows the average turnover among engineers is around 20 percent. PrecisionLender’s combined voluntary and involuntary turnover is a mere 3 percent.

For Ryden, creating and adapting an AI-backed software platform with enterprise-class security to serve commercial banks on two continents is actually the easy part. It’s the people part — especially the hiring part — that is the hardest to scale. “People don’t scale at all but are incredibly flexible and empathetic. Software scales like mad, but is not very flexible and not very empathetic at all,” Ryden admits. But PrecisionLender is scaling, and when Software Executivechecks in with Ryden three years from now, it’s a safe bet that its headcount will be growing faster than it was for the company’s first three years in business.