Entrepreneur Bill Harding launched Static Object to build a product that he personally needed: a more effective way of tracking his programmers’ progress without relying on standup meetings or interrupting them for updates, jarring their coding flow.
Static Object, a Seattle-based company with seven employees, has a tool for managing engineers by analyzing the code that they write and host on GitHub, the most popular platform for that purpose. Static Object reviews and summarizes the code, teasing out the fraction of the work that has a meaningful impact on the code base. It can also highlight roadblocks and help reveal how good, or poor, a new programming hire is.
“It’s a higher level of transparency than is available in any other way,” said Harding, and it provides an efficiency boost that he greatly needs. Harding splits his time equally between two CEO roles: leading Static Object and Bonanza, a profitable online marketplace that launched in 2008 and provides an alternative to Amazon, Etsy and eBay.
Harding began developing the idea for Static Object in 2016. It had a soft launch at the end of 2017 and came out of beta earlier this year. About 100 companies are using the product, though not all are paying customers. A monthly subscription to the service is $295-395 for companies with fewer than 10 people.
Static Object’s main competitor is GitPrime, Harding said, which also generates engineering metrics, but provides less specific information about the actual coding.
Static Object has an obvious appeal for managers, but how do developers respond to the idea of Big Brother keeping close tabs on their coding?
Given that Harding uses Static Object at Bonanza, “we’re still figuring that out ourselves,” he said with a laugh. Many developers have been receptive, but “there is a contingent that is wary about being measured and what they might perceive as managers meddling. But the flip side is they get to see what their team members are working on, and what they’re working on, in a quick format.”
The tool is also good for helping developers recall what they worked on nine months ago, and can identify what sort of coding — front-end, back-end or documenting — they’re best at.
Based on the analysis by Harding’s team, he asserts that 95 percent of coding has little impact. That includes moving files or functions to new places, adding blank lines and spaces that make the code easier to read, repeating lines of code by copying and pasting, and code that’s “churned” or written then replaced.
It’s not that the work is unnecessary, he said, or that you would even want to shift the percentages.
“But it doesn’t amount to moving the ball forward for the company,” he said. “I would have figured 50 percent of code is meaninglessly written. It’s way bigger than I would have guessed.”
There are sure to be engineers and managers who will push back against these percentages and characterization of the code’s value. And the analysis doesn’t score the code on whether it’s scalable or efficient. Advocates of standup meetings and the benefits of face-to-face interactions will bristle at the idea that these analyses can replace those conversations.
But Harding estimates that the time and money savings are significant, particularly for larger teams. If you can eliminate three-times-a-week, half-hour standup meetings with 10 developers who earn on average $150,000 a year, that saves $4,500 a month.
“This is the tool that has become all of the things that we need,” Harding said, “from a development management standpoint.”
We caught up with Harding for this Startup Spotlight, a regular GeekWire feature. Continue reading for his answers to our questionnaire.
Explain what you do so our parents can understand it: We provide a way for managers to understand what their programmers are working on. Companies usually find ways to get more meaningful output from their expensive programming team when they can understand what those programmers are working on.
Inspiration hit us when: Our marketplace business, Bonanza, grew to 10 developers, and it became hopeless to understand what they were working on in a given day — let alone over a week, month or year. And yet management is expected to write an annual review that summarizes the engineer’s output, plus choose a new salary that reflects the value of the work done over the past 12 months, relative to their peers. I’m lucky if I can remember what was going on three months ago.
VC, Angel or Bootstrap: Bonzstrapped! We created the popular online marketplace Bonanza around 10 years ago, and are allowing its success to fund the creation of this new product. That leaves us in a financial situation that closely approximates “bootstrapped.” We’re making resources for Static Object because we desperately need it to exist for the sake of Bonanza.
Our ‘secret sauce’ is: We can identify the 95 percent of engineering work that’s irrelevant, and pluck out the 5 percent that is a good signal for when a developer is making an impact on their project.
The smartest move we’ve made so far: Having conviction about emphasizing a single metric (line impact), and then allowing that to get sliced and diced in a bunch of interesting ways. Having a single good metric that we can trust tells us that our work-from-home day doesn’t reduce productivity, while our Freedom February benefit (which sends employees and their families to a tropical locale to work) does have a significant negative impact on our productivity — but we’re going to keep doing it anyway.
The biggest mistake we’ve made so far: Thinking it’d be easy to launch a product in a category (“code metrics”) that hasn’t previously existed. It has been the opposite of easy to help engineering managers understand the benefits of improved engineering transparency.
Which leading entrepreneur or executive would you most want working in your corner? Ashton Kutcher invested in Airbnb, right? If we could get a testimonial from him to stick on our homepage, and we could add a tagline like “Ashton Kutcher, Airbnb/Static Object Investor” that would go a long way in driving the increased page views we crave. Visitors would have no idea if or when if they might come across Ashton again after signup, which also boosts time on site.
Our favorite team-building activity is: Quiet conversations, thoughtful analyses.
The biggest thing we look for when hiring is: Serendipity! All of the studies that I’ve read, such as this one, indicate that the best interview format from a “prediction power” standpoint is the structured interview. This “best method” has predictive power on par with a coin flip. It’s hard-to-impossible to predict who will thrive based on a few hours of conversation and a LinkedIn profile.
The struggle to quickly recognize impactful developers is partly why Static Object exists. We have a cohort chart specifically dedicated to helping managers compare the strengths of a new hire to past developers at the same point in their career.
What’s the one piece of advice you’d give to other entrepreneurs just starting out: The less money you can raise, the more freedom you’ll have to adapt your business without drawn-out explanations to a board of directors. So many fundraising stories make headlines that it’s easy to forget that most businesses are quietly growing and improving under your nose. A benefit like Freedom February would never survive a Series B round of fundraising. Too many lawyers, too much pull toward convention.