Josh Millet, Criteria: Bootstrapped to Nearly $5M Before Funding

1Mby1M Research · 1Mby1M Case Study

Interviewer: Sramana Mitra

This interview is part of the case study-based research and entrepreneurship education work of 1Mby1M, the global virtual accelerator founded by Sramana Mitra.

Abstract

Josh Millet co-founded Criteria, a SaaS-based employment assessment company, in 2006 after earning a Ph.D. in medieval French history from Harvard and working in hiring at a Los Angeles software company. With co-founder David Sherman, Millet raised only a small amount of angel funding before launching the product in 2007, just months before the global financial crisis forced the company into nine years of necessity-driven bootstrapping.

Focused initially on small businesses through Google Paid Ad-driven customer acquisition, Criteria grew to roughly $4-$5 million in revenue, 16 employees, and about 2,000 customers before raising its first institutional round from Level Equity in 2015, followed by a larger round from Sunstone Equity Partners in 2019 that helped the company shift upmarket into enterprise accounts. This case illustrates how nine years of disciplined, cash-flow-focused bootstrapping through a difficult launch window can build a durable customer base and product-market fit before outside capital is introduced, reinforcing the 1Mby1M principle that validation should precede fundraising.

Transcript of the Interview

Sramana Mitra: Let’s start at the beginning of your journey. Where are you from? Where were you born and raised? What kind of background did you have?

Josh Millet: I was born in Toronto, Canada. I have dual citizenship of US and Canada. I came to the States for school 20 years ago and have been here ever since.

I had an indirect route to software entrepreneurship. I did a graduate degree in medieval French history. I was intent on being a professor at that point. At the end of grad school in 2000, I started a test preparation website with a friend of mine. This was when the first dotcom boom was crashing. We started that venture and sold it a year and a half later. That was when I was bit by the entrepreneurial bug.

Sramana Mitra: Where was this grad school?

Josh Millet: This was at Harvard.

Sramana Mitra: Did you finish your Ph.D.?

Josh Millet:Yes. I finished it in seven years. I spent the better part of my twenties doing that.

Sramana Mitra: This was probably the same time that I was in MIT.

Josh Millet: Yes, maybe there was some overlap there. The test preparation company that I started was acquired by a software company out of Los Angeles when we were only 18 months in. The dotcom bubble had just burst back then, so it seemed like the right move for us to do this buyout. That’s how I came to live in Los Angeles, and I have lived here ever since.

Sramana Mitra: Did you work for this company that acquired you?

Josh Millet: That’s right. There were five of us in that startup. I was the one who volunteered to come out and work for the acquirer. I ran our startup as a division of this bigger company based in Culver City. It was a great experience. I learned a lot from Allen, the CEO of the acquirer. He became my mentor.

That leads to the origin story of my current company. As part of my role there, I was put in charge of hiring. I had no experience at that time. That led me to start my company Criteria three years later. This is a SaaS-based software company that specializes in employment assessment. We help companies and organizations make decisions by using evidence-based tools and assessments.

Sramana Mitra: What year was that?

Josh Millet: I moved to California in 2002 and worked for that other company for three years. We founded Criteria in 2006.

Sramana Mitra: Let’s double-click down on the starting of Criteria. What did you start with? How was the journey? Who was involved? What was the bootstrapping process?

Josh Millet: When I was working at that other company, I started participating in interviews with applicants. We weren’t good at hiring sometimes. A lot of companies struggle with that.

There was this one interview that it was clear to me that it was not a fit on either side. It made me start thinking about ways in which companies could use data and assessments to prevent some of those terrible interviews. I didn’t have any funding.

My co-founder, David Sherman, and I wrote a business plan and started shopping for funding. We decided to bootstrap although we did raise a small amount of angel funding to get going.

The goal for us was to take a field that has been around for a long time like pre-employment assessment and make it more technology-driven. We wanted to make it accessible to companies of all sizes.

We launched the product in 2007. That time ended up being uniquely terrible for launching a hiring software because, within six months, the global financial crisis hit. We were forced to bootstrap out of necessity for the first years.

It took a while to find a product-market fit. We have a great CTO in Wayne Twine who built our product in those early years. We ended up bootstrapping the business for nine years before we took any other outside money.

Sramana Mitra: How did you bootstrap your company? Was it customers buying your products or did you have to do other creative things?

Josh Millet: We got the customer to pay for it. When you are bootstrapping, you have to manage for cash flow and not for growth. There were four of us when we started the company in a tiny office. It was three years before we hired our fifth employee.

We were conservative with the small amount of money that we raised. Eventually, we broke even and that is when we hired another employee. We were always managing cash flow in those early years.

Sramana Mitra: What was the revenue level you reached with bootstrapping before you switched gears?

Josh Millet: We first raised venture capital in 2015. At that point, we had about 16 employees and between $4 million and $5 million in revenue.

Sramana Mitra: How many customers did you have at this level of revenue?

Josh Millet: It would have been about 2,000 customers. We specialized in small businesses. Since then, we have moved upmarket. We now serve customers of all sizes. We have a lot of enterprise customers now as well but, at that point, mostly small businesses were using the product. The SMB market is hard work, but the customers are great to have.

Sramana Mitra: You went after the startups because they were hiring a lot. Intuitively, it would strike me that the large enterprise market would be more conducive to hiring evaluation solutions.

Josh Millet: Yes, we probably should have had you as an advisor back then. We would’ve figured that out quickly. When we looked at the market initially we thought that the enterprise market was pretty well-served.

Often with the assessment, they are sold as part of a consulting package. It’s a lot of effort to implement. We saw an opportunity to switch that process to a product approach and make it accessible to smaller businesses. Our approach was, “Hey, let’s take these services that Fortune 500 companies have been using for many years and make those types of products accessible to small businesses.”

Like so many companies, we got pretty good at that. Then after several years, we realized that enterprise customers are going to be great to have. The companies that can do this at scale, from a SaaS perspective, are valuable customers to have because they don’t churn at all. You can grow those accounts and those relationships over time.

About three or four years ago, we started concentrating on that enterprise market and that is where a lot of our growth has come from.

Sramana Mitra: Who did you raise your first round of funding from?

Josh Millet: They were a growth equity firm out of New York called Level Equity. That was in 2015. Ironically, we didn’t feel like we needed funding at that point. We were getting a few inbound calls at that point.

There had been some M&A and IPO activity in our space of human capital management; so there was investor focus on our space. We thought that we had been committed to bootstrapping at that point. We were getting some inbound calls.

One of the interests was from one of our customers. We eventually decided that this would help us invest in the growth of the business and get help in areas that we are objectively under-resourced at that point. We had about 16 employees at this point, so it was small.

We decided to take outside capital. It was a great partnership with Level. They helped us build the business in some key areas. We partnered with them for four years and then did a bigger round with Samara Equity Partners in 2019. It’s been a great relationship with them. They helped us expand much further and move aggressively in the enterprise market.

Sramana Mitra: Your business model is Software-as-a-Service?

Josh Millet: That’s right. We sell our service on an annual basis. The customer pays us a flat annual fee and they access the service through the web.

Sramana Mitra: What is the average deal size?

Josh Millet: That is one of the things that has radically changed in the last four years as our customer mix has shifted. When we did that first deal with Level, our average deal size was about $2,000 per year.

It’s now north of $5,000 a year because our mix of customers is changing. We are getting more midsize to enterprise-level customers who pay more for the service. In three or four years, we expect that to grow to an average deal size of $8,000 to $10,000.

Sramana Mitra: What is the customer acquisition strategy?

Josh Millet: When we were bootstrapping, we built our business on the back of Google. A lot of customers came in through the web. We did spend quite a bit of money on Google with paid ads. Over time, we built a great content strategy.

What started as paid search has now turned into an organic search. We started telling our story on the web with a blog, which was a hit. It has been web-driven and search-oriented. We get a lot of small businesses because they saw us on the web. We have a free trial of our services. They get to try it out and look at all these different assessments.

As we move up to more enterprises, it’s more difficult as we do a lot of outbound prospecting. In an enterprise, we have different ways of reaching customers as well. It’s still, to a large extent, web-driven.

Sramana Mitra: How do you analyze the competitive landscape today?

Josh Millet: The field of pre-employment assessments is not a new category. It’s been around for 50 years. When we first came to the market, we were competing with a lot of companies that were good at publishing assessments and validating them, but they weren’t great at customer acquisition on the web or even technology in general. They were more like content publishers.

That was the one area that we had an advantage in. All of the three founders of my company had a tech background. Initially, we were competing against these older companies that were new to the web. In the last three or four years as we have moved to enterprise, we still compete with great assessment businesses.

Now, we also compete with AI-driven companies that use some little assessment science along with general machine learning and AI-driven approaches to create selection algorithms with other inputs. When we are in a bake-off with other companies for big enterprise accounts, it’s going to be us, one other assessment company, and AI-driven companies.

The AI companies are more focused on finding correlations wherever they exist. We have a definite perspective on that. We do some machine learning ourselves and we have early growth. Part of our mission and what we believe is that assessments help you reduce the level of unconscious bias in a company’s hiring process.

Some of the high-profile AI attempts in the hiring space have had the opposite effect. You might have heard the story about Amazon abandoning algorithms that were supposed to promote gender equity. It had the opposite effect.

There are some problems with those approaches and, certainly, some of those are interesting. We focus on using selection algorithms where the applicants are very much aware of how they are being measured. We get nervous anytime there’s AI that is black box and opaque to the user.

Our approach is to use assessment science and combine social science with cutting-edge data science. This a good approach to take rather than just leaning on data science alone. It’s interesting because the set of competitors over the years has dramatically changed. Now, we are competing with a lot more AI companies.

Sramana Mitra: Here’s a synthesizing question. What can your technology do that an AI-based approach cannot do?

Josh Millet: In an objective and transparent way, our technology measures key competencies, abilities, and behavior traits that we know are predictive of job success and business outcomes.

Decades of research in social science shows that cognitive ability is a key predictor of success in the workplace, so some of our assessments measure cognitive abilities. In certain roles, we know that certain personality traits are linked to success. There is a set of characteristics, for example, in sales roles that tend to predict success.

Recently, we acquired an Australian company that is a leading game-based assessments provider. One of their assessments is focused on emotional intelligence which is a new area of research. The field of emotional intelligence is interesting and fast growing in popularity these days. In certain managerial positions, for example, emotional intelligence is important to have.

In a lot of respects, because of COVID, it’s become even more important to have AI in your organizations. We are leaning toward assessment science – both the older and some of the newer ones. That is how we would be significantly differentiated to AI-driven approaches that only look at the different data they gather.

Sramana Mitra: Are you sure that what you are doing cannot be modeled in AI?

Josh Millet: Certainly some of it can. There is interesting stuff going on there. I’m going to give you one example. Our assessments measure some personality traits that are relevant to job performance. There have been some AI-driven attempts to glean the same personality characteristics from all sorts of places.

For example, one company has claimed that they can model a person’s personality profile successfully from social media profiles. That is an interesting thing where they use semantics to construct a big five personality profile of an applicant without them having to take an assessment.

They are claiming to get the same data from different sources. We have our perspective on whether that should be done or whether it is ethical or not.

Sramana Mitra: Is there anything else that I should have asked you so far?

Josh Millet: This has been thorough. I know you are a huge proponent of bootstrapping yourself. It’s a thrill for me to talk to you. It was an interesting transition for me going from bootstrapped for so long to outside funding.

It is a reorientation that you have to do. Some of your priorities change a bit. Some habits also need to be unlearned. I think it’s important to have a bootstrap mentality even after securing funding.

Sramana Mitra: Thank you for your time.

Key Takeaways

• Josh Millet bootstrapped Criteria for nine years out of necessity, launching the product in 2007 just before the global financial crisis hit, and raised only a small angel round before institutional funding.
• The company grew by strictly managing cash flow over growth, taking three years to hire a fifth employee, and reached roughly $4-5 million in revenue with about 2,000 small-business customers before raising a first institutional round from Level Equity in 2015.
• Early customer acquisition was built on Google paid search that evolved into an organic content and blog strategy, targeting small businesses with a free-trial model before the company later shifted focus toward enterprise accounts.
• After bootstrapping, Criteria raised growth equity from Level Equity in 2015 and a larger round from Sunstone Equity Partners in 2019, using the capital to expand upmarket, grow average deal size, and compete against both traditional assessment publishers and newer AI-driven hiring tools.