Prior to switching into Corporate Venture Capital (a position I left in May of 2025), I led the teams acquiring job seekers and employers for Indeed. I started way back in 2010, left, came back, and then did that role until 2022 (when I switched into corporate venture capital). During that time, our paid search spend went up 10x, easily. Other channels were found, but just buying ads on keywords went up 10x, into low/mid 7 figures.
When I started, the team would use Excel formulas to calculate and adjust the bids. Then we automated the Excel updates with Python scripts. Eventually we ended up building a whole machine learning and bidding team to automate the process. This was fairly advanced, at the time.
While we were doing this, Google (now Alphabet) and Facebook (now Meta), were building out automated systems to automatically target keywords and manage bids on behalf of advertisers.
(this is going somewhere relevant, I promise).
The rise of automated advertising
We experimented with these systems a number of times, but they weren’t better than what we had. Until about 2021. I had spoken with other leading advertisers that had switched over, but we had a significant amount of in-house expertise. By 2021, we were able to work with Google to help them through issues their system was having, and we were able to drive positive results. Eventually, we could just turn over all of the tactical campaign performance to Google’s Automation and Smart Bidding. Meta developed a similar system.
Hyper simplistically, Google and Meta combine the information they have about users, conversion signals (did someone buy/take a desired action), and machine learning to build these systems. These automated targeting and bidding systems became the clear way for online advertising companies to grow companies.
Reddit, Snap, AppLovin have all taken the roadmap from Alphabet and Meta on how to significantly increase revenue. And, it works for them. Eric Seufert of MobileDevMemo..com has documented these efforts, across numerous posts.
ZipRecruiter explores advertising automation
ZipRecruiter looks like it’s trying a similar approach. During its earnings call, it noted the performance of its automated campaign management. This system is driving an increase in performance revenue (as opposed to subscription-based revenue). ZipRecruiter has some similar elements.
ZipRecruiter has core components to build this out
It knows a decent amount about users from resumes and online job search activity. It has then worked to build out integrations with over 180 applicant tracking systems (ATS). Integrations with ATSes let ZipRecruiter get disposition data (data about whether a candidate was interviewed, rejected, hired, etc). With this information, the behavioral signals, and machine learning, ZipRecruiter could start working on a traditional automated, machine learning system like the other ad platforms.
Why jobs can be like eCommerce
One point of view is that this won’t work. This is not like eCommerce or Travel where multiple people can buy the same item (ecommerce) or seat on an airplane. Many companies are hiring one role, looking for one person. This is partially true. However, there are companies that are hiring thousands of people for highly similar roles (but spread out geographically) or have dozens of open positions around similar types of roles, but in different use cases.
ZipRecruiter serves blue collar and front line work (as evidenced by their Breakroom acquisition), not the type of highly specialized knowledge work roles that LunkedIn does.
Meta and Alphabet’s automation technology doesn’t work perfectly for every use case either. However, where it does work, it drives better performance for the advertisers and for Meta and Alphabet.
If ZipRecruiter is successful, clients should see improved cost per hire and faster speeds and ZipRecruiter will see increased budget share.
ZipRecruiter General Performance Q3 2025
Financial Performance:
• Q3’25 revenue: $115.0 million (2% sequential growth)
• Performance-based revenue grew 12% quarter-over-quarter - the largest sequential growth since Q1’22
• Performance-based revenue now represents 24% of total revenue (up from 22% in both Q3’24 and Q2’25)
Automated Campaign Adoption:
• Enterprise customer adoption of automated campaign optimization increased 19% quarter-over-quarter in Q3’25
• ZipRecruiter has integrated with over 180 applicant tracking systems (ATS)
• These integrations provide the critical disposition data needed to train their machine learning models