Read Your Competitor's Job Postings to See Their Next Move Before They Make It

A competitor's job postings preview their strategy. A company has to hire the people to build something months before it ships, so the roles it opens, the skills it lists, and the cities it hires in map to product bets, technology shifts, and geographic expansion. You can read all of it for free.

These signals appear weeks or months before any press release. Hedge funds and institutional investors already buy hiring data as a dataset worth millions. A small business owner can read the same signal by hand, at no cost, from a competitor's public careers page.

Why is a job posting more honest than a press release?

A press release tells you what a company wants you to believe. A job posting tells you what it is paying to build. The ad has to be accurate, because an inaccurate one attracts the wrong candidates and wastes the recruiter's time. That accuracy is what makes it useful to an outsider.

The academic evidence supports this. The peer-reviewed study "Are Online Job Postings Informative to Investors?", published in Management Science in 2020, found that changes in the number of job postings are positively associated with changes in future performance, and that the relation is stronger when postings likely represent growth rather than replacement. The analysis used data scraped from more than 14,000 public and private companies. Companies that later went bankrupt showed a reduction in online job postings just before filing.

What signals should you actually read?

There are five reads inside a posting. First, required skills point at technology direction. A job description has to name the real stack to attract the right person, so when a competitor starts requiring vector databases, RAG (retrieval-augmented generation), or LLM fine-tuning, they are building a capability that will surface in the product later. Named tools recur in real AI postings, including Pinecone, Milvus, and pgvector.

Second, seniority tells you how big the bet is. A director-level hire signals a funded initiative rather than an experiment. Third, growth versus replacement changes the meaning. A net-new role signals expansion. Backfilling an existing title signals retention. Fourth, location reveals geographic strategy. A new office shows up as a cluster of local roles before any announcement, so a domestic competitor suddenly posting in London, Singapore, or Sao Paulo is signaling market entry. Fifth, and most important, the volume trend over time is the intelligence. Any single posting is noise. Three senior AI roles in January, five in February, eight in March is an acceleration curve pointing at a real plan.

How far ahead does hiring run before the product ships?

Hiring is a forward-dated signal because the seat takes time to fill and the work takes longer to finish. SHRM benchmarks the average U.S. time-to-fill at roughly 44 days, and technical and senior roles run longer. Filling the seat is one phase. Sourcing and onboarding add time before the person is productive, and building the product follows all of that. A posting describes work that is still months from shipping, which is what makes it an early warning.

Is reading a competitor's careers page legal?

Reading publicly posted listings yourself is on solid legal ground. On April 18, 2022, the U.S. Ninth Circuit in hiQ Labs v. LinkedIn reaffirmed that accessing publicly available website data does not violate the Computer Fraud and Abuse Act's "without authorization" clause, because a public site has no access permissions to prevent access.

One honest caveat. The same litigation established that a site's terms of service can still be enforced through breach of contract, and hiQ ultimately settled with a $500,000 judgment. Automated scraping at scale can run into those terms. Reading a public careers page by hand is a different activity, and that distinction is what keeps a small business safe. This is general guidance, so confirm how it applies to your own situation before you rely on it.

What did this look like in real companies in 2025?

The direction of hiring tracked the direction of the business across 2025 alt-data reporting. Booz Allen Hamilton openings fell from more than 1,600 in March 2025 to around 700 before earnings, a 66% year-over-year decline, and its shares fell 16.5% in May 2025. Moderna's postings fell more than half over a year, a 0.86 correlation with its stock, ahead of a layoff of about 10% of staff. CoreWeave nearly doubled postings in the year before its March 2025 IPO. Klarna cut postings from more than 150 in 2023 to 20 to 30 in 2025 as its valuation fell from nearly $50 billion toward $13 to $14 billion. In each case, the hiring curve moved before the headline did.

This is why institutional money pays for the signal. Per a Neudata report cited in 2025 coverage, over 90% of institutional data buyers expect to maintain or increase spending on alternative datasets, with many allocating more than $5 million a year. Dedicated providers exist to sell it. Revelio Labs' COSMOS dataset holds more than 5 billion postings from over 7 million companies.

How does a small business do this for free?

Pick three to five direct competitors. Bookmark their careers pages and their LinkedIn Jobs listings. Once a month, open each one and log the role titles and the count in a simple spreadsheet. That is the whole system. After two or three months you have a slope, and the slope is where the intelligence lives: a cluster of net-new roles around one capability, a jump in senior titles, a new city appearing in the location field. Hedge funds pay millions for a machine-scaled version of this. You are reading the same public pages by hand.

Common questions

How can I tell if a competitor is about to launch a new product?

Watch for a cluster of net-new roles tied to one capability, such as several AI or mobile engineers plus a product manager for the same area, appearing over two to three months. Because roles fill in roughly 44 days on average and building follows, that cluster typically precedes a launch by a few months to over a year.

Can a competitor's job ad really tell me their technology choices?

Yes. The required-skills section names the actual stack, because the ad has to match the work. A new requirement for a tool like Pinecone, Milvus, or LLM fine-tuning shows a capability they are building now that will appear in the product later.

Is it legal to monitor a competitor's job postings?

Reading publicly posted listings yourself is on solid legal ground. The Ninth Circuit held in hiQ v. LinkedIn (2022) that accessing public web data does not violate the Computer Fraud and Abuse Act. Automated scraping at scale can still run into a site's terms of service, so for a small business, manual reading of public pages is the safe path. Treat this as general guidance and confirm it for your own case.

The principle underneath all of this is simple. A company votes for its future with its payroll, and that vote is public. You do not need a dataset or a subscription to read it. You need a spreadsheet, five competitors, and the patience to check once a month and watch the slope.

• Management Science: Are Online Job Postings Informative to Investors?, Peer-reviewed study finding job posting changes predict future corporate performance, stronger for growth than replacement.
• UC Irvine Paul Merage School of Business, Study summary noting data from more than 14,000 companies and reduced postings before bankruptcy filings.
• Morgan Lewis on hiQ v. LinkedIn, Law firm analysis of the Ninth Circuit ruling on accessing public web data under the CFAA.
• hiQ Labs v. LinkedIn case record, Case history including the settlement and $500,000 judgment on the breach-of-contract question.
• Job Board Doctor: When Job Postings Predict Market Moves, 2025 company examples (Booz Allen, Moderna, CoreWeave, Klarna) and the Neudata alt-data spending figure.
• Revelio Labs COSMOS launch, Dataset scale of more than 5 billion postings from over 7 million companies and alt-data industry context.
• PageCrawl: Competitor Job Posting Monitoring, Framing of pattern-over-time and geographic hiring signals from posting clusters.
• SHRM Talent Acquisition benchmarking, Source of the roughly 44-day average time-to-fill benchmark, reported at HRchitect.
• Firecrawl: Best Vector Databases, Reference for named vector-database tools including Pinecone, Milvus, and pgvector.