Indeed sponsored jobs vs free. How to measure what sponsorship actually adds

KEY TAKEAWAYS

1. Indeed's uplift figures run from 60% to 6.5x depending on the page, and none publishes a method.
2. The free baseline those figures compare against has shrunk, so the ratio can grow without sponsorship getting better.
3. The billed event depends on market, budget type and contract. Read the invoice, not a pricing guide.
4. Judge sponsorship on cost per extra qualified applicant from a paired-location test, with a range, not on cost per application.

Seasonal hiring on Indeed started early this year. On 15 October Indeed's Hiring Lab said seasonal postings were about 10% above the same point in 2025, and that they usually keep building until mid-November. Meanwhile, the last of Indeed's feed cutoffs for agencies lands on 31 December. Recruit, Indeed's parent, told investors in August that US revenue per job posting rose 35% in a year. Many hiring teams are about to put more money into sponsored jobs. Few can say what that money adds beyond what a free post would have brought in.

Why do Indeed's own uplift figures disagree?

Indeed publishes several numbers for sponsored jobs against free ones. They do not agree.

Only one line gives a period. None gives a method, a sample or a control group. The spread runs from 1.6x to 6.5x, which is too wide to plan a budget on.

There is a second problem. These are comparisons between jobs that were sponsored and jobs that were not. Employers choose which jobs to sponsor, and those jobs may differ in pay, urgency or title. A randomised study of a freelance marketplace shows how much that choice alone can move a number. In its placebo group, where paying did nothing at all, the applications that workers chose to boost were still about 101% more likely to end in a hire. That was pure selection. Indeed's figures may contain some of the same.

Is the free baseline still the same free baseline?

No. The thing the multipliers compare against has been cut back three times.

• December 2025. Job Board Doctor reported that free hosted jobs dropped from 120 days of visibility to 30, with a cap of three a month. Indeed's help page now says linked accounts under one parent share that single cap. Closing a free job early still uses its slot. Reopening one needs sponsorship.
• 31 March 2026. Jobs sent through single-source XML feeds became visible only when sponsored, where an integrated ATS could have sent them instead.
• 30 June and 31 December 2026. Agency feeds lose visibility in two further steps. After December, agencies manage integrated ATS jobs only through Indeed's Job Update API.

Some free reach survives. Jobs from an integrated ATS do not count toward the three-post cap. Indeed told HR Dive it will keep scraping public career sites, and that employers who do not migrate will still get organic traffic through them.

The point is arithmetic. If free reach shrinks and sponsored reach stays the same, the ratio between them grows. A "3X" today and a "3X" two years ago may describe different products. Search results are also personalised. ERE reported a franchisor that sponsored 50 Panera roles and still could not find them for exact searches, and Indeed said sponsorship does not guarantee visibility for any one search. Searching for your own job proves nothing either way.

Is sponsorship getting more expensive?

The direction is clear. The size is not.

On its 7 August call, Recruit reported US HR Technology revenue up 30.0% to US$1.64 billion, with US job postings down about 4%. Revenue per job posting rose 35%, and Recruit expects about 30% for the year. It named Premium Sponsored Jobs as the main driver of US growth.

Many guides read that 35% as a price rise. Recruit says otherwise: the measure is "not the average unit price per sponsored job ad." For small employers, management said growth came roughly one third each from more paying clients, more sponsored jobs per client and higher unit prices. So prices are rising, but by much less than 35%.

Appcast's 2026 benchmark points the same way from the buyer's side. Across about 1,200 US employers and 27.4 million applications in 2025, the median cost per application on job boards was $19.32. The report says cost per application is increasingly shaped by how job boards price and sell media.

What exactly are you paying for?

This is where pricing guides go wrong most often. Pin and other vendor guides describe sponsored jobs as pay per click. That is half right. SHRM reported that Indeed ended pay per completed application on 18 December 2023, but kept pay per started application.

Indeed's own pages do not agree with each other in 2026:

• The US pricing page says a daily budget charges when someone clicks to view, and a monthly budget charges when someone clicks to apply. It lists minimums of $5 a day or $150 a month.
• A US help article says Indeed does not offer monthly budgets. It says a daily budget is an average, and weekly spend can reach 7.5 times the daily figure on an open-ended job.
• The pay per started application help page says qualifying jobs are enrolled automatically and cannot switch back to pay per click.

A started application can be a click on the apply button, a QR code scan or a chatbot opening. It is not a finished application. Appcast measured completed applications at 5.19% of clicks across job boards in 2025. Your billed event and your useful event may sit far apart. Before you test anything, read one invoice line and write down which event it charges for.

Where is your free baseline?

Your posting route decides how much free reach you have, so it decides what sponsorship should be compared against.

The last row matters. CareerPlug's help notes say the "Indeed Sponsored" and "Indeed" source labels only work with Indeed Apply. Without it, all Indeed traffic is labelled the same.

How do you measure what sponsorship adds?

The cleanest test available to most employers is a paired-location test. It is a small geo test, the same logic as the designs in our piece on choosing a geo test by the question it answers.

1. Pick one role open in many locations. Use indexed jobs from your ATS, because a parent company's three free direct posts a month cannot fill a free arm.
2. Pair locations by past volume. Match locations on applicants per opening over the last 90 days. Within each pair, pick the sponsored one at random.
3. Start both arms on the same day. Appcast found over half of applications arrive in a job's first 10 days, so a free job posted a week earlier is not a fair comparison.
4. Keep everything else equal. Same title, pay, description and screener questions. Test Standard against Premium as a separate arm, not at the same time as the free comparison.
5. Define a qualified applicant first. For example, passed the screener and was offered an interview. Write it down before the test starts.
6. Count applicants from every source per location. This catches applicants who would have come through your career site anyway. The Indeed source split is a diagnostic, not the answer.
7. Compute a range, not a point. Divide spend by the extra qualified applicants, and report the low and high ends.

How many applicants make a result believable is the same question as in any split test. Our note on what a 65% confidence winner is worth walks through it.

A worked example

These numbers are hypothetical. A retailer hires the same hourly role in 20 locations. It runs the test for four weeks, 10 sponsored and 10 free.

• Sponsored locations: $3,000 spend, 420 applications, 126 qualified.
• Free locations: 190 applications, 66 qualified.

The naive reading says sponsorship produced 2.2 times the applications at $23.81 per qualified applicant. The incremental reading says sponsorship added 60 qualified applicants, at $50 each. That is about twice the naive cost.

Then add the noise. Treating the counts as Poisson, the 60 has a rough 95% range of 33 to 87. The cost per extra qualified applicant sits between about $34 and $91. Differences between locations make the real range wider. If your next best channel delivers a qualified applicant for $40, this test has not settled the question. Run it again before you move budget.

When is the test not worth running?

Some employers can skip it.

If you have one or two openings at a time, Noon AI's guide argues the free allowance often covers common local roles. A test needs volume you do not have.

If your jobs have no free reach, there is no baseline to beat. Compare Indeed's cost per qualified applicant with your other channels instead.

If you already get plenty of applicants, more may not help. In a randomised study of over 45,000 job openings on an online labour market, capping applications at 50 or five days cut applications by about 11%. The share of jobs filled did not change. Faster screening may be worth more than more reach.

The line that matters

Indeed's multipliers describe its marketplace, measured against a free tier it controls. Your number comes from your own locations, your own definition of qualified and your own invoice. Job search is also moving into AI assistants. Indeed launched an app in ChatGPT in February. How your employer brand shows up there is part of the same visibility picture. If you want that tracked, the waitlist is open.

Questions

Do sponsored jobs on Indeed get more applications than free jobs?

Indeed says so, but its own figures vary widely. Its UK page says about 60% more applicants worldwide, its German page 80% more applications, its US page 3X for hosted jobs and its Australian page 6.5x at the recommended budget. None states a method, and the free baseline behind them has been cut since December 2025.

How does Indeed charge for sponsored jobs in 2026?

It depends on market, budget type and contract. Indeed's US pricing page says a daily budget charges for clicks to view and a monthly budget for clicks to apply. Its pay per started application help page says qualifying jobs are enrolled automatically and cannot switch back. Pay per completed application ended on 18 December 2023. Check your invoice.

Do free Indeed job posts still get seen?

Some do. In the US and 11 other countries a parent company gets three free direct posts a month, each visible for up to 30 days. Jobs indexed from a career site or an integrated ATS do not count toward that limit. Single-source feeds became sponsored only on 31 March 2026, and affected agency feeds lose visibility on 31 December 2026.

How do I know if Indeed sponsorship is worth the money?

Run a paired-location test. Sponsor the same role in randomly chosen locations, leave matched locations free, start both on the same day and count qualified applicants from every source. Divide the spend by the extra qualified applicants. Compare that cost with your other channels and with the cost of a role staying open.

Sources

1. Indeed, Sponsored Job plans (US) indeed.com
2. Indeed, Free vs Sponsored Jobs on Indeed (UK) uk.indeed.com
3. Indeed, Free vs Sponsored Jobs on Indeed (Australia) indeed.com
4. Indeed Germany, Sponsored Jobs de.indeed.com
5. Indeed Help, Free job post limit indeed.com
6. Indeed Help, Single-Source Feed Policy (Effective March 31, 2026) indeed.com
7. Indeed Help, Pay per started application (PPSA) overview ca.indeed.com
8. Indeed Help, How daily budgets work for Sponsored Jobs indeed.com
9. Indeed Hiring Lab, Seasonal Hiring Is Off to a Fast Start, With an Eye Toward 2027 hiringlab.indeed.com
10. Indeed, Indeed launches app in ChatGPT indeed.com
11. Recruit Holdings, Transcript of Earnings Results Call for Q1 FY2026 recruit-holdings.com
12. Recruit Holdings, Earnings Summary for Q1 FY2026 recruit-holdings.com
13. Appcast, 2026 Recruitment Marketing Benchmark Report info.appcast.io
14. HR Dive, Cost-per-hire and cost-per-application rose sharply in 2025, report says hrdive.com
15. HR Dive, Visibility ends for certain free single-source XML feeds on Indeed hrdive.com
16. SHRM, Indeed Pay-Per-Application Pricing Ends Dec. 18 shrm.org
17. Job Board Doctor, Indeed Hosted Jobs, No More Free Lunch jobboarddoctor.com
18. ERE Weekly, Indeed's Search Experience Is Creating an Opening newsletter.ere.net
19. CareerPlug Support, Identifying clicks and applicants from your Indeed sponsored jobs support.careerplug.com
20. Pin, Indeed Pricing 2026 pin.com
21. Noon AI, How to Post a Job on Indeed (2026) noon.ai
22. Horton, Job-Seekers Send Too Many Applications, Experimental Evidence and a Partial Solution john-joseph-horton.com
23. Filippas and co-authors, Preference Signaling in Labor Markets, Evidence from a Field Experiment apostolos-filippas.com

NEXT STEP

See where your brand stands.

Your Digital Authority Score, who AI recommends for your target questions, and the placements that close each gap. Your first report is free.

Get your free report →