What moves gold

Data Revisions and Gold: Why Revised Jobs Numbers Move XAUUSD, and the Payroll Benchmark Explained

Most traders watch the headline number on jobs day: how many jobs did the US add? But every report also changes the numbers for the two months before, and once a year the whole series is reset by the benchmark revision. Sometimes the revisions matter more than the headline. A strong headline with big downward revisions can turn gold's first move around within minutes. Here is what revisions are, real examples from the Bureau of Labor Statistics, and how gold traders read them.

Gold spot$3,528.2−99.8 (−2.75%)NextUS Q3 GDP (second estimate) and October PCE inflationin 4d 20h · Wed 13:30 GMTMarket opencloses in 4h 34mRange$123.2
Gold price, last 12 monthsGold price, last 12 monthsHigh $5,318 (29 Jan 2026) · low $3,536 (20 Nov 2026) · 12-month change −14.2%$4,000$4,500$5,000$3,536Dec ’25Feb ’26Apr ’26Jun ’26Aug ’26Oct ’26
COMEX gold, daily closes, last 12 months · data to 20 Nov 2026 · xauusdglobal.com

The short answer

Every US jobs report does two things: it gives a first estimate for the latest month, and it revises the two months before. Once a year, the benchmark revision also lines up the whole payroll series with near-complete tax records.

Revisions matter for gold because gold reacts to what jobs data mean for US interest rates. A strong headline with big downward revisions can be weaker than it looks, which can flip gold’s first move. So on jobs day, read the headline, the revisions, wages and the unemployment rate together.

What a revision is

The monthly jobs report (often called NFP, for nonfarm payrolls) comes from a survey of employers run by the US Bureau of Labor Statistics (BLS). When the report comes out, not every employer has replied yet. As more replies arrive in the following weeks, BLS updates its numbers.

So each month you get:

  • A first estimate for the latest month.
  • A second estimate for the month before.
  • A third estimate for the month before that.

The changes between these estimates are revisions. They are not mistakes; they are the normal result of more information arriving.

A real example: October 2026

In the October 2026 report, published on 6 November 2026, payrolls rose by 165,000, against a consensus of about 65,000. BLS also revised August and September up by a combined 84,000, to 186,000 and 60,000. So the report was strong twice: a big beat on the headline, and upward revisions on top.

The report before that showed the opposite: July was revised from +21,000 to −10,000, so on the newer count the economy had lost jobs that month. Our November NFP preview explains why that mattered.

The benchmark revision: the big annual reset

Once a year, BLS compares its survey-based count with a much fuller count of jobs from unemployment-insurance tax records, which cover nearly all employers. It uses March as the reference month.

Stage When What it does
Preliminary benchmark Late summer (28 August in 2026) An early estimate of how far the March level is off. Official numbers are not changed yet
Final benchmark With the January jobs report, published in February The whole series is revised, often going back more than a year

Real benchmark numbers from BLS

Benchmark Revision to total nonfarm jobs Published
March 2025 (final) −898,000 (−0.6%), seasonally adjusted 11 February 2026, with the January 2026 report
March 2026 (preliminary) −79,000 (−0.1%); private sector −178,000 28 August 2026

Source: BLS CES benchmark article and preliminary benchmark release. The final March 2026 figure is due with the January 2027 jobs report in February 2027.

BLS says the average absolute benchmark revision over the past 10 years is 0.2%. So the 2025 revision, at 0.6%, was unusually large: it meant the US had about 898,000 fewer jobs in March 2025 than the monthly reports had said. The 2026 preliminary estimate, at 0.1%, was small.

Why revisions move gold

Gold pays no interest. When markets expect higher US interest rates, gold tends to fall; when they expect lower rates, gold tends to rise. The Federal Reserve watches the job market closely when it sets rates.

So the chain is:

What the revisions show What markets may think Usual pressure on gold
Big downward revisions The job market was weaker than we thought; rate cuts more likely, or hikes less likely Up
Big upward revisions The job market was stronger than we thought; rates may stay high or rise Down
Small revisions Nothing new Little effect; the headline and wages decide

Our guide to Fed rate hikes and gold explains why rate expectations matter so much.

Why gold’s first move can reverse

At 8:30 am New York time, fast traders and algorithms react to the headline within seconds. Over the next minutes, people read the details: revisions, the unemployment rate and wages. If the details tell a different story, the first move can reverse.

Example (hypothetical numbers): the headline is +150,000, beating a forecast of 100,000. Gold drops $15 in a minute on “strong jobs”. Then traders see that the previous two months were revised down by a combined 90,000. Net, the economy added fewer jobs than the market thought before the report. Gold climbs back and ends the hour higher.

These numbers are made up to show how revisions can change the reaction. They are not a real report.

What our data shows on two real days

Jobs report What was in it Gold in the release hour (COMEX, hourly bar)
11 Feb 2026 (January report, with the final −898,000 benchmark) Headline and a large downward reset of past levels Bar opened $5,107.5, fell as low as $5,036.3, closed $5,089.6: a $78.7 range
6 Nov 2026 (October report) +165,000 vs about 65,000 expected, revisions +84,000 Bar opened $4,071.4, fell as low as $4,035.1, closed $4,053.9: a $41.6 range

Release at 13:30 UTC on both days. Source: COMEX gold futures hourly bars via Yahoo Finance, from our site data. Many things are in each report, so these moves cannot be put down to revisions alone.

On 6 November 2026, a strong headline plus upward revisions both pointed to a firmer job market, and gold fell in the release hour. Our October NFP reaction article covers that day in detail.

Other revised data gold traders watch

Jobs are not the only revised numbers:

  • GDP: the US Bureau of Economic Analysis publishes an advance estimate, then a second and third estimate for each quarter, and larger annual updates.
  • Retail sales, durable goods and many other monthly reports revise the previous month or two.
  • CPI is not revised month to month in the same way, but seasonal factors are updated once a year, which can change the recent monthly pattern.
  • Jobless claims revise the previous week.

The same rule applies to all of them: the revision can change the meaning of the headline.

A jobs-day reading order

  1. Before the release: know the forecast for the headline, the unemployment rate and wages. The NFP time page shows the release time in your zone.
  2. At the release: do not trade the first seconds. Spreads widen.
  3. Read in this order: headline vs forecast, then revisions to the previous two months, then unemployment rate, then average hourly earnings.
  4. Add the headline and the revisions together. A +150,000 headline with −90,000 of revisions is really +60,000 of “new news”.
  5. Watch whether gold’s first move holds after the details are digested, usually within 15 to 30 minutes.
  6. In February, check the benchmark. The January report carries the annual reset, which can change the story of the whole past year.

Our guide to trading gold on NFP day covers entries, stops and sizing around the release, and the NFP event page lists past and upcoming release dates with gold’s reactions.

Jobs days are among gold’s most volatile. In the two release hours above, gold’s hourly range was $41.6 and $78.7. On a 0.1 lot position, a $40 move is $400. Keep size small, and never hold a position through the release without a stop.

Key takeaways

  • Every US jobs report revises the previous two months; that is normal, not an error.
  • The annual benchmark lines payrolls up with tax records. The final March 2025 revision was −898,000 (−0.6%), far above the 10-year average of 0.2%.
  • Revisions change what the data mean for Fed rates, and so for gold.
  • A strong headline with big negative revisions can flip gold’s first move.
  • Read the headline, revisions, unemployment and wages together before acting.

Sources

  • BLS: CES National Benchmark Article (March 2025 final benchmark, −898,000, published 11 February 2026)
  • BLS: CES Preliminary Benchmark (National) news release, 2026 (March 2026 preliminary, −79,000; 10-year average absolute revision 0.2%)
  • BLS Employment Situation, October 2026, as summarised in our 7 November 2026 news digest (165,000 jobs; August and September revised up by a combined 84,000)
  • XAUUSD Global analysis of COMEX gold futures (GC=F) hourly bars via Yahoo Finance (site data file event-reactions.json)

Educational content only, not financial advice. Trading gold with leverage carries a high risk of loss.

Frequently asked questions

What are NFP revisions?

Each monthly US jobs report gives a first estimate for the latest month and also updates the estimates for the two months before, as more employer surveys come in. Those updates are revisions. They can add or remove tens of thousands of jobs from what was reported before.

What is the payroll benchmark revision?

Once a year the Bureau of Labor Statistics lines up its survey-based payroll numbers with a near-complete count of jobs from unemployment-insurance tax records, for the month of March. A preliminary estimate comes out in late summer and the final one is built into the January jobs report, published in February.

How big was the last benchmark revision?

The final benchmark published on 11 February 2026 lowered March 2025 nonfarm employment by 898,000, or 0.6%, seasonally adjusted. The preliminary estimate for March 2026, published on 28 August 2026, was minus 79,000, or minus 0.1%. BLS says the average absolute benchmark revision over the past 10 years is 0.2%.

Why do revisions move gold?

Because gold reacts to what the jobs data mean for Federal Reserve interest rates. Big downward revisions suggest the job market was weaker than thought, which can make rate cuts more likely or hikes less likely, and that tends to help gold. Big upward revisions point the other way.

Do revisions matter more than the headline?

Sometimes. If the headline beats forecasts but the previous two months are revised down by more than the beat, the overall picture is weaker than it looks. Traders often see gold’s first move on the headline reverse once they read the revisions, wages and unemployment rate.

Comments

No comments yet.

Leave a comment

Comments are checked before they appear. Be respectful: no promotion, no trade signals. Rules · Privacy. No email needed.