After Orders Grew 252%, Vertiv Stopped Disclosing Orders

On 11 February 2026, Vertiv (NYSE: VRT) published its Q4 2025 earnings release. The headline numbers were quarterly organic order growth of +252%, a book-to-bill of 2.9x, and backlog jumping from $9.5bn to $15.0bn — an increase of $5.5bn in a single quarter.

That was the last time the company disclosed any of those three figures.

In the Q1 release of 22 April 2026 and the Q2 release of 29 July 2026, the words "backlog", "book-to-bill" and "organic orders" each appear zero times in the body.

The method first, because this is where the mistake usually happens

The easy error with a finding like this is mistaking "I did not find it" for "it is not there". So here is exactly how it was checked, reproducible by anyone:

1. Pull every 8-K from data.sec.gov/submissions/CIK0001674101.json and filter for those whose items include 2.02 (results of operations).
2. For each, fetch its index.json and locate the exhibit 99.1 htm file — filenames are not standardised, so match loosely on exhibit99.
3. Flatten the HTML to plain text and cut the trailing "Forward-Looking Statements" block. That block contains the standing risk-factor phrase "failure to realize sales expected from our backlog of orders and contracts". Leave it in and every filing matches on "backlog", producing exactly the opposite conclusion.
4. Count terms in what remains.

Step 3 is the whole game. The 2026 release bodies are 6,008 and 6,408 characters respectively, and all three terms return zero — while "backlog" does appear in the complete file, buried in legal boilerplate no one reads.

Ten quarters of continuous disclosure

Source: SEC EDGAR, Vertiv Holdings Co (CIK 0001674101), twelve 8-K exhibit 99.1 filings from Q3 2023 to Q2 2026, parsed filing by filing for this piece.

Look at the middle rows. In three quarters of 2024 the company did swap the quarterly figure for a trailing-twelve-month one, or omit an absolute backlog number. But all three metrics had never gone at once. The two 2026 quarters are the first time.

Why it matters: this is the company's only forward indicator

Vertiv sells power and thermal equipment for data centres, paid out of the customer's capital expenditure budget. The 10-K states that most of the backlog "is expected to be shipped within the next 12 to 18 months".

So reported revenue is the result of orders placed a year to eighteen months earlier. Revenue is the result; orders are the cause. Anyone reading only the income statement is looking at demand from eighteen months ago.

Orders, backlog and book-to-bill were the only quarterly-verifiable forward indicators on that chain. With them gone, the most recent verifiable reading stands at 31 December 2025 — eight months before this was written.

One knock-on effect: the annual backlog disclosure survives, in the 10-K. So the metric has not vanished; its check frequency has fallen to a quarter of what it was.

Do not cross the line the evidence supports

Everything above is verifiable fact. What follows is the step many people take and should not: reading it as "the company is hiding bad news".

This piece does not make that inference, because the contemporaneous numbers do not support it.

Source: Vertiv Q1 and Q2 2026 10-Q filings and earnings releases; gross margin computed from revenue less cost of sales.

A company covering up a demand collapse does not raise full-year guidance in the same document — because guidance gets tested next quarter, and a disclosure convention does not.

Discontinuation has several legitimate motives, each a real corporate practice:

• The metric loses comparability. When a single quarter can print +252% because a handful of hyperscale projects signed, the number stops measuring "demand" and starts measuring "did a mega-deal land this quarter". Publishing a metric that swings twentyfold invites misreading.
• Avoiding a de facto guidance number. Orders and backlog are non-GAAP operating metrics. Once the market anchors on them, the company is tied to a figure it does not control.
• Legal caution. "Backlog" has no GAAP definition, and the company's own 10-K says it is "not necessarily indicative of our net sales for any future period".

Behaviour is verifiable; motive is not. A release can establish which words appeared and which did not. It cannot establish why.

How to actually use this

Not as a reason to be bearish — as a reason to reweight. Three specifics:

One: timestamp every "the backlog is full" argument. Any judgment resting on it has as its latest evidence 31 December 2025. Eight-month-old data is not bad data, but it is eight months old, and its weight should decay accordingly.

Two: promote the intermediate-layer metrics. Orders are gone, but the traces of capacity and delivery are still in the statements: inventory days (74 in Q2 2025 → 113 in Q2 2026), deferred revenue on the balance sheet (+100.2% in six months), and capex ($285.9m in H1 2026, already above the $220m of all of 2025). All three appear in every 10-Q and anyone can pull them.

Three: make "does disclosure return" a tracked item in its own right. If the three terms are back in the Q3 2026 release, that is a repair of information availability and a positive signal by itself. If Q3 comes and goes without them, "no longer disclosed quarterly" has moved from a one-off to the new normal, and every framework that depends on order data has to be rebuilt.

The general method

Most people read filings for what a company said. But in a periodic disclosure, what a company has stopped saying is information too — and almost nobody counts it.

The procedure is simple: flatten eight to twelve consecutive quarterly releases from the same company to plain text, build a term-frequency matrix over a fixed keyword set, and look for the row that suddenly goes to zero. It takes under twenty minutes and catches exactly what reading filing by filing misses, because a reader's attention sits on what is present, not on what was there last time and is not now.

There is one trap, the same one from the top of this piece: strip the legal boilerplate first. Forward-looking statements and risk factors will match almost any keyword and wash the trend out completely.

Where this sits in the report

This check forms the body of Chapter 2 of our Vertiv (VRT) deep dive, second edition. It also overturned and corrected a statement in the first edition, which said backlog was disclosed annually only. That was wrong: it was disclosed every quarter until it stopped in 2026. The correction is printed in Appendix C of the report.

The report ships as two PDFs (35 pages in Chinese, 29 in English; 14 chapters plus three appendices, 14 figures and 20 data tables), with a reference date of the 21 August 2026 close. Three other findings in it come purely from decomposition: EMEA organic growth of −14.8% underneath consolidated organic growth of +19.9%; 27.5 of the 80.8 percentage points of first-half EPS growth coming from the effective tax rate falling from 28.8% to 16.1%; and twelve months of zero insider buying, in which only $100.8m of the $142.1m sold is a genuine reduction.

Related method pieces: what actually counts as insider selling and the blank field in a Form 144. Our evidence grading and process are described in the research method.

Data and sources. All facts come from primary SEC EDGAR filings, accessed 24 August 2026: twelve 8-K exhibit 99.1 filings by Vertiv Holdings Co (CIK 0001674101) from Q3 2023 to Q2 2026, plus the Q1 and Q2 2026 10-Q filings. Term counts were taken on the body text after removing the "Forward-Looking Statements" block.

Disclaimer. This is research content. It does not constitute investment advice, an offer or a solicitation, and is not tailored to any particular investor's financial situation, objectives or risk tolerance. We hold no investment-adviser licence in any jurisdiction. Past performance does not indicate future results.