Every annual report includes a section where management talks about what comes next. Revenue targets. Capital expenditure plans. Expansion ambitions. Margin trajectories. Project timelines. Most investors read it, form an impression, and move on.
The Q Factor takes a different approach. We extract every forward-looking commitment from every annual report we cover, then check the following year to see what actually happened. After several years of tracking, the database holds more than 19,000 commitments — 6,024 from the most recent reports alone — across 411 companies on four exchanges. This article presents what that data shows.
Key Insight: Across 411 companies and four exchanges, the delivery rate for forward guidance ranges from 35% to 81%. The gap is explained less by management quality than by what the commitment was about. Financial outcomes (revenue, margins, earnings) get delivered around 85% of the time. Operational milestones — expansion, product launches, project delivery — get delivered closer to 45%.
What forward guidance actually means in an annual report
Forward guidance is the formal name for the forward-looking statements that appear throughout an annual report. They are concentrated in the chairman's letter, the chief executive's report, and the strategic review section, though they also appear in financial commentary and risk disclosures.
A forward commitment, in our framework, is a specific, falsifiable statement about what management intends to do or expects to happen. "Revenue is expected to grow by 8 to 10 percent in FY26" is a commitment. "We are confident in the company's long-term prospects" is not.
We distinguish between commitments and aspirational language because the latter cannot be verified. A statement is only useful as a tracking signal if it admits two outcomes — delivered, or not delivered.
Forward guidance covers a wide range of subjects: revenue and earnings projections, capital expenditure plans, new market entry, product launches, margin targets, dividend policy, cost reduction programmes, and operational milestones. Each carries different weight, different verifiability, and as the data will show, different likelihood of being delivered.
How we tracked 6,000+ commitments across 411 companies
The Q Factor uses Claude AI to read every annual report we cover. As part of that analysis, the model is instructed to extract every commitment that meets two criteria: specific enough to be testable, and tied to a defined timeframe (next year, three years, by 2027, and so on).
Each commitment is stored with its source quote, the year it was made, the category it falls under, and whether the company indicated it was a measurable target or directional intent. Twelve months later, when the next annual report is published, we check.
The check itself is structured. The new report either acknowledges the previous commitment (delivered, partially delivered, missed, abandoned), or it does not — meaning the commitment has disappeared from the narrative. Each commitment receives an outcome status. Where the answer is genuinely too early to tell, it is recorded as pending.
The current snapshot — companies whose most recent annual report is on file — contains 6,024 commitments across 411 companies. The full multi-year history contains 19,426 commitments, covering up to seven years of data per company in the older parts of the universe. The patterns presented below are drawn from the full dataset, with the most recent year shown separately where the comparison is informative.
Delivery rates by exchange: a 47-point spread
The first finding is that delivery rate varies enormously by market.
| Exchange | Companies | Commitments (current year) | Delivery rate |
|---|---|---|---|
| US | 152 | 456 | 81.4% |
| ASX | 154 | 2,979 | 58.8% |
| SGX | 12 | 193 | 45.1% |
| NZX | 93 | 2,396 | 34.6% |
Delivery rate is calculated on decided outcomes only. A commitment too early to assess (pending) is excluded from both numerator and denominator. This matters: a company that makes longer-horizon commitments appears to deliver less if pending outcomes are counted as failures.
The US figure of 81.4% requires a caveat. The US coverage in The Q Factor is currently a curated 152-company sample of well-known names. It is not yet representative of the broader US market, and the headline figure should not be extrapolated to S&P 500 averages.
The other three exchanges are more complete. ASX delivers 58.8%, just under three-fifths. SGX comes in at 45.1%. The NZX delivers 34.6% — just over a third.
The NZX figure is the article's hardest finding. The likeliest explanations are not pleasant. NZX companies tend to make more numerous and more specific commitments per report than their ASX counterparts, partly because the NZX disclosure regime encourages it. Once a commitment is on the page, it can be checked. The more you commit to, the more there is to miss.
That said, market-level averages obscure individual performance. There are NZX companies with delivery rates above 90%. There are ASX companies below 25%. The exchange figure is not a judgment of any individual company within it.
The financial-vs-operational divide
The second finding is more important than the first.
Delivery rate varies far more by what the commitment was about than by which exchange it came from. We grouped commitments by category and calculated delivery rates for each.
| Category | Commitments | Delivery rate |
|---|---|---|
| Revenue | 4,485 | 89.0% |
| Margins | 3,946 | 87.7% |
| Investments | 3,912 | 82.8% |
| Earnings | 552 | 81.7% |
| Cost | 379 | 64.0% |
| Capital expenditure | 559 | 63.0% |
| Expansion | 734 | 57.4% |
| Operational | 3,388 | 46.6% |
| Dividend | 802 | 45.2% |
| Product | 350 | 42.1% |
The pattern is clear. Commitments about financial outcomes — revenue, margins, investments, earnings — are delivered between 82 and 89 percent of the time. Commitments about operational milestones — expansion plans, product launches, operational targets — are delivered between 42 and 57 percent. Capital allocation commitments sit in between.
The reason is not difficult to understand. Financial outcomes are externally validated. Revenue is reported quarterly. Margins are visible in the cost line items. The numbers either grow or they do not, and the auditor signs off. Management has every incentive to commit conservatively to financial targets, because committing aggressively and missing is publicly humiliating.
Operational milestones are different. A factory expansion that slips by six months is announced quietly in the next operational review. A product launch delayed from H1 to H2 rarely makes the headlines. A capital project that overruns by 20% on cost is folded into the next year's capex line. The penalty for missing is lower, and management can therefore be more ambitious in what they promise.
This has implications for how investors should read forward guidance. Take revenue and margin commitments seriously — they are usually delivered, and when they are missed it is genuinely informative. Take expansion plans, product roadmaps, and project timelines with significant scepticism. Roughly half of them will be late, abandoned, or quietly redefined.
The dividend surprise: 45% delivery
The most counter-intuitive finding sits in the table above. Dividend commitments are delivered only 45.2% of the time.
This is initially surprising. Dividends are arguably the most visible commitment a company makes. Boards announce them publicly. Investors track them closely. Yield-focused holders depend on them. One would expect the delivery rate to be very high.
The data tells a different story. There are three plausible explanations.
First, dividend commitments are often made as multi-year progressive policies. "We expect to grow the dividend by 5 to 10 percent per annum" is effectively a five-year commitment in a single sentence. If any year in the cycle requires a cut, the entire commitment is recorded as not delivered even if four of the five years met the target. The granular tracking penalises the long-dated promise.
Second, dividends are the first cash to be cut when conditions tighten. Capex can be deferred. Hiring can be paused. But the dividend forecast is the company's last line of defence when cash flow shrinks. In our data, dividend miss rates spike in years of sector downturn — REITs in rising-rate environments, energy companies in commodity declines, retailers in consumer pullbacks.
Third, dividend commitments are sometimes interpreted as more specific than they were intended. A chairman writing "we remain committed to returning capital to shareholders" intends a directional statement; the extraction process may record it as a verifiable commitment. We continue to refine the distinction in the extraction layer.
The investor takeaway is not that dividends are unreliable. In most years, most companies do deliver. The takeaway is that "committed to a progressive dividend policy" should be read as guidance under favourable conditions, not as a guarantee.
How to read forward guidance yourself
The point of presenting this data is not to suggest that investors should distrust forward guidance. Most of it is genuinely delivered. The point is that not all commitments carry equal weight, and a few heuristics make the section much more useful to read.
1. Separate the financial commitments from the operational ones. Revenue and margin targets are usually delivered. Expansion timelines and product roadmaps are usually not. Apply different scepticism to each.
2. Look for specificity. A commitment that names a date, a percentage, or a dollar figure is one the company expects to be held to. A commitment that uses words like "around", "broadly", or "in line with" is hedged and should be weighted less.
3. Track what disappears. The most useful tracking signal is not the missed commitment — which the company will usually acknowledge in the next year's discussion — but the disappeared commitment. A specific target from last year's report that simply does not appear in this year's report is a quiet downgrade. Most companies do this at least once per cycle. Some do it routinely.
4. Focus on the strategic review, not the chairman's letter. Chairman letters are written for shareholders and tend toward optimism. Strategic review and operational sections are written for the regulator and tend toward defensibility. Commitments in the latter are usually firmer.
5. Treat a single miss as information, not condemnation. A commitment abandoned because the market moved is different from a commitment abandoned because the company never had a plan. The pattern across years matters more than any single miss.
The Q Factor tracks more than 19,000 commitments across 411 companies because the pattern is what carries the signal. Any single year reveals little. Five years of delivery data reveals everything.
Forward guidance, read carefully, is one of the more honest sections of an annual report. The companies that consistently deliver on the commitments that matter — revenue, margins, capital allocation — are usually run by management teams worth taking seriously. The companies that consistently fall short, particularly on commitments they made specifically and publicly, deserve closer scrutiny than the headline numbers might suggest.
Browse the management credibility scores and delivery patterns for any company in our universe at theqfactor.io/stocks. For the methodology behind how we read and score annual reports, see our guide to how to read management promises and the explanation of our management credibility scoring. For the broader case behind combining qualitative and quantitative analysis, the 70/30 Split article goes deeper.
This educational content is part of The Q Factor's methodology documentation. This is not financial advice. Past patterns may not predict future performance. Always conduct your own research before making investment decisions.