A retailer’s annual report is one of the few documents where a company has to describe its own business under a legal standard of accuracy, and it is also one of the least read. Most coverage of a big retail filing lasts a single news cycle and focuses on two numbers: headline revenue and earnings per share. The parts that actually predict the next twelve months (how comparable sales are defined, whether inventory is growing faster than demand, what the lease book commits the company to, and which risk factors were rewritten this year) sit further back, in sections that look like boilerplate until you know what changed. This guide walks through a retail annual report the way an analyst reads it, section by section, with the traps that catch first-time readers.
In short
- The annual report is two documents in one: a designed marketing narrative and a statutory filing. For US-listed retailers the statutory filing is Form 10-K, filed with the Securities and Exchange Commission, and it is the version that carries legal weight.
- Comparable sales is not a defined accounting term. Each retailer sets its own rules for which stores qualify, how long a store must be open first, and whether digital orders count. Two chains reporting “comps up 3%” may be measuring different things.
- The inventory-to-sales spread is the single most useful early warning in retail. When inventory grows materially faster than sales for two or three consecutive quarters, markdowns usually follow.
- Lease commitments moved onto the balance sheet under current US and international lease accounting standards, but the useful detail (remaining term, discount rate, maturity ladder) still lives in the notes rather than on the face of the statements.
- Risk factors matter only as a diff. Compare this year’s list to last year’s: new items, reordered items and newly specific language are the signal, and the unchanged pages are the noise.
The structure of a retail annual report
Before reading anything, work out which document you are holding. Retailers routinely publish a designed “annual report” with photography, a chief executive letter and selected highlights, and separately file a dense statutory document with the regulator. They overlap, but they are not the same, and only one of them is prepared under a liability standard.
The US filing: Form 10-K
For a US-listed retailer, the document that matters is Form 10-K, filed with the SEC’s EDGAR system. Its structure is prescribed, which is exactly what makes it useful: the same information sits in the same place every year and at every company. Item 1 describes the business, Item 1A lists risk factors, Item 2 covers properties (in retail, that means the store estate), Item 3 covers legal proceedings, Item 7 is Management’s Discussion and Analysis, and Item 8 holds the audited financial statements and the notes.
The notes are where most of the real work happens. A retail income statement is four or five lines; the notes behind it run for dozens of pages and contain the segment breakdown, the lease disclosures, the inventory accounting policy and the revenue recognition detail. Readers who stop at the statements miss almost everything that distinguishes one year from another.
The glossy annual report versus the statutory filing
The designed annual report is a communications product. It is not worthless: the chief executive letter tells you which two or three themes management has chosen to own publicly, and a shift in those themes between years is itself information. But the selected figures in it are chosen, and the adjusted metrics in it are often presented without the reconciliation that the statutory filing is required to include.
A practical habit: read the letter first for framing, then ignore it and work only from the filing. If a number in the letter does not appear in the filing in the same form, treat it as a management-defined metric and find out how it was built before using it.
Where UK and EU retailers differ
Outside the US the packaging differs. UK-listed retailers publish an annual report and accounts that combines the narrative, the strategic report, governance and remuneration disclosures, and the audited accounts in one volume, and they report under IFRS rather than US GAAP. European filers vary by market. The underlying questions in this guide do not change, but the labels do: “like-for-like sales” is the common UK term for what US retailers call comparable sales, and lease accounting sits under IFRS 16 rather than the US standard.
| Section | What it actually tells you | Common trap |
|---|---|---|
| Business description (Item 1) | Segment definitions, channel mix, store formats, sourcing footprint | Reading it as marketing copy and skipping the segment definitions, which set how everything later is grouped |
| Risk factors (Item 1A) | What management is contractually willing to admit could go wrong | Reading this year’s list in isolation instead of diffing it against last year’s |
| Properties (Item 2) | Store count, square footage, owned versus leased split, distribution centers | Assuming store count is comparable year to year after a format change or an acquisition |
| MD&A (Item 7) | Management’s own attribution of what moved sales and margin | Accepting the attribution without checking it against the inventory and markdown data in the notes |
| Financial statements (Item 8) | The audited numbers | Stopping at the four headline lines and not opening the notes |
| Notes to the accounts | Leases, inventory policy, segments, commitments, contingencies | Treating the notes as technical appendix rather than the primary source |
If you are new to the sector, the vocabulary in these filings assumes a working model of how the industry is organised. Our explainer on what the retail industry is today and how it really works covers the channel and format definitions that annual reports use without defining, and the wider piece on how retail news shapes the global e-commerce industry sets out why a single filing often moves an entire category’s narrative for a quarter.
Comparable sales and why the definition matters
Comparable sales, comps, like-for-like sales and same-store sales all describe the same idea: growth stripped of the effect of opening and closing stores. The idea is sound. The execution is company-specific, because comparable sales is not defined by US GAAP or IFRS. It is a management-defined metric, and management writes the definition.
What “comparable” actually excludes
Find the definition before you use the number. It is usually a short paragraph in MD&A or in a glossary at the back. The variables to check are the waiting period before a new store enters the base (commonly 12 to 15 months, but not always), whether remodeled or relocated stores stay in the base or drop out, how temporarily closed stores are handled, and whether the calculation is on a constant-currency basis for international operations.
Each of these is defensible on its own. Together they create real differences. A chain that removes remodeled stores from the base during the remodel is reporting a cleaner comp than one that leaves them in while they trade at reduced capacity, and neither approach is wrong. What matters is that you know which you are reading before comparing two companies.
Definition changes deserve specific attention. When a retailer changes its comp definition, the change is normally disclosed, often briefly, and the prior-year comparative may or may not be restated. A definition change that improves the reported number in a weak year is not evidence of anything by itself, but it is a reason to rebuild the comparison from underlying data rather than trusting the trend line.
Digital sales inside the comp base
The largest single source of confusion in modern retail comps is where digital revenue sits. Practice varies: some retailers include all digital orders in the comparable base, some include only orders fulfilled from or collected at stores, and some report a separate digital growth figure alongside a store-only comp.
The distinction matters most for retailers with heavy click-and-collect volume, because an order placed online and collected in store can legitimately be counted as either. When a company includes it in store comps, a shift in customer behavior toward collection can flatter store performance without any change in demand. Read the definition, then read what the company says about channel mix in the same section, and check whether both can be true at once.
Traffic versus ticket
A comp number is a product of two things: how many transactions happened and how large each one was. Most retailers disclose the split, either as transaction count and average transaction value or as traffic and ticket. This decomposition is where the useful reading is.
Comp growth driven by ticket alone, with traffic flat or negative, is usually price rather than demand. In an inflationary period that can persist for several quarters and look healthy in the headline. Comp growth driven by traffic with flat ticket is the harder and more durable achievement. Comp growth where both rise is rare and worth checking against the promotional calendar. Retail segments behave differently here, and our breakdown of retail industry segments from grocers to luxury is a useful reference for what a normal traffic and ticket mix looks like in each.
| Definition variable | Typical options | Effect on the reported number |
|---|---|---|
| New store waiting period | 12, 13 or 15 months, sometimes a full fiscal year plus one period | A longer wait excludes the opening-volume spike and produces a more conservative comp |
| Remodeled stores | Kept in base, removed during works, or removed for a set number of months after reopening | Removing disrupted stores raises the reported comp during an active remodel program |
| Relocations | Treated as continuing, or as a closure plus an opening | Treating them as continuing keeps the new, usually better, location in the base sooner |
| Digital orders | All included, store-fulfilled only, or excluded and reported separately | Full inclusion blends two very different growth rates into one number |
| Currency | Reported or constant currency | Constant currency isolates volume but hides real cash effects for the group |
| Fiscal calendar | 52 or 53-week years, shifted holiday weeks | A 53rd week inflates the annual total and requires an adjusted comparison |
Inventory levels versus sales growth
If you read only one relationship in a retail annual report, read this one. Inventory is the sector’s dominant working capital item, it is bought months before it sells, and it is the mechanism by which a demand miss becomes a margin miss. The relationship between inventory growth and sales growth is visible in every filing and requires no adjustment to compute.
The spread that matters
Take the year-over-year change in period-end inventory and subtract the year-over-year change in sales for the same period. A small positive spread is normal and can be deliberate: a retailer opening stores, extending assortment or building ahead of a known supply constraint will carry more inventory. A large positive spread sustained across consecutive periods is a different matter, because the goods have to clear somehow, and the usual route is markdown.
Read the spread alongside the explanation. Management almost always addresses it in MD&A, and the explanations fall into a few recognizable categories: timing of receipts near the period end, a deliberate pull-forward ahead of supply chain or tariff risk, an assortment expansion, or an acquisition. Each is checkable. Timing explanations should reverse in the following quarter; if the same timing explanation appears twice, treat it as a trend rather than a timing effect.
Inventory turns and days on hand
Turns (cost of goods sold divided by average inventory) and days inventory outstanding (365 divided by turns) put the level into context. Absolute values are not comparable across formats: a grocer turns inventory many times a year, a jeweller or a furniture retailer turns it very few times, and neither is better. Compare a retailer to its own history and to direct format peers only.
Use average inventory rather than a single period-end figure where you can. Retail balance sheet dates often sit just after a major selling season, which is exactly when inventory is at its lowest point of the year, and a year-end snapshot can make the book look leaner than it was for the preceding eleven months.
Where inventory quality hides
The level tells you how much. It does not tell you how old or how saleable. Three disclosures help. First, the inventory accounting policy note states the method (variants of cost or retail inventory method) and the basis for reserves. Second, any change in the reserve or in the estimation approach is disclosed and is worth reading closely. Third, in some filings a composition split between raw materials, work in progress and finished goods is given, which matters for vertically integrated retailers.
Gross margin, markdowns and what moved them
Gross margin is where merchandising decisions become visible. In a retail MD&A, the margin discussion is usually written as a bridge: a list of the factors that moved the rate up or down, sometimes with basis points attached to each. That bridge is the highest information density paragraph in the document.
Mix, markdown, freight and shrink
Four factors do most of the work. Mix is the change in what sold, and it moves margin without any change in pricing discipline: a strong season in a low-margin category drags the blended rate down even in a good year. Markdown is realized discounting, and it is the direct read on whether the inventory position was correct. Freight and other supply chain costs move with fuel, carrier rates and mode choice, and they can swing the rate materially in a volatile year. Shrink is inventory loss, and where a retailer quantifies it, the disclosure is usually driven by the number having become large enough to require explanation.
Separate the factors management can control from the ones it cannot. A margin decline driven by freight is a different business problem from one driven by markdown, even if the basis points are identical. The first tends to reverse when rates normalize; the second reflects a buying decision that has already been made and may be repeated.
Where occupancy sits
Check whether occupancy costs (rent, depreciation on store assets, utilities) are inside cost of sales or in operating expenses. Both treatments exist and both are permitted, and the choice materially changes the reported gross margin rate. A retailer that puts occupancy in cost of sales will show a structurally lower gross margin than a peer that does not, with no difference in underlying economics.
This is the most common error in cross-company margin comparisons. The accounting policy note states the treatment. Read it before building any comparison table, and if the treatments differ, compare operating margin instead, which is unaffected by where the line sits.
Adjusted numbers and their reconciliation
Retail filings are full of adjusted measures: adjusted EBITDA, adjusted operating income, adjusted earnings per share. US rules on non-GAAP financial measures require that such measures be reconciled to the most directly comparable GAAP measure, and that the GAAP measure is presented with equal or greater prominence. The reconciliation table is therefore in the document by requirement, and it is the part to read.
The question is not whether adjustments are legitimate. Many are. The question is whether the same adjustment appears every year. A restructuring charge excluded once is an exclusion; the same charge excluded in five consecutive years is an operating cost the company has chosen to present separately. Recent coverage of how retailers deploy one-off gains, including our analysis of why the tariff-refund windfall is likely to end in buybacks rather than price cuts, shows how much of the story sits in items management describes as non-recurring.
Store counts, openings and closures
The store estate is the part of a retail annual report that most directly forecasts the next two years, because opening and closing decisions are made well in advance and are expensive to reverse. The data is usually in Item 2 (properties) and in a store activity table in MD&A.
Gross versus net, and the roll-forward
Net store count is the weakest available figure. A chain that opened 40 and closed 38 reports the same net change as one that opened 3 and closed 1, and the two businesses are doing completely different things. Look for the roll-forward: opening count, openings, closures, acquisitions, conversions, closing count. Most retailers provide it; when one does not, that is worth noting.
Closure programs are announced in stages, and the annual report usually gives the total program size, the number completed and the charge taken. Compare the program as originally announced to the current figure, because expansions of a closure program are a stronger signal than the original announcement. The pattern is visible across the sector: our report on Cato tripling its store closure program to 120 locations is a clear example of a target that moved well beyond its initial framing within a single year.
Square footage and sales per square foot
Store count without square footage is incomplete, because format changes are common and a chain can shrink its selling space while growing its store count. Total selling square footage, average store size and sales per square foot together describe whether the estate is being grown, densified or shrunk.
Sales per square foot deserves the same caution as comps. Check whether the denominator is total or selling square footage, whether it is a period-end or average figure, and whether digital sales are in the numerator. Retailers that attribute online orders to the fulfilling store can report productivity gains that reflect a fulfilment decision rather than store performance.
Lease obligations and off-balance-sheet commitments
For most retailers the lease book is the largest long-term commitment the business has, larger than its debt in many cases. Lease accounting changed substantially in recent years, and understanding what changed is necessary to read older filings alongside current ones.
What the current lease standards changed
Under the current US standard issued by the Financial Accounting Standards Board, and under IFRS 16 for international filers, operating leases are recognized on the balance sheet as a right-of-use asset and a corresponding lease liability. Before those standards took effect, most retail operating leases were disclosed in the notes rather than recognized on the balance sheet, which is why analysts historically built their own capitalized-lease estimates.
The practical consequence for a reader today is that balance sheet comparisons spanning the transition are not like for like, and that leverage ratios calculated on pre-transition and post-transition years are not comparable without adjustment. Check which basis a multi-year table in the filing is using, because retailers do not always restate the full history.
The disclosures that still matter
Recognition put a number on the balance sheet but the analytical detail remains in the notes. Three items are worth extracting every year. The weighted-average remaining lease term tells you how locked in the estate is: a short average term means flexibility to exit but also imminent renewal negotiations at current market rents. The weighted-average discount rate affects the size of the liability and is a rough read on the company’s incremental borrowing cost. The maturity ladder of undiscounted lease payments shows the cash commitment year by year, which is what actually has to be paid regardless of accounting treatment.
Commitments and guarantees
Beyond leases, the commitments note typically covers purchase obligations (inventory commitments, technology contracts, media buys), construction commitments, and any guarantees the company has given. For retailers with franchise or wholesale partners, guarantees of partner obligations are a genuine off-balance-sheet exposure and are disclosed here rather than in the statements.
Risk factors: separating boilerplate from signal
Risk factors have a reputation for being lawyer-written padding, and much of the section deserves it. A typical retail risk factor list warns about consumer spending, competition, supply chain disruption, cybersecurity, key personnel and weather, and most of those paragraphs are near-identical across the sector and across years. The signal is not in the content. It is in the delta.
Read it as a diff, not as a list
Download last year’s filing and this year’s, extract Item 1A from both, and compare them mechanically. Three kinds of change matter. New risk factors mean something entered the company’s disclosure calculus during the year, and adding one is a deliberate decision made with counsel. Removed risk factors mean a risk was judged resolved or immaterial, which is equally deliberate. Reordered factors matter because most filers put what they consider most significant first.
The fourth and subtlest change is specificity. A risk factor that moves from “we may be affected by changes in trade policy” to naming a particular tariff schedule, a compliance deadline or a share of cost of goods exposed has been rewritten for a reason. Vague to specific is the direction that carries information; the reverse is rare.
Cross-check against the rest of the filing
A risk factor is worth more when it is corroborated elsewhere. If a new risk about supplier concentration appears in Item 1A, check the business description for a customer or supplier concentration disclosure, and check the commitments note for related purchase obligations. When three sections point at the same exposure, it is an operating reality rather than protective drafting.
The same technique works in reverse. A risk factor that appears with no corroboration anywhere else in the document is usually genuine legal caution and can be discounted. For the wider context of which risks the sector is actually carrying into the next cycle, our 2026 retail industry outlook maps the exposures that have been appearing across filings rather than in any single one.
A reading order that works in about 45 minutes
You do not need to read a retail annual report front to back, and almost nobody does. The following order gets the substance out of a filing efficiently, and it works because each step gives you the context needed to judge the next.
- Open the store roll-forward and square footage table first. Two minutes, and it tells you whether this is a growing, stable or shrinking estate, which frames every other number.
- Find the comparable sales definition and read it in full before looking at the comp number itself.
- Pull the comp decomposition into traffic and ticket, and note which one is carrying the growth.
- Compute the inventory-to-sales spread from the balance sheet and income statement, then read what MD&A says about it.
- Read the gross margin bridge and write down the factors in order of magnitude.
- Open the lease note and extract three figures: weighted-average remaining term, weighted-average discount rate, and next year’s undiscounted payment.
- Diff Item 1A against last year’s filing. New, removed, reordered, and vague-to-specific.
- Read the non-GAAP reconciliation and check whether any adjustment repeats across years.
- Only now read the chief executive letter, and note where the narrative and the filing disagree.
Repeat the same sequence every year for the same company and the exercise compounds, because the value is almost entirely in the year-over-year comparison rather than in any single reading. Keeping a short standing file per retailer, with those nine outputs recorded, turns each new filing into a 20-minute update. The way a single filing propagates through the sector’s coverage is covered in our guide to how retail news shapes the global e-commerce industry, which is useful context for judging how much of a post-results move is information and how much is framing.
A note on scope: this article is general information and education about how retail financial disclosures are structured, and it is not legal, accounting, tax or investment advice. Reporting requirements, accounting standards and disclosure rules change, they differ by jurisdiction, and the summaries here are simplified. Anyone relying on a filing for a decision should read the original document from the official source (EDGAR for US filers, the relevant national register or the company’s own investor relations pages elsewhere) and consult a qualified accountant, financial adviser or attorney about their specific situation. Nothing here is a recommendation to buy or sell any security.
FAQ on reading retail annual reports
What is the difference between an annual report and a Form 10-K?
The annual report is often a designed communications document containing a leadership letter, highlights and selected figures. Form 10-K is the statutory annual filing that US-listed companies submit to the SEC, with a prescribed structure and audited financial statements. Some companies wrap the 10-K inside the annual report and distribute them together. When the two disagree in emphasis, the filing is the authoritative version.
Is comparable sales an official accounting measure?
No. Comparable sales, like-for-like sales and same-store sales are management-defined metrics rather than measures defined under US GAAP or IFRS. Each company sets its own rules for which stores qualify and whether digital orders are included, and it discloses that definition in the filing. Read the definition before comparing the figure to another retailer’s.
Which single number best predicts a retailer’s next few quarters?
There is no single number, but the inventory-to-sales spread is the most useful one available without adjustment. Inventory growing materially faster than sales across consecutive periods usually precedes markdown activity and gross margin pressure, because the goods have to clear. Read it alongside the management explanation rather than in isolation.
Where do I find a retailer’s lease commitments?
The lease liability and right-of-use asset appear on the balance sheet under current US and IFRS lease standards, but the useful detail is in the lease note: the weighted-average remaining term, the weighted-average discount rate, the split between fixed and variable payments, and the maturity ladder of undiscounted payments by year.
How should I treat adjusted earnings figures?
Treat them as a starting point that needs checking. US rules require non-GAAP measures to be reconciled to the closest GAAP measure, so the reconciliation is in the document. The test is consistency: an adjustment made once may well be genuinely one-off, while the same adjustment repeated across several years is better read as a recurring cost presented separately.
Are risk factors worth reading at all?
Only as a year-over-year comparison. The bulk of a risk factor section is standard sector language that changes little. What carries information is which factors are new, which were removed, which moved up the order, and which were rewritten from general language to something specific about amounts, counterparties or dates.
How do UK and US retail filings differ in practice?
UK-listed retailers report under IFRS and publish a combined annual report and accounts containing the strategic report, governance and remuneration sections alongside the audited accounts. US filers use the prescribed Form 10-K structure under US GAAP. The analytical questions are the same, but terminology differs, including like-for-like versus comparable sales, and UK reports include a directors’ viability statement with no direct US counterpart.
Where can I get these filings for free?
US filings are freely available on the SEC’s EDGAR database, including full-text search across filings. UK and other international filers publish annual reports on their investor relations pages, and many markets maintain a national filing register. Company investor relations sites also carry results presentations and transcripts, which are useful supplements but do not replace the filing itself. Background on the metric conventions is available in general references such as the same-store sales entry on Wikipedia.