How to analyze royalty performance: the licensor's portfolio guide

What royalty performance analysis is, and what it is not

Every licensor receives royalty reports one at a time, and most licensing teams are good at reviewing them one at a time: does the report foot, are the rates the contracted rates, are the deductions permitted, does the cash match the payable. That review is necessary and it is covered in the guide on how to read a royalty report, linked at the end. Royalty performance analysis starts where the single-report review stops. It takes every report the program has received, puts them on one grain, and asks questions no individual report can answer: which licensees carry the program, which are drifting, which will miss their guarantees, which properties and categories are growing, and where the income the agreements should produce is not arriving.

The distinction matters because the two reviews find different things. A report review finds errors inside a submission: a stale rate, a deduction that the agreement does not name, an advance balance that does not roll. A performance analysis finds patterns across submissions: a licensee whose deduction rate has risen for four consecutive quarters while staying inside its cap, a category whose unit volume has been flat for a year while its gross sales grew, a guarantee that has been met every year by a shortfall payment rather than by sales. Each of those can sit inside a series of reports that all pass review individually. A program can pass every report review and still be underperforming, mispriced or concentrated in ways that only the portfolio view shows.

The analysis is also not forecasting, though it feeds it. A licensee forecasting its royalty expense projects what it will owe each licensor; that process is covered in the guide on forecasting royalty expense. The licensor's analysis looks back first, at what was earned and why, and projects forward only far enough to see the guarantee positions and the shortfall exposure the current year is building. And it is not an audit. An audit tests a licensee's records against its reports under the agreement's audit clause. Performance analysis is what tells a licensor which licensees, which periods and which questions an audit should be pointed at, and which ones a query will settle without one.

The output of a good analysis is short. A ranked view of licensees by royalty income and by change; the guarantee position of every agreement, with the projected shortfall where there is one; a list of signals, each tied to an agreement, a period, a measure and a proposed next step; and the handful of conversations the licensing team should have before the next quarter closes. Everything in this guide is in service of producing that view reliably, every quarter, from data that can be traced back to the reports it came from.

Who runs the analysis, and what each reader is deciding

The same analysis serves several readers, and it is worth being clear about what each one is deciding, because that determines which measures lead the review. The licensing team is deciding where to spend relationship time: which licensees to grow, which to push on product and distribution, which to bring into a renewal conversation early, and which to let lapse. For that team the leading measures are royalty income by licensee and property, same-period growth, MG coverage and the mix of channels and categories each licensee sells through.

The licensor's finance team is deciding what the program will deliver and whether the cash matches the entitlement. It leads with guarantee positions, projected shortfall exposure, the split between earned royalties and shortfall payments in the year's income, reporting timeliness and the value of restatements. A finance team that sees the portfolio only as the sum of payments received cannot tell an agreement that is earning its guarantee from one that is paying it, and the difference is the most important single fact about the agreement's renewal.

Licensing agents running programs for licensors are deciding both at once, across many licensees, and usually reporting upward to the licensor on a fixed cadence. The guide on how licensing agencies manage royalty reporting covers the validation machinery an agency builds; this guide covers the analysis that machinery feeds. Where a brand owner runs licensing as a division, the readers include the brand team, which is deciding whether licensed product is positioned where the brand wants it: an apparel extension that has drifted into off-price channels may be meeting its guarantee and still working against the brand's own distribution strategy.

Finally, a licensee running the same kind of analysis across its own licensors is answering a mirror-image question: which licenses earn their cost, which guarantees are at risk, and which renewals deserve better terms. Most of the measures transfer directly, and a short section near the end of this guide sets out what changes from the licensee's seat. The core of the guide stays on the licensor side, because that is where the portfolio view is least often built and most often needed.

The dataset: one grain for every report

Analysis cannot start until every report the program receives has been put on one grain. Licensees report in different templates, at different levels of detail, on calendars that may or may not match the agreement's reporting periods, and in currencies the agreement may or may not name. The first job is to bring all of it into a single table with a consistent row: one row per agreement, property, product category, channel, territory and reporting period, carrying units, gross sales, deductions by type, net sales, the contract rate that applied, earned royalty, and the currency the figures were reported in.

Each of those columns earns its place. The agreement column is what ties every figure to a set of terms: the rate card, the deduction permissions and caps, the guarantee, the advance and the contract year. Property matters wherever one agreement licenses several marks or one licensee holds several agreements: a league program with team marks, a collegiate licensee reporting across many institutions, an entertainment licensee selling product for several films or series under one agreement. Category and channel are where rates and deduction behavior differ. Territory is where currency and grant questions live. Units are needed for every per-unit measure and for the price checks later in this guide; a dataset without units can analyze dollars and nothing else.

Two date columns matter more than they look. The reporting period is the period the agreement says the report covers; the contract year is the measurement period the guarantee, any tier thresholds and any annual caps are measured against. Tag every row with both. A licensor whose contract years start on July 1 for one licensee and January 1 for another cannot compare guarantee pacing, tier progress or year-over-year growth without knowing which contract year each row belongs to, and reconstructing that later is far harder than recording it when the report arrives.

Keep versions. A report that is restated, or corrected by a prior-period adjustment on a later report, changes figures in a period that has already been analyzed. The dataset should hold the figures as originally reported and as currently restated, with the date each version arrived, so that the analysis can be run on current figures while the restatement history remains measurable in its own right. Overwriting the original figures destroys one of the most useful signals a licensor has about a licensee's reporting quality.

Finally, record what is missing. A report that arrives without the category or channel detail the reporting clause requires cannot be placed on the grain, and the temptation is to allocate its totals across categories using last period's mix. Resist it, or at least flag every allocated row as allocated. An analysis built on allocated rows will reproduce last period's mix faithfully and find nothing, which is the outcome a licensee withholding detail would choose.

The measures, and the formula for each

The table below lists the measures this guide uses, each with its formula and what it is for. Most are ratios rather than totals, because totals in royalty reporting are almost always explicable on their own and ratios move only when something about the business or the reporting has changed. Every ratio has to state its denominator, and the denominators here are chosen deliberately: gross sales for the effective rate and the deduction rate, because gross is the figure deductions have not yet touched; the guarantee for coverage; units for every per-unit measure.

Three conventions apply throughout. Earned royalty means the royalty the agreement produces on reported sales at contract rates, before any guarantee top-up; paid royalty adds any shortfall payment. Every measure is computed per agreement first and only then rolled up, because a ratio computed on portfolio totals is a weighted average that can move when nothing in any agreement moved, simply because the weights did. And every comparison is between the same period in two contract years unless the measure is explicitly a pacing measure, for reasons the sections on period comparison set out.

The sections that follow take the measures in turn: what each one shows, what moves it legitimately, and what kind of movement deserves a question. The worked example near the end then computes every one of them for a three-licensee portfolio, so the formulas can be checked against figures.

Royalty income by licensee, property, category and territory

The first cut is the simplest: earned royalty, summed along each dimension of the grain, for the current period, the same period of the prior contract year, and the contract year to date. By licensee, it shows who carries the program. By property, it shows which marks earn and which are licensed but dormant. By category, it shows where the program's product actually is, which is often not where the licensing strategy says it should be. By territory, it shows how much of the income depends on the markets the licensor knows least.

Keep earned, paid and received as three separate figures. Earned royalty is what the agreement produces on reported sales. Paid royalty adds any guarantee shortfall settled in the period. Received is the cash that arrived, which can differ from both because of timing, short payments or recoupment against an advance. A program report that shows only received cash will show nothing at all for a licensee during its recoupment period, and will show a sudden jump in the year a large shortfall is settled; neither is a fact about sales. An analysis of performance runs on earned royalty, and reports paid and received alongside it so finance can reconcile.

Property-level income needs care where a licensee reports one agreement across several marks. A league licensee may sell the marks of every team in the league under one agreement; a collegiate licensee may report dozens of institutions on one statement; an entertainment licensee may carry several films, series or characters at different rates. If the reports carry property detail, the analysis uses it. If they do not, the absence is itself a finding: a licensor that cannot see which properties are selling cannot tell a growing program from one riding a single hit, and cannot allocate income to rights holders where its own obligations require that.

Territory income has a currency question attached, which the section on currency treats at length. For the income cut itself the rule is simple: convert every territory's figures to the reporting currency at one fixed rate when comparing periods, and report the effect of actual exchange rates separately. Otherwise a strengthening home currency makes every foreign territory look as if it were shrinking.

The income cut is descriptive, and the temptation is to stop there because it produces a satisfying chart. Its job is to set the context for the ratios that follow. A licensee whose royalties grew is not necessarily performing; one whose royalties fell is not necessarily failing. The ratios say whether the movement came from volume, price, mix, rates or deductions, and only that decomposition supports a conversation.

Effective royalty rate against the contract rate

The effective royalty rate is the single most informative ratio in the analysis, provided its two halves are stated. In this guide effective rate on gross = earned royalty ÷ gross sales, measured per agreement per period. The glossary entry on the effective royalty rate explains why the denominator must always be named; gross is the right default for a licensor because it is the figure the deductions have not yet touched, so the gap between the effective rate on gross and the contract rate captures everything that happened between the invoice and the royalty.

Under an agreement with one flat rate, the gap decomposes exactly. Earned royalty is the contract rate times gross sales less deductions, so the effective rate on gross equals the contract rate times one minus the deduction rate. At a 12% contract rate and an 8% deduction rate, the effective rate on gross should be 12% × (1 − 0.08) = 11.04%. If the measured figure matches, the whole gap is deductions, and the deduction rate is the measure to watch. If the measured figure is lower, something other than deductions is lowering the royalty: a line priced below its contract rate, a tier applied in the wrong period, or a channel carrying a lower rate than the report claims.

Under an agreement with several rates, which is normal in licensed apparel, a second ratio is needed: the rate on net, earned royalty divided by net sales. If every line is priced at its contract rate, the rate on net equals the expected blended rate, which is the weighted average of the contract rates with each line's net sales as its weight. Compute the expected blended rate from the licensee's own reported mix and the rate card, and compare. A rate on net below the expected blended rate means at least one line is priced below contract, whatever the deductions did. The difference times net sales is what the mispricing is worth.

Across the portfolio the same measure becomes a ranking. List every agreement's effective rate on gross beside its contract rate and its deduction rate, and sort by the gap between the measured effective rate and the figure the contract rate and deduction rate predict. Agreements where the two agree are explained. Agreements where they disagree are the first queries of the quarter. This ranking is useful precisely because it normalizes away the differences between agreements: a 9.68% effective rate is neither good nor bad until it is set beside an 11% contract rate and a 12% deduction rate, at which point it is exactly what the contract produces and the question moves to the deductions.

Keep the earned and paid versions apart. A paid effective rate, which adds the guarantee shortfall to the numerator and is measured per contract year, answers a different question: what the agreement costs the licensee in total. An agreement whose paid effective rate sits far above its contract rate is one where sales do not support the guarantee, which matters for the renewal even though every report is correct. Mixing the paid and earned versions in one trend line produces a jump at every guarantee settlement that has nothing to do with reporting.

Three legitimate movements need to be recognized before any change is treated as a signal. Mix moves the blended rate when a period is heavy in a lower-rate category or channel, with every line correct. Tiers move it when cumulative contract-year sales cross a threshold. And channels with deliberately different rates, such as an FOB rate set to protect per-unit income on lower invoice prices, move the effective rate on gross in a direction that can look wrong while the royalty per unit is unchanged; the section on channel mix works through that case.

MG coverage, to date and projected

Minimum guarantee coverage measures how far an agreement's earned royalties have gone toward its guarantee for the measurement period. MG coverage to date = earned royalty to date in the measurement period ÷ minimum guarantee for that period. Projected MG coverage replaces the numerator with earned royalty to date plus projected earned royalty for the periods that remain. Both are computed per agreement, against the guarantee as the agreement defines it: the amount, the measurement period, whether it steps up by contract year, and whether anything other than earned royalty counts toward it.

Coverage to date is only meaningful against an expectation for the point in the year. After three quarters of an evenly spread year, an agreement would be at 75% coverage if it is on track to exactly meet its guarantee. Licensed products are rarely evenly spread. Football and basketball apparel peak with their seasons and postseasons, collegiate programs with back-to-school and bowl season, entertainment programs with release windows, and gifting categories with the holiday quarter, so the honest expectation uses each licensee's own seasonal shape: the share of last contract year's earned royalty that had arrived by the same point. Write the expected coverage down beside the measured coverage, with the assumption that produced it.

Projected coverage needs a projection of the remaining periods, and the projection method should be the simplest one the data supports. One method takes the same remaining periods from the prior contract year and adjusts them by the year-to-date trend. Another takes the licensee's own forecast where the agreement requires one, which some agreements do for exactly this reason. A third, for a first-year agreement with no history, uses the business plan the licensee presented when the guarantee was negotiated, discounted by however far the year to date has run behind it. Every projection is a planning figure, not a fact, and the analysis should carry the method and its inputs beside the number so the next quarter can test it.

Coverage is a per-agreement measure and must not be averaged. A portfolio whose total earned royalties exceed its total guarantees can still be carrying large shortfalls, because one licensee's surplus does nothing for another's guarantee: separate agreements with separate licensees do not cross-collateralize, because each guarantee is one licensee's own obligation. The worked example shows a portfolio projected at close to full aggregate coverage that is nevertheless building several times as much shortfall as the aggregate gap suggests.

Two structural details change the arithmetic and should be on the agreement record. Where an advance is credited against the guarantee, the advance does not change coverage, because coverage measures earned royalty, but it does change the settlement, because the cash already received counts toward the floor. And where a guarantee is measured over the full term rather than each contract year, coverage to date is measured against the term guarantee, and an apparently low figure early in a long term may be on pace.

Shortfall exposure across the portfolio

Shortfall exposure is the guarantee top-up a measurement period is projected to require. For each agreement, shortfall exposure = the greater of zero and (minimum guarantee − projected earned royalty). It is a planning measure: the figure a licensor expects to invoice or receive at the measurement boundary if the projection holds. Summed across agreements, it is the part of the year's projected royalty income that depends on guarantees rather than on sales.

The sum has to be built agreement by agreement, from the positive shortfalls only. Netting a surplus on one agreement against a shortfall on another produces the aggregate gap between total guarantees and total projected earnings, which is a meaningless number for a portfolio of separate agreements. Where an agreement genuinely cross-collateralizes properties or contract years, as some multi-property agreements with a single licensee do, the pooled basket is the unit of measurement and the shortfall is computed for the pool. The guide on cross-collateralization covers how pooled baskets work; for the portfolio analysis, the rule is to measure exposure at whatever level the agreement settles at, and nowhere else.

Shortfall exposure says something about income quality that coverage alone does not. A program whose projected income is mostly earned royalty is being paid by the market. A program whose income increasingly includes shortfall payments is being paid by the guarantees, which hold only as long as the licensees can and will pay them and renew on similar terms. Track the shortfall share of paid royalty income, shortfall exposure divided by projected paid royalty, by licensee and for the program, and watch its trend across contract years.

Exposure also has a sensitivity. An agreement projected at 97% coverage has a small shortfall that a modestly stronger final quarter would remove; an agreement projected at 70% will settle a large shortfall unless something unusual happens. Agreements near the floor deserve watch status, because they flip between earning and paying their guarantee as projections move, and because below the floor a reporting understatement changes what the shortfall payment is called rather than what the licensor receives. Above the floor, the same understatement is a loss. An agreement hovering near its guarantee is therefore where reporting quality matters most and where it is easiest to overlook.

Finally, exposure is a conversation starter rather than an invoice. A shortfall projected in the third quarter can still be influenced: a licensee can add a program, open a channel the agreement permits, or bring forward an order. A shortfall discovered at the measurement boundary can only be settled. A licensor that shares projected positions with its licensees during the year tends to get fewer disputed settlements and earlier signals about renewals, because a licensee that is going to struggle with its guarantee usually knows it before the licensor does.

Deduction rate: the gross-to-net measure across licensees

The deduction rate is total deductions divided by gross sales, and it is the portfolio-level measure of how much of each invoice never reaches the royalty base. Compute it per agreement, by deduction type and by category, every period. The guide on gross-to-net royalty deductions works through each line of the deduction stack and the rule that only contractually permitted deductions may be taken; the analysis question is different. It is not whether each deduction is permitted, which the report review tests, but whether the pattern of deductions makes sense over time and across licensees.

Start with each licensee's own history, because deduction rates vary legitimately between agreements. Different net sales definitions permit different deductions; direct-to-consumer volume brings consumer returns, wholesale volume brings markdown allowances and chargebacks, and categories with high shipping cost per invoice dollar carry more freight where freight is permitted. A deduction rate means something only against the agreement's terms and the licensee's own past. No general benchmark replaces that comparison, and any range quoted without a named source should be treated as illustrative.

Then decompose by type. A rising deduction rate is either a business change or a reporting change, and the decomposition usually says which. Returns rising as a licensee shifts toward its own e-commerce site are consistent with the shift; allowances rising faster than the wholesale sales that generate them are not. A new deduction type appearing without an agreement change is a query on its own. And a deduction type claimed at exactly its cap period after period suggests the licensee is reporting the cap rather than the amount.

At portfolio level, compute the deduction rate on total program gross sales as well, but treat it as a weighted average with all the weaknesses of one. A licensee whose deduction rate jumps by three points can disappear inside a portfolio rate that barely moves if that licensee is a small share of gross sales or if mix moves the other way. The worked example decomposes the portfolio deduction rate into a mix effect and a rate effect to show how this happens; the practical rule is to rank agreements by the change in their own deduction rate and read the portfolio figure last.

The value of a deduction rate change is easy to compute and should be written beside it. The extra deductions are the change in rate times gross sales, and the royalty they removed is that amount times the contract rate on the lines where they were taken. For a licensee near its guarantee, set the figure beside the projected shortfall: a deduction rate increase that pushes an agreement below its guarantee costs the licensor almost nothing this year, because the guarantee holds, and the full amount in any year the agreement is above it.

Reporting timeliness and restatements

Two measures describe how a licensee reports rather than what it reports, and both predict trouble earlier than the financial measures do. Reporting timeliness is the number of days between a report's due date and the date it arrived complete, with complete meaning carrying every schedule and level of detail the reporting clause requires. Track it per report and summarize it per licensee as the share of reports on time and the average days late. A report that arrives on time without its required detail should be counted as late until the detail arrives, or the measure will reward licensees for sending totals first.

Lateness matters beyond the cash timing it causes. A licensee that starts reporting late is often a licensee whose finance function is under strain, whose systems have changed, or whose business is in difficulty, and any of those can precede a missed guarantee or a dispute. A licensee that is consistently a few days late is a process conversation; one whose lateness is growing is a relationship conversation, and should be raised before the guarantee boundary rather than after it.

Restatements are the second measure. Count the reports later restated or corrected through a prior-period adjustment, as a share of reports submitted, and measure the value of the restatements against the royalty originally reported. Separate the restatements that follow the agreement's own mechanics, such as returns attributed back to their original period, from those that correct errors, such as a rate or a category mapping. The first kind is expected under most agreements and should be predictable from the licensee's returns pattern. The second kind is a direct measure of reporting quality.

Direction matters as much as frequency. Restatements that move in both directions are consistent with an imperfect but honest process. Restatements that only ever reduce royalty, especially near guarantee or tier boundaries, deserve attention: a licensee that restates downward after a quarter in which it crossed a tier threshold, or restates a contract year's final quarter after the guarantee settlement, is moving the base where the agreement's thresholds make it most valuable. The pattern is visible only in a dataset that kept the originally reported figures.

Both measures belong in the quarterly review beside the financial ones, because a licensor deciding whether a financial signal is worth a query will weigh it differently for a licensee with a clean reporting record than for one whose reports are late, incomplete and frequently restated.

Royalty per unit and invoiced price drift

Dollar measures hide price. A licensee's royalties can rise because it sold more units, because its prices rose, or because its mix moved toward higher-priced product, and each of those means something different for the program. Two per-unit measures separate them. Royalty per unit = earned royalty ÷ units, and gross sales per unit = gross sales ÷ units, the licensee's average invoiced price to its customers. In retail planning the abbreviation AUR usually means average unit retail, the price at the shelf; on a royalty report the price is the licensee's invoice to its own customer, which for a wholesale licensee is a wholesale price, so this guide uses gross sales per unit to avoid confusing the two.

Both measures need category-level computation. A licensee selling tees, fleece and outerwear has a gross sales per unit that moves whenever its mix moves, so a portfolio of category-level figures is far more informative than one figure per licensee. Within a category, a steady rise in gross sales per unit is usually a price increase, a move up in fabric or decoration, or a shift toward direct-to-consumer sales invoiced at retail; a sharp fall is usually markdowns, off-price disposal, a shift toward FOB programs invoiced below domestic wholesale, or a change in what is counted as a unit.

One signal belongs in every analysis: unit volume flat while revenue rises. It has three common explanations. The licensee raised prices or moved up-market, which can be checked against its price lists and is good news for a percentage royalty. The mix moved toward higher-priced categories, which the category-level figures will show. Or the unit count changed meaning: multi-packs or sets counted as one unit, bundles reported as a single item, units reported net of returns in one period and gross in another. Only the third is a reporting problem, but it is the one that matters most under per-unit royalty terms, because a per-unit royalty paid on an undercounted unit base is understated by exactly the undercount.

The opposite pattern, units rising while gross sales per unit falls, is the classic off-price signal: a season of inventory clearing through closeout accounts at low invoice prices. Whether that is a problem depends on the agreement. Many agreements restrict off-price and closeout sales, price them differently, or set a deemed minimum price, and a report that carries low-price units on ordinary wholesale lines may be avoiding those terms. The category-level price trend is how a licensor sees it without line-level account detail.

Royalty per unit is the measure to trend when channels carry different rates. If an agreement sets a higher FOB rate on lower FOB invoice prices to keep per-unit income level, royalty per unit should stay roughly stable as the mix moves between domestic wholesale and FOB, while gross sales per unit and the effective rate on gross both move. A royalty per unit that falls as FOB volume grows says the FOB rate is not doing what it was set to do, which is a renegotiation point rather than a reporting finding.

Channel mix shift: off-price, direct-to-consumer and FOB

Channel mix is the share of each licensee's net sales in each channel the agreement recognizes: wholesale to approved accounts, the licensee's own e-commerce and stores, marketplaces, team and campus stores, event retail, off-price and closeout accounts, and FOB programs where a retailer imports directly. Compute it per agreement every period, because a mix shift moves the royalty, the effective rate and the deduction rate even when every line on every report is correct.

A short illustrative example shows the size of the effect. Take a licensee with $4,000,000 of net sales in each of two years under an agreement paying 12% on wholesale and 8% on approved off-price sales. In the first year 10% of net sales are off-price: royalty is $3,600,000 × 12% + $400,000 × 8% = $432,000 + $32,000 = $464,000, a rate on net of 11.6%. In the second year 25% are off-price: royalty is $3,000,000 × 12% + $1,000,000 × 8% = $360,000 + $80,000 = $440,000, a rate on net of 11.0%. Net sales are unchanged, every line is priced correctly, and the licensor's income is $24,000 lower. The figures are illustrative; the mechanism is general.

FOB moves the measures in the other direction. Suppose an agreement prices domestic wholesale at 12% and FOB at 16%, and a product invoices at $40.00 domestic and $30.00 FOB. Royalty per unit is $40.00 × 12% = $4.80 domestic and $30.00 × 16% = $4.80 FOB. If a licensee moves 10,000 units from domestic wholesale to FOB, its gross sales fall by 10,000 × $10.00 = $100,000, its royalty is unchanged, and its effective rate on gross rises. Again the figures are illustrative. A rising effective rate is not automatically good news, and a falling one is not automatically a problem; the channel mix says which one the licensor is looking at.

Direct-to-consumer volume has its own profile. Sales invoiced at retail raise gross sales per unit and royalty per unit at a percentage rate, unless the agreement sets a separate direct-to-consumer rate or a deemed wholesale price; consumer returns raise the deduction rate; and markdown allowances fall, because no retailer is receiving them. A licensee growing its own site will show all three movements at once, and the analysis should attribute them to the channel before testing any of them.

Off-price is where mix analysis turns into a grant question. Where an agreement restricts off-price and closeout sales, a rising off-price share is something the licensing team should already have approved. Where the share is rising without approvals on file, or where off-price indicators appear on lines reported as ordinary wholesale, the issue is not the royalty arithmetic but whether the licensee is operating inside the grant. For a brand licensor that issue can outweigh the royalty entirely, because licensed product clearing through off-price channels at volume affects the brand's own pricing position.

Comparing periods honestly: contract year against calendar year

Most of the errors in royalty performance analysis are comparison errors rather than arithmetic errors. The figures are right; they are being compared with the wrong thing. The first rule is to compare the same contract-year-to-date period against the prior contract year, not calendar year against calendar year, wherever the agreement's contract year is not the calendar year.

The reason is that almost everything that drives a royalty is defined on the contract year. The guarantee is measured on it. Rate step-ups usually take effect on its anniversary. Tier thresholds accumulate within it and reset at its start. Annual deduction caps are measured over it. A calendar-year comparison of an agreement whose contract year runs from July to June combines the second half of one contract year with the first half of the next, which may carry a different rate, a different guarantee and a different tier position. A licensee that stepped from 10% to 11% on July 1 will show calendar-year growth in royalties that is partly rate, and the analysis will attribute it to sales unless the rows carry their contract year.

Portfolio roll-ups make the problem harder, because different agreements have different contract years. A licensor with some agreements on calendar contract years, some on July-to-June years tied to a sports season, and some on anniversary years starting whenever the agreement was signed cannot produce one meaningful year-to-date figure for the whole program. The workable approach is to produce two views. The first is a calendar-period view of income, for finance, which reconciles to the cash and the books. The second is a contract-year view of performance, per agreement, in which each agreement is compared with its own prior contract year at the same point. Growth and coverage come from the second view; income totals come from the first.

Seasonality is the second reason for the same-period rule. Licensed products are seasonal enough that comparing a quarter with the quarter before tells the analyst mostly about the calendar: football and basketball seasons, postseason runs, back-to-school, the holiday quarter and the returns that follow it. Compare volumes and income against the same period last contract year. Ratios such as the deduction rate and the effective rate move much less with the season, so for ratios the prior period is a reasonable comparison and a large jump between adjacent quarters deserves an explanation.

Event-driven programs need a third adjustment. A championship, a title run, an anniversary, a film release or a one-time event program produces a spike that does not repeat. Comparing next year with the spike makes a healthy program look as if it collapsed; comparing the spike with the year before makes an ordinary licensee look exceptional. Tag event programs in the dataset, report them separately, and compare the base business with the base business.

Comparing periods at a fixed exchange rate

Wherever royalties are reported or owed in more than one currency, period comparisons mix two things: what the licensees sold and what the exchange rate did. Compare periods at a fixed exchange rate — usually the rate applied to the comparison period, applied to both — and report the effect of actual rates as a separate line. That separation is the whole of constant-currency analysis, and it is what lets a licensing team discuss performance without a currency argument in every conversation.

An illustrative example shows the effect. A territory produced €250,000 of earned royalty in the prior contract year, converted at an average of 1.10 US dollars per euro, which is $275,000. In the current year it produced €262,500, a 5% increase in euro terms, but the average rate fell to 1.04, so the current figure in dollars is $273,000. Reported in dollars, the territory declined by $2,000, or about 0.7%. At the prior year's fixed rate of 1.10, the current year is €262,500 × 1.10 = $288,750 against $275,000, an increase of 5%. The licensees in that territory grew; the currency hid it. All figures are illustrative.

Two different exchange rates are involved in royalty reporting, and they should not be confused. The first is the rate the agreement specifies for converting reported sales into the contract currency, with its source and its date; that rate decides what the licensee owes, and the guide on multi-currency royalty reporting covers it. The second is the analyst's fixed rate for comparison, which decides nothing about what is owed and exists only to make periods comparable. A licensor that reports performance at the contractual conversion rates has measured the obligation correctly and the performance badly.

Currency also affects the guarantee. Where a guarantee is denominated in one currency and royalties are earned in another, coverage moves with the exchange rate even if sales do not, and a projected shortfall can appear or disappear on a currency move. The analysis should compute coverage in the guarantee's own currency, as the agreement measures it, and show separately how much of any change in projected coverage came from exchange rates.

Comparing figures that are still moving: returns, restatements and as-of dates

Royalty figures for a period are not final when the report arrives. Returns attributed back to their original period, prior-period adjustments and restatements all continue to change a period's figures for some time after it closes. A comparison between the current quarter, as first reported, and the same quarter last year, after a year of adjustments, compares a gross-of-returns figure with a net-of-returns one, and will show the current year growing even when it is not.

The fix is to compare at matched maturity. Either compare the current period as first reported against the prior-year period as first reported, using the original versions the dataset kept, or compare both periods at the same age, such as as of ninety days after the period closed. Which one a licensor uses matters less than using it consistently and stating it. The original-version comparison is easier and is usually good enough; the matched-age comparison is better where a licensee's returns are large or slow, as they can be in wholesale apparel and in gifting categories.

Keep the adjustments visible in their own right. A licensee whose periods consistently lose a predictable share of their first-reported royalty to later returns is behaving as its business suggests. A licensee whose adjustments are large, irregular or concentrated around guarantee and tier boundaries is not. Measuring adjustments as a share of first-reported royalty, per licensee, across periods, turns a source of comparison noise into a reporting-quality signal.

The same principle applies to year-end. A contract year's final figures are not known until the last period's adjustments have settled, so the first analysis of a completed year should be labelled preliminary and rerun once the returns window the licensee's history suggests has passed. Renewal conversations built on preliminary figures risk arguing about numbers that will move.

Segmenting the portfolio: concentration

Once the measures are computed per agreement, segmentation turns them into decisions. The first segmentation is by size. Rank licensees by earned royalty for the contract year to date and compute each one's share of program income. The top licensee's share, and the combined share of the top three or five, describe how dependent the program is on a few relationships. A program where one licensee earns more than half of the royalties has a renewal risk that no amount of growth elsewhere offsets in the short term.

Where a single number is useful, the Herfindahl-Hirschman index summarizes concentration: square each licensee's percentage share of income and add the squares. A program with one licensee scores 10,000; a program split evenly among ten scores 1,000. The index is a summary, not a judgment, and for a small program the shares themselves are easier to discuss. It is most useful as a trend: a rising index means the program is becoming more dependent on its largest relationships, whatever the total is doing.

Concentration should be measured along every dimension that carries risk, not only by licensee. By property, it shows a program depending on one team, one character or one franchise. By category, it shows a program that is really a headwear program or a tee program with some other licenses attached. By channel, it shows dependence on one retailer or one marketplace, which matters because the loss of one large account can move several licensees at once. By territory, it shows exposure to one market's currency, regulation and retail health.

Concentration is not automatically bad. A league or a major entertainment property may reasonably build its apparel program around one master licensee, and a brand licensor may prefer one strong partner per category. The point of measuring it is to make the dependence explicit, so that the renewal calendar, the guarantee structure and the attention the licensing team gives each relationship reflect it.

Segmenting the portfolio: guarantee coverage against growth

The second segmentation places every licensee on two axes: projected MG coverage for the current measurement period, and same-period growth in earned royalty against the prior contract year. The two axes are chosen because they answer the two questions a licensing team has to settle for every agreement — is this agreement earning what it was sized to earn, and is it heading in the right direction — and because together they suggest a different next step for each quadrant.

Licensees above their guarantee and growing are the program's core. The conversation is about protecting and extending the relationship: early renewal on terms that reflect the growth, new categories or properties, and a guarantee that steps up with the business rather than staying where it was set. Licensees above their guarantee but declining are still earning their floors, but the trend says that will not last; the conversation is about why, and whether product, distribution or the licensee's own business is the cause.

Licensees below their guarantee but growing are often agreements whose guarantee was set too high for the business they could build in the time available: a first-year license, a new category, a new territory. The conversation is about sizing. A guarantee that the licensee pays every year from its own margin is a guarantee that will be renegotiated or abandoned at renewal, and a licensor that raises the question first usually gets a better outcome than one that waits for the licensee to raise it. Licensees below their guarantee and declining are the hardest quadrant. The guarantee protects this year's income, but the agreement is not working, and the conversation is about whether it can be fixed, restructured or allowed to end on the agreement's own terms.

Add the reporting measures as a third dimension. A licensee below its guarantee whose reports are also late and frequently restated is a different risk from one with the same coverage and a clean record. The first may be heading toward a payment problem as well as a sales problem; the second is probably a sizing conversation between two well-run businesses. The table below sets out the four quadrants with the measures that usually confirm each reading and the conversation each implies.

Segmenting by property, category and channel across licensees

The licensee is not the only unit of analysis. A licensor with several licensees selling the same property, or the same category under different properties, can learn more by turning the portfolio sideways: one property's income across all its licensees, or one category's effective rate across all the licensees that sell it. Cross-licensee comparison is the licensor's one genuinely independent check, because no single licensee can see the others' figures.

Within a category, compare deduction rates, effective rates and price per unit across licensees selling similar product through similar channels. The comparison is loose — licensees differ in customers, price points and channel mix — but a licensee whose deduction rate is twice that of every peer in the same category and channel deserves a closer look at its deduction stack, and one whose price per unit is far below its peers' may be reporting closeout volume as ordinary wholesale. The comparison is a reason to ask, never a basis for a finding by itself.

Across properties, the same cut shows which marks are earning. A league program can compare its team marks by income per licensee and growth; a collegiate licensing office can compare institutions; an entertainment licensor can compare properties against their release and broadcast calendars; a brand licensor can compare categories against the brand's own core business. Properties that are licensed widely but earn little are either under-merchandised or over-licensed, and the distinction is a strategic question that the royalty data can raise but not answer.

Across channels, the cut shows where the program's product is actually sold. A licensor that sees its program shifting toward off-price accounts across several licensees at once is seeing a market condition, such as excess inventory across a category after a weak season, rather than a reporting problem at any one licensee. The right response to a market condition is a program-level decision about channel policy, not a series of individual queries.

Signals that warrant a conversation or an audit

Every measure in this guide can produce a signal, and nearly every signal has an innocent explanation. Licensees change prices, enter channels, lose accounts, restate in good faith and make process errors that a query resolves. A signal is a reason to ask a question, not proof of underpayment, and the first response to almost every signal in the table below is a query with the arithmetic attached.

Three signals carry the most weight because they point at the two places an understatement can come from, the rate and the base. An effective rate on gross drifting below the contract rate times one minus the deduction rate means something other than deductions is lowering the royalty; the rate on net against the expected blended rate says whether it is a pricing problem. A deduction rate rising period after period while staying inside its cap means the base is shrinking, and the decomposition by type says whether the business explains it. And unit volume flat while gross sales rise, or units rising while royalty per unit falls, means either the business changed or the unit base did, and the category-level price trend says which.

Escalation toward the agreement's audit clause is justified when queries stop resolving: the answer does not arrive, arrives without support, or the same signal recurs after it was answered. It is also justified where a signal suggests that sales are missing rather than mispriced, because no amount of internal consistency on a report can reveal a population of sales that never reached it. The audit clause sets the lookback, the notice, the frequency limit, who may conduct the audit and who bears its cost when an underpayment is found; each of those terms is contract-specific, and the licensing team should know them before raising the possibility. The guide on common royalty audit findings describes what audits tend to find from the licensee's side.

A portfolio view makes audit selection better in two ways. It ranks candidates by estimated exposure, which can be built from the reports themselves: the rate gap on the affected net sales, the deduction excess at the line's rate, the unit undercount at the per-unit rate, period by period. And it shows which signals are licensee-specific and which are program-wide, so that an audit is pointed at a licensee whose figures diverge from its peers rather than at one that is behaving exactly like everyone else in a difficult market.

Worked example: the portfolio and its data

The example that follows runs a small portfolio through every measure in this guide. All figures — rates, guarantees, sales, units, deductions and projections — are illustrative, chosen so the arithmetic can be followed. They are not benchmarks, not category norms and not drawn from any licensor, licensee or agreement. The portfolio has been built so that each licensee shows a different pattern, which real portfolios seldom arrange so neatly; the point is the method.

The licensor runs an apparel program for one sports property through three licensees, each under its own agreement. Licensee A holds fan apparel — tees and fleece — at 12% of net sales with a $600,000 minimum guarantee. Licensee B holds headwear at 11% of net sales with a $300,000 guarantee. Licensee C holds replica jerseys at 14% of net sales with a $400,000 guarantee. All three agreements use a calendar contract year, report quarterly, measure the guarantee at contract-year end with any shortfall payable after the year closes, carry no advance and do not cross-collateralize. Each agreement's royalty is a flat rate on net sales, so every report can be priced with one multiplication.

The review takes place after the third-quarter reports have arrived. The table below shows the contract year to date, the first three quarters, as reported and as restated. Net sales are gross sales less deductions; earned royalty is net sales times the contract rate; coverage to date is earned royalty divided by the full-year guarantee. Licensee A: $5,000,000 − $400,000 = $4,600,000 of net sales, and $4,600,000 × 12% = $552,000 earned. Licensee B: $2,400,000 − $288,000 = $2,112,000, and $2,112,000 × 11% = $232,320. Licensee C: $1,500,000 − $90,000 = $1,410,000, and $1,410,000 × 14% = $197,400. The program total is $552,000 + $232,320 + $197,400 = $981,720 of earned royalty on $8,900,000 of gross sales.

The same three quarters of the prior contract year provide the comparison. Licensee A reported $4,500,000 of gross sales with $360,000 of deductions, so $4,140,000 of net sales and $496,800 of earned royalty. Licensee B reported $2,400,000 of gross sales with $216,000 of deductions, so $2,184,000 of net sales and $240,240 earned. Licensee C reported $1,200,000 of gross sales with $72,000 of deductions, so $1,128,000 of net sales and $157,920 earned. The prior-year program total is $496,800 + $240,240 + $157,920 = $894,960 on $8,100,000 of gross sales.

Two reporting facts complete the picture. Licensee A's three reports arrived on time and complete. Licensee B's second-quarter report arrived 12 days late and its third-quarter report 9 days late. Licensee C restated its first-quarter report once, before the second quarter closed, to add a jersey style that had been left off the original submission; the figures in the table are as restated.

Worked example: income, rates and what the portfolio average hides

Start with income. Program earned royalty rose from $894,960 to $981,720, an increase of $86,760 or about 9.7%. Licensee A contributed $552,000 − $496,800 = $55,200, an increase of about 11.1%. Licensee C contributed $197,400 − $157,920 = $39,480, an increase of exactly 25%. Licensee B fell by $240,240 − $232,320 = $7,920, about 3.3%. Shares of current program income are about 56.2% for A ($552,000 ÷ $981,720), 23.7% for B and 20.1% for C, so one licensee earns more than half the program's royalties, and the renewal of A's agreement is the single largest income event on the licensor's calendar.

Now the effective rates. For A, $552,000 ÷ $5,000,000 = 11.04%, and the flat-rate prediction is 12% × (1 − 0.08) = 11.04%; the two agree, so the whole gap between A's contract rate and its effective rate is deductions, and A's deduction rate of 8.00% is unchanged from the prior year's $360,000 ÷ $4,500,000. For C, $197,400 ÷ $1,500,000 = 13.16%, and 14% × (1 − 0.06) = 13.16%; again they agree, with C's deduction rate at 6.00% in both years. For B, $232,320 ÷ $2,400,000 = 9.68%, and 11% × (1 − 0.12) = 9.68%; they agree as well. Every line is priced at contract. B's effective rate has nevertheless fallen from $240,240 ÷ $2,400,000 = 10.01% to 9.68%, and the entire fall is the deduction rate, which rose from 9.00% to 12.00%.

The portfolio figures show how little of this is visible from the top. Program effective rate on gross was $894,960 ÷ $8,100,000 = 11.05% last year and is $981,720 ÷ $8,900,000 = 11.03% this year: a movement of two hundredths of a point, while one licensee's effective rate fell by a third of a point. The program deduction rate rose from $648,000 ÷ $8,100,000 = 8.00% to $778,000 ÷ $8,900,000 = 8.74%, which looks like a modest program-wide drift and is in fact one licensee.

The deduction rate decomposes cleanly. At last year's deduction rates applied to this year's gross sales, program deductions would have been $400,000 for A, $216,000 for B (9% of $2,400,000) and $90,000 for C, a total of $706,000, or 7.93% of $8,900,000. That is the mix effect: because C, with the lowest deduction rate, grew fastest and B did not grow, mix alone would have lowered the program rate slightly. B's extra three points on $2,400,000 add $72,000, which is 0.81 points of program gross, and 7.93% + 0.81% = 8.74%. The whole increase in the program deduction rate is Licensee B, and mix was working the other way.

B's deductions break down by type in its reports. Last year: returns $96,000 (4.0% of gross), outbound freight $48,000 (2.0%) and markdown allowances $72,000 (3.0%), totalling $216,000. This year: returns $96,000, freight $48,000 and markdown allowances $144,000 (6.0%), totalling $288,000. Returns and freight are flat; allowances have doubled on unchanged gross sales. B's agreement, in this example, permits documented markdown allowances up to 6% of gross sales for the contract year. The allowances are inside the cap and sitting exactly at it — the pattern behind two of the signals in the table above, from one line of one licensee's reports; the quarter-by-quarter allowance figures will show whether it has held every period. The royalty they removed is $72,000 × 11% = $7,920, which is the whole of B's decline.

Worked example: guarantee coverage and shortfall exposure

Projected coverage needs a fourth-quarter assumption for each licensee. The licensor projects fourth-quarter gross sales of $1,500,000 for A, $600,000 for B and $600,000 for C, from each licensee's prior-year fourth quarter adjusted for the year-to-date trend, and applies each licensee's year-to-date deduction rate. The assumptions are illustrative and would be written down beside the projection in a real review.

Licensee A: $1,500,000 less 8% deductions of $120,000 is $1,380,000 of net sales, and $1,380,000 × 12% = $165,600 of fourth-quarter royalty. Projected full-year earned royalty is $552,000 + $165,600 = $717,600 against a $600,000 guarantee: projected coverage of 119.6%, no shortfall. Licensee B: $600,000 less 12% of $72,000 is $528,000, and $528,000 × 11% = $58,080. Projected earned is $232,320 + $58,080 = $290,400 against $300,000: coverage of 96.8% and a projected shortfall of $9,600. Licensee C: $600,000 less 6% of $36,000 is $564,000, and $564,000 × 14% = $78,960. Projected earned is $197,400 + $78,960 = $276,360 against $400,000: coverage of about 69.1% and a projected shortfall of $123,640.

Now the portfolio. Projected program earned royalty is $717,600 + $290,400 + $276,360 = $1,284,360 against total guarantees of $1,300,000. The aggregate gap is $1,300,000 − $1,284,360 = $15,640, and a program report that showed only totals would describe the program as about 98.8% covered and roughly $16,000 short. The real shortfall exposure is $9,600 + $123,640 = $133,240, more than eight times the aggregate gap, because A's $117,600 surplus over its guarantee does nothing for B or C. The arithmetic ties: $133,240 − $117,600 = $15,640.

Projected paid royalty for the year is $717,600 from A, $300,000 from B and $400,000 from C, a total of $1,417,600, which equals projected earned royalty of $1,284,360 plus shortfall exposure of $133,240. About 9.4% of the program's projected income is guarantee rather than sales. For C alone the figure is $123,640 ÷ $400,000, about 30.9%. Paid effective rates on projected full-year gross sales make the same point from the licensee's side: A pays $717,600 ÷ $6,500,000 = 11.04%, B pays $300,000 ÷ $3,000,000 = 10.00%, and C pays $400,000 ÷ $2,100,000, about 19.05%, against a 14% contract rate.

B's position deserves one more calculation. Had B's deduction rate stayed at last year's 9% for the full year, its projected earned royalty would be $3,000,000 × 91% × 11% = $300,300, just above its guarantee. B's allowance increase is what moves it below the floor. Because the guarantee holds, the licensor's projected income from B is $300,000 either way this year, so the allowance increase costs the licensor about $300 this year — and its full royalty effect in any later year in which B's sales carry it above the guarantee. To reach its guarantee at its current deduction rate, B would need fourth-quarter earned royalty of $300,000 − $232,320 = $67,680, which at 11% of net and 12% deductions requires about $699,200 of gross sales, roughly 16.5% above the projection.

Worked example: units, price per unit and growth

The reports carry units, which lets the analysis separate volume from price. Licensee A sold 250,000 units in the first three quarters of both years. Its gross sales per unit rose from $4,500,000 ÷ 250,000 = $18.00 to $5,000,000 ÷ 250,000 = $20.00, and its royalty per unit from $496,800 ÷ 250,000, about $1.99, to $552,000 ÷ 250,000, about $2.21. A's entire 11.1% growth is price or mix; its unit volume is flat.

That is one of the signals in the table above, and it resolves in the way it usually does. A's category detail shows the tee and fleece split roughly unchanged, so the movement is not category mix; A's current wholesale price list, which A supplied with its annual product submissions, shows price increases on carryover styles. Under a percentage royalty on net sales, a price increase raises the licensor's income per unit with no reporting question attached. Had A's agreement carried a per-unit royalty, or had A's units changed definition — multi-packs counted as one unit, say — the same pattern would have required a query about the unit base.

Licensee B sold 160,000 units in both years at $15.00 of gross sales per unit, so its volume and price are both flat; its royalty per unit fell from $240,240 ÷ 160,000, about $1.50, to $232,320 ÷ 160,000, about $1.45, and the whole difference is the allowance increase already traced. Licensee C sold 16,000 units last year and 20,000 this year, a 25% increase, at $75.00 per unit in both years, with royalty per unit of $9.87 in both. C's growth is pure volume, priced and deducted exactly as last year.

So the three licensees grew in three different ways. A grew on price, B shrank on deductions, and C grew on volume from a base too small for its guarantee. The program's 9.7% growth in earned royalty is the sum of those three stories, and none of them is visible in the program total.

Worked example: segmenting the three, and the conversations that follow

Placed on the coverage-against-growth axes, the three licensees fall into three different segments, and each implies a different conversation before the fourth quarter closes.

Licensee A is above its guarantee and growing: projected coverage of 119.6%, earned royalty up 11.1%, on-time reports, an effective rate exactly where its contract rate and deduction rate predict. The conversation is about protecting the relationship and pricing the next term. A's growth has come from price rather than volume, which is worth understanding before the renewal: a licensee that has raised prices on flat units may be near the limit of what its retail accounts will accept. A's agreement is also more than half of program income, so an early renewal conversation is worth having on the licensor's timetable rather than at the end of the term.

Licensee B is just below its guarantee and declining, but its units and price are flat and the decline is entirely one deduction type at exactly its cap, so the quadrant's usual fix-or-exit conversation waits on a query. The query goes with the arithmetic attached: the allowance amounts by quarter, the gross sales they relate to, the cap computation, and a request for the documentation the agreement requires for each allowance. B's late reports in the second and third quarters add weight to the query without proving anything. If the allowances are documented and reflect real markdown support to B's retail accounts after a slow season, the finding is commercial — B's accounts are under pressure — and the conversation becomes one about B's business. If they cannot be documented, they come out of the base, and the agreement's terms on underpayment apply. Either way the licensor wants the answer before B's year ends below its guarantee and the issue is folded into a shortfall settlement.

Licensee C is far below its guarantee and growing fast. C's reports are clean and its growth is genuine volume, but its projected paid effective rate of about 19.05% against a 14% contract rate says that C is paying for a guarantee its business cannot yet support — roughly 45% more full-year gross sales than projected would be needed to earn it. The conversation is about sizing. C may be in the early years of building a jersey program, in which case a guarantee that steps up with the business might serve both parties better than one set at its eventual target. C may also have been over-optimistic when it negotiated, in which case it will know already. Either way, a licensor that raises the guarantee structure before renewal keeps the initiative.

None of the three conversations is an audit. Every report in the example is internally consistent and priced at contract rates; the findings are a deduction pattern to query, a guarantee to resize and a price-led growth story to understand. That is the normal output of a portfolio review, and it is why the review should be run quarterly rather than reserved for audit selection.

How the analysis changes by type of licensing program

The measures are the same in every program; what changes is which ones lead, which comparisons are fair, and what the seasonal and event calendar does to the figures. The sections that follow take the program types licensed apparel analysis meets most often — professional sports leagues and teams, collegiate programs, entertainment and character licensing, and brand licensing — and then the adjacent categories that most apparel programs eventually include. In each case the point is not a different method but a different set of traps.

Professional sports leagues and teams

League programs are seasonal by sport and volatile by result. Football, basketball, baseball, hockey and soccer each have a season, a postseason and a draft or transfer window that drive demand, and a team's championship run can move a quarter's royalties by more than any licensee's performance. Compare the base business with the base business: tag championship, playoff and event programs separately, and compare a team's regular-season royalty with its regular-season royalty last year. A licensor that does not separate them will read every championship year as growth and every following year as decline.

Property-level analysis carries most of the value in a league program, because one licensee usually sells many teams' marks under one agreement. Income by team across licensees shows which marks the market is buying and which licensees are under-merchandising which teams. Player-identified product adds a cooperative-mark question: where product carries both league or team marks and player identity, a second rights holder is owed a royalty too, and the analysis on the licensor's side should check that player-identified categories appear where the product descriptions say they should, at the rate the agreement sets for them.

Channel analysis in league programs has its own pattern. Team stores, venue retail on game days, league and team e-commerce, national sporting goods chains, mass retail and off-price all carry licensed product, often at different rates or under different approval regimes. A licensee whose venue and team-store share is rising is following the team's on-field year; one whose off-price share is rising after a weak season is clearing inventory, which the agreement may restrict. Read the channel mix against the team's results before treating either as a reporting signal.

Replica and authentic jerseys are their own category almost everywhere: high price per unit, a narrow range of styles, demand tied to roster moves, and frequently a higher rate than fan apparel. In the worked example the jersey licensee's royalty per unit is several times the fan apparel licensee's, which means a small change in jersey units moves the program more than a large change in tee units. Price per unit is especially informative here: a falling jersey price per unit with flat units usually means markdowns on last season's roster, which is legitimate, or a channel shift, which may not be.

Collegiate licensing programs

Collegiate programs run on the academic and athletic calendar: back-to-school, football season, conference championships, bowl season and the basketball tournament, with graduation and reunion product in spring. The contract year often follows the academic or fiscal year of the institution rather than the calendar, which makes the contract-year comparison rule especially important. A calendar-year analysis of a collegiate program splits football season across two contract years.

Many collegiate licensees report many institutions on one statement, and many institutions license through a common administrator or consortium. The analysis an institution or its administrator runs is therefore usually property-level: income by institution, by licensee and by category, against the same period last year. The comparisons that matter are within peer groups — institutions of similar size in similar conferences — and over time, because a program's royalty income tracks enrollment, alumni base and athletic results in ways that differ sharply between institutions.

Collegiate programs also carry more small licensees than most, including local and regional businesses with modest volumes and minimal reporting infrastructure. Reporting timeliness and completeness matter more here, because a long tail of small licensees produces a long tail of late, incomplete and restated reports. Segmenting by size first, and applying the full ratio analysis to the licensees large enough to move the program, is a reasonable allocation of effort, provided the small licensees are still tested for completeness and timeliness.

Finally, collegiate product is heavily event-driven. A bowl appearance or a deep tournament run produces a burst of event-marked product with its own approvals and sometimes its own rates. Tag it and analyze it separately, as with league championship programs, or it will distort every comparison for two years: the year it happened, and the year after, when it did not.

Entertainment and character licensing

Entertainment programs are driven by release windows. A film's theatrical release, a series season premiering on a streaming service, a game launch or an anniversary re-release creates demand that rises before the release, peaks around it and decays afterward, and the shape of that curve differs by property. Compare a property against its own release cycle, not against the calendar: a property in the quarter after a major release should be compared with the equivalent post-release quarter of its previous cycle, or with similar properties at the same point in theirs.

Evergreen characters behave differently from release-driven properties, and the analysis should separate them. A classic character with steady demand can be analyzed like a brand, with same-period comparisons and stable ratios. A new release property cannot: its first year is a launch and its second year is a decay, and a guarantee sized on launch-year expectations will often look badly covered in year two without any reporting problem at all. Guarantee coverage for release properties should be read against the release calendar the guarantee was negotiated on.

Entertainment agreements often carry several properties under one licensee, sometimes with different rates and sometimes with pooled or cross-collateralized guarantees. Where a guarantee is pooled across properties, coverage and shortfall exposure are measured for the pool, and the property-level figures are for understanding rather than settlement. Where each property carries its own guarantee, the property-level coverage is the settlement figure, and a strong property does nothing for a weak one.

Style guide and approval compliance matter more in entertainment than in most programs, because character integrity is the asset. Royalty analysis cannot test approval compliance directly, but it can flag product categories, styles or channels that appear on reports and not in the approvals record — a cross-check worth running every quarter for any licensee whose category mix changes.

Brand licensing and brand extensions

Brand licensing programs extend a consumer brand into categories the brand owner does not make itself: an apparel brand licensing footwear, eyewear or home, a lifestyle brand licensing apparel, a heritage or corporate brand licensing merchandise. For a brand licensor, channel and price positioning can matter as much as royalty income, because licensed product sits beside the brand's own product in the market and shapes how consumers see it.

That makes the price-per-unit and channel-mix measures leading measures rather than supporting ones. A licensee whose price per unit is falling and whose off-price share is rising may be meeting its guarantee and still eroding the brand's pricing, and the brand team will care about that before the finance team does. Analyze licensed categories against the brand's own distribution: a licensee concentrated in channels the brand itself avoids is a positioning question, whatever the royalty arithmetic says.

Brand programs also tend to have fewer licensees with broader category grants, which makes concentration high by construction and category-level analysis within each licensee more important than licensee-level comparison. A single licensee holding several categories should be analyzed category by category, with its own rates, deduction profile and price trend per category, because a strong category can hide a weak one inside a licensee total.

Because brand extensions are often new categories for both parties, guarantee sizing is a frequent issue in the early contract years. The coverage-against-growth segmentation applies directly: an extension below its guarantee but growing is a sizing conversation, and one below its guarantee and declining is a question about whether the brand belongs in that category at all.

Adjacent categories: footwear, headwear and home

Most apparel programs eventually license adjacent categories, and each brings its own vocabulary and its own analysis traps. Licensed footwear is analyzed by model year and size run: a footwear licensee's units depend on how fully its size runs and width fits are stocked, its price per unit moves when a model year turns over or when carryover styles are marked down, and broken size runs at the end of a model year produce closeout volume that the agreement may price or restrict differently. Compare model year with model year, and read a falling price per unit against the model-year calendar before treating it as a signal.

Headwear moves with team and fashion cycles, has low price per unit and high unit volume, and tends to carry a larger share of freight relative to the invoice than apparel where freight is a permitted deduction, so its deduction rate is not comparable with a fleece or jersey licensee's. Fitted and adjustable product often sell through different channels at different prices, and a mix shift between them moves price per unit without any change in reporting.

Home and fan gear — blankets, drinkware, décor, furniture and outdoor goods — brings long lead times, container minimums and, for furniture and larger goods, dealer prebooks and retail floor sets that commit volume months ahead of sale. Royalties follow shipments, so a home licensee's quarterly income reflects prebook and container timing more than consumer demand; a quarter-to-quarter swing is often a shipping calendar artifact. Compare contract-year-to-date figures rather than quarters, and read the licensee's prebook position, where the agreement or the relationship provides it, as the leading indicator.

In every adjacent category the same discipline applies: compare each licensee with its own history first, then with peers in the same category and channel, and only then with the program. Ratios that look abnormal against an apparel licensee are often entirely normal for a footwear, headwear or home licensee, and the reverse is also true.

The licensee's mirror view: analyzing royalty cost across licensors

A licensee holding agreements with several licensors runs the same analysis from the other side of the table. The measures transfer almost unchanged — effective rate, coverage, shortfall exposure, deduction rate, price and royalty per unit, channel mix — but the questions change. The licensee is asking which licenses earn their cost: which agreements carry a paid effective rate far above their contract rate because the guarantee is not being earned, which properties grow and which decline, where its own channel mix is moving royalty cost, and which renewals it should approach early to restructure.

Two measures matter more from the licensee's seat. The paid effective rate, guarantee top-ups included, is the true cost of each license, and comparing it across licensors shows which agreements are expensive in practice rather than on paper. And the margin after royalty by license, which only the licensee can compute because only it knows its own cost of goods and cost to serve, shows whether a license that looks healthy by royalty volume is profitable once its royalty, guarantee and compliance cost are included.

This is the side of the analysis Royalty Reporting is built for. Its analytics and insights feature page describes multi-period breakdowns across licensors, schools, customers and product categories, period-over-period deltas (MoM, QoQ, YoY), side-by-side comparison such as DTC versus wholesale within a single licensor, and drill-through from any analytics view to the underlying sales transactions. The method in this guide does not depend on any particular tool: it depends on data at one grain, comparisons that are fair, and measures computed per agreement before they are rolled up.

A quarterly royalty performance review checklist

The checklist below puts the guide in order. It is written to be run every quarter, after the period's reports have been reviewed individually, with each step recorded so the next quarter's review starts where this one finished. Steps one to four build the dataset, five to nine compute the measures, ten to twelve segment the portfolio, and thirteen to fifteen turn the analysis into action.

Where time is short, run every step at reduced depth rather than dropping steps. A coverage projection for every agreement with a guarantee, a ranking of every agreement by the change in its deduction rate and a comparison of each effective rate with its flat-rate prediction take a short time to compute from a clean dataset, and between them they cover the guarantee, the base and the rate.

From analysis to action: renewals, guarantee sizing and audit selection

An analysis that ends in a report is half finished. The point is to change what the licensing team does, and three decisions benefit most from a portfolio view.

Renewals come first. Every renewal conversation should start from the agreement's full performance history: earned royalty by contract year, coverage and shortfall payments, the paid effective rate, deduction and price trends, channel mix and reporting quality. A licensor that walks into a renewal with that history can price the next term on evidence; one that walks in with this year's total can only negotiate on impressions. Schedule the conversation by the renewal calendar and by concentration, starting with the agreements the program depends on most.

Guarantee sizing is the second. A guarantee is a forecast both parties signed, and the history of coverage across a portfolio is the best evidence of how well the licensor's forecasts have worked. If most first-year agreements pay a shortfall, the licensor's first-year guarantees are set too high for the businesses its licensees can build; if most mature agreements cover their guarantees by a wide margin, the guarantees are not protecting much. Neither conclusion needs a benchmark — it needs the licensor's own coverage history, read across agreements and years. The guide on how to set a royalty rate covers how guarantees and advances are sized at negotiation.

Audit selection is the third. A licensor with an audit budget for a limited number of licensees per year should spend it where the analysis says the exposure is: signals that survived queries, patterns that diverge from peers, estimated exposure large enough to matter, and audit clauses whose lookback still reaches the affected periods. The analysis does not replace the audit. It makes sure the audit is pointed at a question the reports raised and could not answer, rather than at the licensee whose turn it happens to be.

Common analysis mistakes

Most of the mistakes in royalty performance analysis come from treating portfolio figures as if they were agreement figures, or from comparing periods that are not comparable. The table below lists the ones this guide has worked through, with the correction for each. Each of them can make a program look healthier than it is, which is why they survive: a mistake that produces good news is rarely questioned.

Where this guide sits

This guide covers the portfolio view. The guides linked below cover the ground on either side of it: how to read one royalty report, line by line, which is the input to everything here; how net sales is built from gross, deduction by deduction; how minimum guarantees and advances differ; how channel and territory decide a rate; how currency conversion works on a statement; how cross-collateralized guarantees are measured; what royalty audits tend to find; how a licensing agency runs validation across a program; and how a licensee forecasts its royalty expense. The glossary entries on the effective royalty rate, the minimum guarantee, the minimum guarantee shortfall and net sales define the terms this guide relies on.