Scope: what this audit covers and what it does not
The question this audit answers is whether marketing in the business's established markets is run so that every edition says what is true, campaigns reach every market in step and markets can be compared fairly, and which weakness should be fixed first. It covers markets that are already live: who decides what between the centre and each market, the shared calendar, campaign rollouts and creative adaptation, localisation resources and suppliers, editions kept in step with the source, prices shown in each market, channels, budgets and distributors, and reporting and the yearly decision on each market.
It does not re-teach the work, which sits in the international marketing operations hub and the localisation and strategy guides it links. Running established markets week to week is the multi-market marketing operations playbook, with a gate per phase; this scorecard rates whether that running works. Readiness for a new market belongs to the international market entry playbook and the new market launch checklist, and one symptom in one market to the international marketing troubleshooting collection.
Locale signals such as hreflang and each edition's indexing are scored by the technical SEO audit, support in each market's languages and hours by the customer service audit, and the evidence behind budget decisions in general by the marketing measurement audit. Legal and tax questions are out of scope. Every scorecard sits in the audit library.
Before you score: access and evidence
Ask for read access to every edition of the site, each market's product feed and ad accounts, the localisation files and the consolidated report with its source data; without the editions and feeds the gate on editions stays provisional. Fix an evidence windowThe fixed period of data an audit scores, agreed before scoring, long enough to cover a full cycle for the business and kept the same at every rescore. that includes at least one campaign rolled out across markets and one change to the source content, and keep the same window for every rescore.
Collect:
- the written decision rights or brand charterA short document every market and agency receives, recording what the centre fixes, what markets may adapt and the route for requesting an exception. between the centre and each market, and the decision log;
- the shared calendar with lead times, freezes and each market's local moments;
- the master briefs, adaptation kits and local plans of recent cross-market campaigns;
- each language's termbase, translation memory and style guide, and the latest supplier reviews;
- the source change log with its triage, and the latest drift check of every edition;
- prices as shown on pages, feeds, ads and emails in each market, with the exchange-rate rule;
- budgets and channel plans by market, distributor agreements and the consolidated report with its data notes.
Before scoring, pick a handful of recent source changes, ideally a price, a claim and a retired page, and follow each into every edition by hand. What you find there is the fastest test of the gate.
Criteria and scoring
Rate each row missing (absent or broken), partial (present but incomplete, inconsistent or out of date) or solid (matches the "what good looks like" column). The scale is your judgement against that column, not a number. Score a row across every live market: a practice that holds in some markets and not others is partial. Rows marked gate decide the band on their own when they score missing, because without them an edition can mislead customers or markets cannot be compared. A row that does not apply is marked not applicable rather than missing.
Decision rights and planning
| Criterion | What good looks like | Score guidance |
|---|---|---|
| Decision rights written | One role is accountable for each repeated decision shared by the centre and the markets, and market owners and agencies all hold the brand charter and a way to request an exception | Missing: decided by whoever asks first / Partial: a charter, but no exception route / Solid: one owner per decision, charter shared |
| Centre and market split fits the team | The team model keeps centrally what variation weakens and locally what uniformity weakens, with the interfaces to sales, product and finance defined | Missing: the split by accident / Partial: split, interfaces undefined / Solid: deliberate, interfaces defined |
| One calendar across markets | One published calendar, filterable by market and reviewed with market owners, holds global and local dates, lead times counted back from them, release times and freezes | Missing: a calendar per market / Partial: one calendar, local moments missing / Solid: one, filtered, reviewed |
Decision rights follow the comparison of standardisation and adaptation, which writes the split into a brand charter with an exception route, and a RACI matrix gives each decision its one accountable role. The team model is judged against the comparison of centralised and decentralised operations, and the calendar against the tutorial on building a marketing calendar for several markets.
Rollouts and creative
| Criterion | What good looks like | Score guidance |
|---|---|---|
| Master brief for every cross-market campaign | Every cross-market campaign begins with a master brief separating the fixed parts from the open ones, plus an adaptation kit, a single timeline and one approver in each market | Missing: translated from the home version / Partial: a brief without kit or approvers / Solid: brief, kit, timeline, approvers |
| Idea kept, execution adapted | Creative keeps the idea and adapts its execution by market, with transcreation where the copy carries the persuasion and local sign-off before release | Missing: word-for-word translation / Partial: adapted, never signed off locally / Solid: idea kept, signed off |
| Content localised by priority | What is localised first is decided by each piece's value and risk in the market, not by what is easiest to translate | Missing: whatever is easiest / Partial: priorities never revisited / Solid: by value and risk |
Rollouts follow the tutorial on rolling out a campaign across markets and creative the guide to adapting creative for different markets, with TranscreationRewriting marketing copy for another language or market from the original brief, so the intended effect survives even when the words change. explained in the guide to translation and transcreation. Both send every claim to whoever owns compliance in each market, with the UK's Advertising Standards Authority as the guides' general direction, never legal advice. Priority follows the tutorial on deciding which content to localise first.
Localisation resources
| Criterion | What good looks like | Score guidance |
|---|---|---|
| Termbase and translation memory owned | Approved terms and sentences for each language sit in files the business owns, approved by a reviewer, and every correction is fed back into them | Missing: held by suppliers / Partial: owned, corrections not fed back / Solid: owned, approved, updated |
| Style guide per language | Each target language has a style guide written with the in-country reviewer on top of the house style and kept alive by corrections | Missing: none / Partial: written once, never updated / Solid: per language, kept alive |
| Suppliers matched and reviewed | Suppliers are matched to the label on each asset, chosen with paid test pieces, given the brief and resources on every job and reviewed on quality and time | Missing: one supplier for everything / Partial: matched, never reviewed / Solid: matched, briefed, reviewed |
| Copy checked by someone who reads it | Every edition that sells is reviewed in the market, and a central approver who does not read the language runs the checks that remain | Missing: published unread / Partial: some editions only / Solid: every selling edition |
The TermbaseA record of the approved translation of each product, feature and brand term per language, with terms not to translate, definitions and rejected alternatives. and Translation MemoryA record, kept by the business, of each source sentence or segment and the translation a reviewer approved for it, so repeated wording comes back the same way. follow the tutorial on building a termbase and translation memory, style guides the tutorial on writing a localisation style guide and suppliers the guide to choosing and managing translation suppliers, which asks where Machine TranslationTranslation produced by software or a language model, now fluent enough for first drafts but used with human post-editing applied by content type. sits in a supplier's process and, citing the GDPR, keeps personal data out of files sent for translation. In-country reviewThe check of translated or adapted content by a native speaker who knows the market and category, who signs off the language before publication. and the checks a central approver can still run follow the guide to reviewing localised copy in a language you do not speak, and each job starts from the localisation brief template.
Editions in step
| Criterion | What good looks like | Score guidance |
|---|---|---|
| Source changes reach every edition (gate) | Each source change is traced to the editions it touches and triaged by risk, so prices, claims, legal and safety lines change everywhere at once | Missing: an edition still states an old price, claim or safety line / Partial: risky changes traced, some editions lag / Solid: traced, triaged, changed together |
| Drift checks owned | A named owner checks every edition against its source on a rhythm, and pages retired in the source leave every edition together | Missing: nobody checks / Partial: checked, retired pages linger / Solid: owned, on a rhythm, retired together |
Both rows follow the guide to keeping localised content in step with the source. Drawing on the hreflang guide's reading of Google Search Central's documentation on localized versions, it takes a page retired in the source out of every edition's hreflang set at once; whether the locale signals themselves are right is left to the technical SEO audit.
Prices, channels and partners
| Criterion | What good looks like | Score guidance |
|---|---|---|
| One price everywhere in each market | Each market's price is shown the same way on every page, feed, ad and email, follows a written rule when exchange rates move and can be explained when customers compare editions | Missing: prices disagree within a market / Partial: agree, but no exchange-rate rule / Solid: one price, a written rule |
| Channels chosen by market | Each market's channel mix comes from its own media landscape and funds the market by its role, not by the home market's habits or last quarter's efficiency | Missing: the home mix exported / Partial: mapped, funded by habit / Solid: by landscape and role |
| Budget moved by written rules | Money moves between markets only by rules written before the period starts and approved by the role they name | Missing: moved on request / Partial: rules unwritten or unapproved / Solid: written first, approved |
| Distributor marketing agreed | Where a distributor sells, its marketing duties, the kit it uses, a shared plan and the ownership of accounts and customer data are agreed in writing | Missing: nothing written / Partial: schedule, but data ownership unclear / Solid: all written down |
Prices follow the guide to setting and showing prices across markets, which notes, from the site's pricing guides, that misleading prices are enforced against in the UK by the Competition and Markets Authority and the Advertising Standards Authority's codes cover price claims in ads, and leaves each market's own rules to local advice. Channels and budget rules follow the guide to choosing media and channels by market, and each DistributorA partner that buys a brand's product and sells it on in a market through its own shops, sales force and online channels, taking on much of the marketing. the guide to marketing through a local distributor.
Reporting and decisions
| Criterion | What good looks like | Score guidance |
|---|---|---|
| Markets reported on shared definitions (gate) | Every market is measured on the same definitions and currency, time zone and period rules, in one consolidated report with each owner's commentary and data notes | Missing: each market its own definitions, or currencies added together / Partial: shared definitions, no data notes / Solid: shared, consolidated, annotated |
| A recorded decision on every market | Each market has a recorded continue, adapt, hand-over, pause or close decision that the report and the decision log can justify | Missing: markets simply continue / Partial: decided, never recorded / Solid: recorded, explainable |
The gate follows the guide to measuring and reporting across markets, which notes that consent rules differ by region, the GDPR among them, so raw conversion counts are not comparable between markets and the data notes record each consent change. Market decisions follow the guide to pausing or closing a market, with each one kept in the Decision LogA running record, one row per decision, of what changed and why, what was expected, who decided and when the result will be checked..
How to score and prioritise
Score the two gates first. If any edition states an old price, a withdrawn claim or a stale safety line, correct it in every edition before scoring further; if markets are compared on different definitions, mark every budget and market decision row provisional.
Then rank the remaining weaknesses in this order:
- Whatever lets markets drift apart: unwritten decision rights, separate calendars and campaigns translated rather than rolled out from a master brief.
- The resources that keep quality in every language: termbase, translation memory, style guides, supplier reviews and review in the market.
- Price display, channel and budget rules, distributors and the yearly decision on each market, which all depend on the fair report the gates protect.
Within each group, rank by the customers a weakness reaches: a gap in the largest selling edition outranks the same gap in an edition few people see.
Reading the result: what to do with each band
Red means a gate scored missing. Hold rollouts and budget moves between markets, bring every edition back in step with its source and rebuild the report on shared definitions before deciding anything else. Rescore on the same evidence window.
Amber means the gates hold at least partially but several other rows are partial or missing, the usual outcome. Work a short ordered list, one area at a time, and confirm each fix in every market rather than the one that raised it.
Green means the gates are solid and most rows are too. Let the multi-market operations playbook carry the rhythm, and rescore each quarter, since editions drift from the first edit after launch.
From findings to a prioritised fix list
Write each partial or missing row as a finding in plain words: what you saw, in which markets, and why it matters to customers or to comparison. Then build one fix list with a line per fix: the finding, the fix, the owner, the guide that explains it and the rescore date, ordered by the priorities above rather than by ease.
Group the lines by who does the work: decision rights and the calendar with the central marketing lead, editions and localisation resources with the localisation owner, prices with whoever owns pricing, channels and budgets with each market owner, and the report with the analyst who builds it. Show the band, the gate findings and the first fixes to whoever owns international marketing, and keep the full scorecard for the rescore.
Illustrative example
A lampshade maker audits marketing in its four markets
Grivoltash Lampshades, an invented lampshade maker, sells in its home market and 3 others in 3 languages. It audits the 6 months that include its autumn campaign. Following source changes into each edition shows a fire-safety claim withdrawn 11 weeks earlier still live on 7 product pages of one edition, so the editions gate scores missing. One market's feed prices a shade at 64 while its product page shows 69, so the price row scores missing. The monthly report adds revenue in 3 currencies into a single total, so the reporting gate scores missing. Nobody owns drift checks, budget moves on request and no market has a recorded decision, so those rows score missing too. The business sells direct everywhere, so the distributor row is not applicable. Of the 18 rows, 1 is not applicable, 6 score missing, 7 partial and 4 solid. The band is red. The fix list puts the claim's removal from every edition and a source change log with risk triage first, the report rebuilt on shared definitions with one reporting currency second, the feed price and a price check third and a drift check owner fourth. A quarter later both gates score partial or better, 2 rows remain missing and the band moves to amber.
Common mistakes
- Scoring the home edition and assuming the rest. Follow source changes into every edition; drift hides where nobody reads the language.
- Treating translation as rollout. A campaign translated from the home version has no master brief, kit or local approver.
- Letting suppliers hold the termbase. Approved terms the business does not own leave with the supplier.
- Adding currencies together. A total across currencies without a written rule compares nothing.
- Moving budget to last quarter's cheapest market. Fund each market by its role, on rules written in advance.
- Leaving markets to run on. A market with no recorded decision is a decision nobody took.
