Running Multi-Country Research Projects Without Losing Control of Quotas, Translations or Timelines

Multi-country research projects multiply everything: languages, quotas, suppliers, time zones, holidays and the number of things that can quietly go wrong. A study that would be routine in one market becomes an operations exercise when it runs in eight. This playbook walks through the phases of an international fieldwork project, from first feasibility check to the combined data file, with the practical detail that keeps a multi-country survey on schedule.

Start with the analysis, not the countries

The single most useful question to ask at the start is: how will the results be compared? If the client wants to compare brand health across markets, the questionnaire, sample definitions and quota logic must be equivalent across countries. If each country will be read on its own, there is more room for local adaptation. Agreeing this early prevents the common situation where data arrives from every market and cannot be combined in the way the client expected.

Write down three things in the project brief:

• The unit of comparison. Country totals, specific segments, or both.
• The definition of the target audience in each market. "Adults who buy the category" may need a different screener in different places.
• What may vary by country. Brand lists, price bands, regional quotas and similar local items.

Phase one: country feasibility and costing

Feasibility in a multi-country survey is never one number. Each market has its own audience size, its own online reach and its own incidence for the screener, and these can differ widely between countries for the same target. Check feasibility per country, per cell, before you commit to quotas or timings in the proposal.

Questions to answer for every country

1. Is the target audience reachable online in this market, or does part of it need another mode?
2. What incidence is realistic for the screener as written in this language and context?
3. How many completes can be delivered in the field window, and is a single supplier enough?
4. Are there public holidays, festivals or election periods in the field window?
5. What local regulations or norms affect the questions, incentives or data handling?

The answers shape both the costing and the timeline. A country with low incidence and limited supply may need a longer field period, a blended sample from more than one supplier, or a conversation with the client about the quota. Our guide to research bid management covers how to present those trade-offs in the proposal.

Audience reach online80

Feasibility at target incidence65

Translation ready and verified90

Supplier coverage70

Field calendar clear of holidays55

A simple readiness score like the illustrative one above, filled in per country, makes it easy to see which markets are likely to cause trouble before launch. The numbers are judgements, not measurements, but they force the conversation.

Phase two: translation and localisation

Translation and localisation are where multi-country research projects most often lose time. The questionnaire is finalised late, translations are rushed, and errors appear only once data arrives.

Separate translation from localisation

Translation renders the source wording into the target language. Localisation adapts content to the market: currencies and price bands, brand lists, regional breaks, education and income categories, date formats and examples that make sense locally. Keep a localisation sheet per country listing every item that changes, who approved it and why.

Build verification into the plan

Back-translation used to be the default check, but it has known limits. In a Pew Research Center article on how it translates survey questions, the Center describes a team approach in which a local agency translates, a separate firm verifies the translation line by line, disagreements are discussed and adjudicated, and translations are pretested before fielding. Commercial projects rarely have the same budget, but the principles scale down well:

• Flag idioms and ambiguous terms in the source questionnaire before translation starts.
• Have someone other than the translator review the translated version against the source.
• Resolve disagreements with reference to what the question is meant to measure.
• Test the scripted survey in each language, on the devices respondents will use.
• Record the decisions so later waves stay consistent.

1. Lock the sourceFreeze the master questionnaire before translation and log any later change.
2. Prepare notesExplain the intent of tricky questions and terms for translators.
3. Translate and verifyTranslate, then have an independent reviewer check against the source.
4. LocaliseApply the country localisation sheet for brands, prices, regions and categories.
5. Script and testCheck each language version in the live script on mobile and desktop.
6. Soft launchOpen each market with a small batch and review the first completes.

If your team scripts the survey in-house, keep every language inside one script rather than separate copies, so logic changes apply everywhere at once. Our post on survey programming logic covers structuring routing so it survives translation.

Phase three: country quotas and supplier allocation

Country quotas are usually a nested structure: country totals at the top, then cells such as age, gender, region or user type within each country. The key decisions are:

Keep every country's quotas in the same project, not in separate trackers. The survey quota management guide explains how to structure nested cells so progress is visible for each country and for the study as a whole.

Phase four: running international fieldwork

International fieldwork rarely moves at the same pace everywhere. Some markets fill in days; others need the full field window and extra supply. A few habits make the difference.

Stagger the launch

Open one or two markets first, ideally ones where the team can read the language. Check routing, translations and redirects on real completes, then open the remaining countries. A problem found in the first market costs one fix instead of eight.

Monitor by country, then by cell

Daily monitoring should start with completes by country against plan, then drill into cells and suppliers. Watch termination and drop-out by country, because a spike in one language version often points to a translation or localisation issue rather than a sample problem. The live fieldwork monitoring routine applies here, run per market.

Respect local calendars and time zones

Plan sends around local working days, weekends and holidays. A quota that looks stuck may simply be waiting for the start of the local day.

Keep one status view for the client

Clients on multi-country studies usually have stakeholders in several markets, each asking about their own country. Send one consolidated status note at a fixed time each day, with every market on the same page: completes against target, cells at risk and actions taken. It saves the field team from answering the same question in eight separate email threads, and it keeps local stakeholders from making requests that conflict with the overall design.

Decide escalation rules in advance

Agree with the client which changes the field team may make on its own, such as rebalancing supplier traffic within a country, and which need sign-off, such as relaxing a country quota or extending field in one market. Write the rules into the project plan so decisions are quick when a hard market falls behind.

• Country feasibility checked per cell
• Translations verified and localisation sheet approved
• Every language tested on mobile and desktop
• Redirects tested for every supplier in every country
• Staggered launch plan agreed with the client
• Daily report shows each country against plan

Phase five: combining and closing the data

The last phase is where the early decisions pay off. Check that every country's file uses the same variable names, codes and labels; recode localised items such as brand lists and regions back to the master structure; and confirm that quality checks were applied consistently across markets. Then reconcile each supplier in each country separately before invoicing, as described in sample supplier reconciliation.

Service standards such as ISO 20252 set requirements for how research providers run their work, and a well-kept multi-country project file, with feasibility notes, translation decisions, quota changes and reconciliations, is the kind of record that supports them.

Where Sampledge fits

Sampledge was shaped by Frequent Research's experience of running fieldwork across many markets. SampAudience lets you define an audience, configure it, check feasibility and cost, then purchase, launch and track, which is a practical way to test country-by-country assumptions before committing to a design. For agency-run studies, SampTrek keeps each country's quotas, suppliers and fieldwork status in one project, so the combined picture and the country detail come from the same record.

To try it on your next international study, create a workspace or talk to our team about your markets and audiences.