Skip to content
Faisal Mahmud TusherDigital marketing, SEO & analytics
with Faisal Tusher

Interactive Tools › SEO Traffic Estimator

SEO Traffic Estimator for International Markets

Using the SEO traffic estimator for a new market: volume researched in the local language and a position a new edition can hold, to compare markets.
Tutorialintermediate7 min readUpdated

Why a new market changes the estimator's inputs

The default page assumes the keyword, the results page and the site already exist in one market the team knows. Entering a new market removes all three. The guide to market entry and prioritisation warns that demand read in the local language is easy to get wrong without noticing: a category name rendered word for word can show little Search Volume because people there say something else, or plenty for a different need. The estimator multiplies whatever volume it is given, so a dictionary error passes straight through to the clicks.

Position changes too. At home the team can look at where its pages already rank; in a new market the edition has no history, and the sites holding the results may be local retailers, marketplaces or publishers the home team has never competed with. The tutorial on keyword research for a new market looks at each head term's results page (SERP) with the market's own region and language settings, and that reading is what sets a believable position.

The conversion rate has no local history at all. An edition whose prices, payment, delivery and proof are not yet set for the market can convert well below home, so the home figure is an upper bound. With three inputs this uncertain, the entry guide uses the estimator's result to rank markets against each other rather than to predict their traffic.

The inputs for an international market

Calculator inputOn the default pageFor an international marketWhere to find it
Monthly search volumeVolume for the keywordVolume for the cluster from research in the market's own language and country settings, never home terms translatedKeyword research by a native-speaking researcher with tools set to the market
Ranking positionExpected ranking positionA position a new edition can plausibly hold after reading the market's own results page for the head termThe market's results pages and who ranks on them
Conversion rateSite conversion rateThe home site's organic rate for the same page type, treated as an upper bound until the edition has its ownHome analytics for the page type

The volume row is the one to derive with care. Brief a native-speaking researcher with the needs each home cluster serves rather than its words, build seeds from how the market already talks, expand them in keyword tools configured for the market, and cluster the results in that language. The research tutorial keeps translations only as candidates to test and drops any whose results page is about something else. Sum the cluster's volume as the market expresses it, which may be one home cluster split in two or two collapsed into one.

For the conversion rate, the diagnosis of an international site that is not converting sets the edition beside the home site one page type at a time, which is why the input borrows the home rate for that page type and labels it a ceiling.

Entering the market inputs step by step

  1. Commission research in the market's language

    Hand a native speaker the needs, not the words, and collect volume per cluster from tools set to the market. Keep the translated home term's figure beside it as a check, not as an input.

  2. Read the results page and choose a cautious position

    Search the head term as a local searcher would and note who ranks, which formats appear and how established they are. Choose a position a new edition could hold, and apply the same caution to every market compared.

  3. Borrow the home rate as a ceiling

    Take the home site's organic conversion rate for the page type that would own the cluster and enter it, labelled as an upper bound.

  4. Run every candidate on the same assumptions

    Enter each market, or each piece of content being considered for localisation, with the same rules for volume, position and rate, and record the outputs side by side with the evidence behind each input.

A worked market comparison with illustrative inputs

Illustrative example — not a real result

Glustex Audio compares two markets for its headphones

Glustex Audio, an invented headphone brand, is weighing two new markets. In the first, the translated head term for its main category shows 900 searches a month; a native-speaking researcher finds 5,400 under the word people there actually use. The results page shows specialist audio retailers and review sites, so Glustex enters a cautious position 6, and its home organic conversion rate of 1.2% for category pages as a ceiling. The estimator assumes a 5% click-through rate and returns about 270 clicks and about 3 orders a month. Built on the translated term, the same market would have shown about 45 clicks and under one order, and been written off.

In the second market, research finds 8,100 searches for the cluster, but its results page is held by large local electronics retailers and a marketplace, so the cautious position is 9: about 219 clicks and about 2.6 orders at the same ceiling. On one set of assumptions the smaller market ranks first, and Glustex puts it in the first wave.

Prefilled with the example's inputs: Monthly search volume 5,400 · Ranking position 6 · Conversion rate 1.2% → Assumed CTR 5% · Estimated monthly clicks 270 · Estimated conversions 3.

All figures above belong to the invented brand. Swap the volume in the embedded estimator for the translated figure to see how far a dictionary-built estimate falls.

Interpreting estimated clicks and conversions for a new market

Read the outputs as a comparison first. The entry guide treats the estimator as a way to compare markets, not a forecast, and its scoring model needs evidence written beside each score. Estimated clicks and conversions from several markets, or several pieces of content, built on one set of rules for volume, position and rate, give a fair ranking even when every absolute figure is uncertain. Two estimates built on different rules cannot be ranked at all.

Then apply the localisation rule. The tutorial on deciding which content to localise first adds local demand to each candidate piece and moves a piece that nobody in the market searches for, and that customers do not need in order to buy or be served, to a later wave, whatever it earns at home. An estimate near zero for a well-loved home guide is a finding, not an error: the market does not ask that question, or asks it in a form the home page does not answer. Compare the Search Intent labels as well, since the research tutorial finds that a need people buy against at home can be one they are still learning about where the category is younger.

Read a low estimate carefully before acting on it. The diagnosis of a new market not gaining traction traces one common failure to demand estimated from translated keywords, and its fix is research in the language before the plan is redrawn. If an estimate rests on a translated list, rebuild it from research before the market is written off.

The estimated conversions output inherits the ceiling, so read it as the most the market could deliver at that position, not an expected figure. The assumed click-through rate is the estimator's illustrative curve, not the market's own; results pages crowded with marketplaces, ads or local features will give a different share.

After launch, the same troubleshooting page verifies the research by filtering Search Console to the market's country and checking the queries the edition is found for against the research sheet. Where the edition shares a language with another, its pages are declared as alternates with hreflang, which the hreflang guide covers; if the market's searchers reach the wrong edition, fix that before reading the clicks. Sibling pages apply the estimator to keyword prioritisation within one market and to site migrations.

Pitfalls when market numbers go into the calculator

  • Translating home keywords for the volume input. A dictionary term can show a fraction of the real demand, or demand for another need; the estimate is only as honest as the research behind it.
  • Assuming the home position in a new market. The edition starts with no history against competitors the home team has never met. Set the position from the market's own results page.
  • Treating the home conversion rate as the market's. Until prices, payment, delivery and proof are local, the home CVR is a ceiling; label the conversions output accordingly.
  • Reading the illustrative click curve as the market's own. The curve is a demonstration assumption, the same for every market; real shares depend on what each results page shows.
  • Comparing markets built on different assumptions. One market at a hopeful position and another at a cautious one ranks the assumptions, not the markets.

Frequently asked questions

Can translated keyword volumes give a first estimate for a new market?

Only as candidates to test. A word-for-word translation may show little demand because people in the market use another word, a borrowed term or a brand name, or it may show demand for something else entirely. An estimate built on it can write off a market that has real demand or favour one that does not. Replace it with research by a native speaker using tools set to the market.

Why treat the home conversion rate as a ceiling rather than the market's rate?

Because a new edition starts without what makes the home site convert: local prices, familiar payment and delivery terms, reviews from people like the visitor and copy written for the market. Until those are in place and the edition has its own data, the home rate for the same page type is the best case, and the estimate built on it is the most the market could deliver.

How can the estimate be checked once the edition is live?

Filter Search Console by the market's country and the edition's pages, then set the search terms that bring the edition impressions beside the research sheet. Queries that match the sheet support the research; pages found mainly for terms the market does not type point back to demand read from translated keywords.

Terms used on this page

Tools to try

Related