> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://docs.beatapi.io/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.beatapi.io/_mcp/server.

# Which X market should I choose for Japan or the UK, and what can I compare?

> Prepare a multi-market editorial review with explicit country inputs, actual snapshot times and comparable evidence instead of global or author-location assumptions.

Call `twitter.web.fetch_trending` once per market: `country:"Japan"` for Japan and `country:"UnitedKingdom"` for the UK. If you omit `country`, the action returns United States trends. Save the collection time of each snapshot, then compare the two markets' topics and possible editorial angles. When both lists show the same trend name, read the underlying posts to confirm that both markets are discussing the same event.

## Ask three questions before comparing markets

| Your question                                                  | What to do                                                         | How to explain the result                                                                                           |
| -------------------------------------------------------------- | ------------------------------------------------------------------ | ------------------------------------------------------------------------------------------------------------------- |
| What is being discussed in each market today?                  | Request each market separately and keep each actual query time     | Two observed trend snapshots, not simultaneous complete topic statistics.                                           |
| Can one angle work for a topic appearing in both?              | Compare the original names, then confirm the meaning through posts | Identical names may concern different events; list positions are not cross-country heat scores.                     |
| Does appearing in Japan's list make matching authors Japanese? | Inspect the actual context of the search results                   | The discovery market does not establish author location; `twitter.web.fetch_search_timeline` has no country filter. |

When a trend list is empty, first separate a successful empty response from a failed request: check the `country` input and the response status. A successful empty list does not show that nobody in the market is discussing anything. Investigate an error before sending a series of paid requests for other countries. When the list contains a name you do not recognize, mark the name as unresolved; do not accept a background story invented by an AI as fact.

Hand over the country, the snapshot time, the original trend name, the source posts and the reason for each follow-up, exclude or unverified decision. The follow-up, exclude and unverified decisions are your editorial judgments; the trend list does not supply them. The [country-trend guide](/guides/x-country-trends) covers local preparation and the unexecuted search plan. The sections below give the actual inputs, files and budget calculation.

## Use Japan. Omitting country selects the US, not worldwide

Request Japan with `twitter.web.fetch_trending` and `{"country":"Japan"}`. The default `country` is UnitedStates. `twitter.web.fetch_trending` takes a country name string; JP, region, a city ID or a worldwide switch do not work. The free Inspect description lists names including Japan, Germany and UnitedKingdom. Check the current input in Inspect before running a request; do not copy another product's country list.

## What should a Japan–UK editorial comparison deliver?

1. Save a Japan snapshot and a UnitedKingdom snapshot, each with the snapshot's actual collection time.
2. Run `prepare_trends.py` with `trend-inputs.json` from the [country-trend guide](/guides/x-country-trends). The script produces `trends.csv`, `selected-searches.json`, `coverage.json` and evidence copies of the original files.
3. In each row, keep the country input, `response_position`, the original trend name, the returned context and description, the request ID and the source hash.

`response_position` is the item's order in one response, not a guaranteed official rank. Do not label the two snapshots as simultaneous top-30 rankings: each snapshot has its own collection time and returned count. When a trend has no description, leave the description blank, and write your editorial explanation in a separate brief so that nobody mistakes your explanation for platform text.

To pick topics for follow-up, identify each topic by country and `response_position`. `prepare_trends.py` JSON-encodes the actual trend name as the search `keyword`, so names that contain quotes stay intact. `selected-searches.json` is an unexecuted plan and makes no calls.

## Do posts matching a Japan trend come from Japan?

No. `twitter.web.fetch_search_timeline` has no country input. Japan identifies the market where you discovered the topic, not the authors' nationality, residence or posting location. Read each post's text and `lang` to choose relevant evidence; `lang` does not show location either. In the brief, record the discovery market, the original topic, the snapshot time, the supporting posts and your editorial judgment as separate fields. Verify news and sensitive events; a trending label does not establish a fact.

## How do I budget four requests?

The recorded free Inspect example prices, which are not live prices, put both `twitter.web.fetch_trending` and `twitter.web.fetch_search_timeline` at `$0.003` per request.

| Plan                                                                          | Successful requests | Estimate |
| ----------------------------------------------------------------------------- | ------------------- | -------- |
| Two country snapshots plus one search page for one selected topic per country | 4                   | `$0.012` |
| Two country snapshots, no search                                              | 2                   | `$0.006` |

Processing saved files locally makes no calls. Each additional topic, search page or refresh is another request; add the request to the budget separately.

The workflow has no recurring collector, so the workflow provides no refresh interval, historical curve, fixed result count or numeric popularity measure. Deliver the actual snapshot first, then decide whether another query is useful.

Free contracts: `data:twitter.web.fetch_trending`, `data:twitter.web.fetch_search_timeline`.