> ## Documentation Index
> Fetch the complete documentation index at: https://docs.adriel.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Data reference

> Metrics, breakdowns, refresh cadence, and limits for the Google Trends data source in Adriel.

## Introduction

Google Trends measures the **relative popularity** of search terms over time and across locations. It returns a normalized index from 0 to 100 rather than absolute search counts — 100 represents the peak popularity of the term in the selected timeframe and location, and other values scale relative to that peak. This connector fetches Trends data through an Adriel-operated AWS Lambda proxy and stores the resulting series in a per-asset cache for reporting in Adriel widgets.

As a research-style connector, Google Trends is best used to compare search interest across keywords, track seasonality, and detect emerging topics — not to estimate absolute search volume.

To connect this data source, see [How to connect Google Trends to Adriel](/data-sources/g-n/google-trends/how-to-connect).

## Data refresh strategy

### Architecture data

This connector does not import a campaign hierarchy. Each Adriel asset represents one Trends query configuration — the set of keywords, timeframe, location, category, search type, and timezone. Any change to the configuration creates a distinct asset.

### Reports data

Adriel re-fetches Trends data on a scheduled cadence via the Lambda proxy and rewrites the per-asset cache on each run.

**Refresh schedule.** Twice daily — approximately 06:00 and 16:00 UTC.

**Refresh strategy.** Atomic full replace — the cached series is rewritten in full on each successful run; partial or in-place updates are not used.

## Metrics

<Note>
  **How to read the columns**

  **Data type** uses this vocabulary: Number, Currency, Percentage, Ratio, Duration, Date, Text, URL, Array, Boolean.

  **API Key** in code style refers to the field name in the underlying `pytrends` payload.
</Note>

### Interest over time

| Metric             | Description                                                                                            | Data type | API Key              |
| ------------------ | ------------------------------------------------------------------------------------------------------ | --------- | -------------------- |
| Interest over time | Normalized search-interest index from 0 to 100 for the keyword in the selected location and timeframe. | Number    | `interest_over_time` |

### By region

| Metric             | Description                                                                                                                                         | Data type | API Key              |
| ------------------ | --------------------------------------------------------------------------------------------------------------------------------------------------- | --------- | -------------------- |
| Interest by region | Normalized search-interest index from 0 to 100 for the keyword across the locations in the query, scaled relative to the region with peak interest. | Number    | `interest_by_region` |

### Related queries

| Metric                   | Description                                                                   | Data type | API Key                  |
| ------------------------ | ----------------------------------------------------------------------------- | --------- | ------------------------ |
| Related queries (top)    | The most-searched related queries for the keyword within the timeframe.       | Array     | `related_queries.top`    |
| Related queries (rising) | Related queries whose search interest is rising fastest within the timeframe. | Array     | `related_queries.rising` |

## Breakdowns

### Query configuration

| Breakdown   | Description                                                                              | API Key   |
| ----------- | ---------------------------------------------------------------------------------------- | --------- |
| Keyword     | Group by keyword — up to 5 keywords per query (Trends limit).                            | `kw_list` |
| Location    | Country or sub-region from the supported location list.                                  | `geo`     |
| Category    | Trends category filter (e.g., Beauty & Fitness, Business & Industrial).                  | `cat`     |
| Search type | Google surface — Web Search, Image Search, News Search, YouTube Search, Google Shopping. | `gprop`   |

### Time & date grouping

<Note>
  Adriel groups Trends results into these standard views for reporting alongside other connectors.
</Note>

| Breakdown           | Description                                                         |
| ------------------- | ------------------------------------------------------------------- |
| Auto Time Breakdown | Automatically picks the best granularity for the date range.        |
| Daily               | Split by day (when the query timeframe supports daily granularity). |
| Weekly              | Split by ISO week.                                                  |
| Monthly             | Split by calendar month.                                            |

## Limitations

* **Normalized index, not search volume.** Values are scaled 0–100 relative to peak popularity within the query window; absolute counts are not exposed ([Google Trends Help — FAQ about Google Trends data](https://support.google.com/trends/answer/4365533)).
* **Re-normalization across timeframes.** The same keyword returns different index values depending on the requested timeframe, because Trends re-normalizes each query ([Google Trends Help](https://support.google.com/trends/answer/4365533)).
* **Up to 5 keywords per query.** The connector fetches up to 5 keywords per comparison query.
* **Granularity depends on timeframe.** Trends selects the returned granularity automatically — daily for short windows, weekly or monthly for longer ones. Adriel time breakdowns are constrained to what the source returned.
* **Rate-limited unofficial source.** Data is fetched via the unofficial `pytrends` library through an Adriel-operated Lambda proxy; heavy or repeated queries may be rate-limited by Google Trends.
* **Low-volume keywords may return empty.** Google Trends suppresses data for terms below its internal search-volume threshold ([Google Trends Help](https://support.google.com/trends/answer/4365533)).

## API references

* [Google Trends](https://trends.google.com/trends/)
* [pytrends library (unofficial)](https://github.com/GeneralMills/pytrends)

## See also

* [How to connect Google Trends](/data-sources/g-n/google-trends/how-to-connect) (paired how-to)
* [Google Keyword Planner data reference](/data-sources/g-n/google-keyword-planner/data-reference) — for search-volume estimates and bid ranges
* [Google Search Console data reference](/data-sources/g-n/google-search-console/data-reference) — for actual search-impression and click data on owned properties