Google Trends Planner: Seasonality, Regions & Rising Searches

Google Trends data built for people who spend media budgets. Give it a list of keywords; get back one summary row per keyword with the numbers you plan with (peak month, low month, a 12-month seasonality index, year-over-year and 3-month change, top regions, breakout searches), plus flat timeline, region and related-search rows you can pivot in a spreadsheet.

$2.50 per 1,000 keywords analysed. No start fee. Up to 120 timeline, region and related rows per keyword are included (the defaults stay inside that). Keywords Google has no data for are free.

What you can use it for

• Budget pacing and flighting: seasonJan … seasonDec is an index (100 = average month). A category at 160 in November and 70 in February tells you where to weight spend.
• Geo-targeting and bid adjustments: interest by country, state, US metro area (DMA) or city.
• New keywords and negatives: "rising" related searches flag breakout terms (isBreakout) before they show volume in your keyword tools, plus the junk queries to exclude.
• Share of search vs competitors: with Add share of search on, every keyword is also put on one common scale (shareOfSearch, shareOfSearchAvg), even past Google's 5-term comparison limit.
• Shopping and YouTube demand: switch Search type to Google Shopping (the closest proxy for purchase intent) or YouTube.

Input

Field	What it does
keywords	Search terms or Trends topic ids (/m/...). With share of search on, put your main term first: it anchors the scale.
geo	Empty = worldwide; US, GB, TH…; or a subregion such as US-CA.
timeRange	Preset (default: past 5 years, needed for YoY and a stable seasonality index), or set startDate/endDate.
searchProperty	Web (default), Google Shopping, YouTube, Images, News.
category	Optional Trends category id to disambiguate ("apple" in Food vs Computers).
regionResolution	AUTO, COUNTRY, REGION (states), DMA (US metros), CITY.
includeTimeline, timelineGranularity	Timeline rows; AUTO gives monthly rows for ranges over a year (60 rows for 5 years), NATIVE gives every weekly/daily point.
includeRegions, maxRegions	Region rows, highest interest first (default top 25).
includeRelatedQueries, maxRelatedPerKeyword	Top and rising searches (default 10 per list).
maxKeywords	Hard cap on keywords per run; your cost ceiling.

Output

Every row has recordType, keyword, geo, timeRange, searchProperty, category. Use the dataset views (Summary, Timeline, Regions, Related) or filter on recordType:

• SUMMARY (one per keyword): status (OK / NO_DATA / ERROR), dataQuality (GOOD / SPARSE), averageValue, latestValue, peakValue/peakDate, lowValue/lowDate, change3mPct, changeYoYPct, peakMonth, lowMonth, seasonJan…seasonDec, shareOfSearchAvg, topRegions, topQueries, risingQueries, message, and trendsUrl to check it in Google Trends.
• TIMELINE: granularity, date (YYYY-MM when monthly), value (0-100 for this keyword), shareOfSearch, isPartial (the current, still-filling period; excluded from the summary maths).
• REGION: regionResolution, regionCode, regionName, latitude/longitude (cities), value.
• RELATED_QUERY: listType (TOP / RISING), query, value, formattedValue (Breakout, +250%), isBreakout.

Example summary row (US, past 5 years, web search):

{"recordType": "SUMMARY", "keyword": "nike", "status": "OK", "dataQuality": "GOOD",
 "averageValue": 49.6, "latestValue": 40, "peakDate": "2026-04-12", "changeYoYPct": 6.1,
 "change3mPct": -24.5, "peakMonth": "Dec", "lowMonth": "Sep", "seasonNov": 121, "seasonDec": 124,
 "risingQueries": "nike mind 001 (Breakout), nike vomero plus (Breakout), ...",
 "trendsUrl": "https://trends.google.com/trends/explore?q=nike&date=today%205-y&geo=US"}

Honest about failures

• No silent empty datasets. A keyword with too little volume gets a NO_DATA row saying so (not charged). A keyword Google refuses after all retries gets an ERROR row with the reason (not charged). If every keyword fails, the run is marked Failed with the reason in the status message.
• Sparse data is flagged. If fewer than half the periods have data, dataQuality is SPARSE and seasonality/change figures are withheld rather than computed from noise.
• No timeouts that eat your results. About 450 keywords an hour (requests are deliberately paced). The default run timeout is 4 hours; if a run nears its timeout it stops cleanly, saves everything done so far and lists the keywords it did not reach.
• Polite by design. Requests are sequential and spaced out; on HTTP 429 the actor backs off (honouring Retry-After) and retries with a fresh session. A RUN_REPORT record in the key-value store lists requests made, rate-limit hits, failures and duration.

Things to know about Google Trends data

• Values are relative (0-100 within the request), not search volumes. Pair them with a keyword volume tool when you need absolute numbers.
• Region scores are normalised by each region's total searches, so small regions (Wyoming, small US metros) can top a list on modest absolute volume. Use them for bid adjustments, not volume forecasts.
• Related topics are off by default: in October 2026 Google returns empty topic lists. Related queries are unaffected.
• Data comes from Google Trends' public web interface, the same data you see at trends.google.com. No login and no Google account are used.

Pricing

Pay per event:

Event	Price	When
keyword-analyzed	$0.0025	One keyword with data: its summary row plus up to 120 detail rows
extra-row	$0.00001	Each detail row beyond 120 for a keyword (only with NATIVE timelines, all metros, etc.)

1,000 keywords on default settings = $2.50. NO_DATA and ERROR keywords cost nothing. Set a maximum charge per run in Apify; the actor stops cleanly at it.

Measured on Apify (31 Oct 2026): 50 keywords, US, 5 years, all row types, finished 50/50 in 6.6 minutes, recovering on its own from the 3 rate-limit responses it met.