Build a repeatable trend-monitoring workflow
Last updated
Maintained by Datapika · We publish this product under the openclawai account on Apify. Product reference
Use Datapika’s Reddit scraper and supported social-video actors to collect repeatable observations, store each run with a timestamp, and compare matching queries over time. This is a workflow assembled from separate products, not a ready-made cross-platform report. The former 30-Day Trend Intel actor was unavailable when checked on September 4, 2026; do not build a new integration against that endpoint.
Start with a question you can measure
Choose a small, consistent set of topics, communities or public URLs. Record the exact search filters, collection time, actor version and output limits for each run. A changing query or sample size can look like a trend even when the underlying interest has not changed.
- Use the same query and filters when comparing dates.
- Keep platform-specific engagement metrics separate.
- Record failures, missing fields and truncated results alongside successful rows.
Collect each source with its documented input
Use the Reddit actor for supported post and comment modes. Use the TikTok, Douyin and Bilibili actor only for its documented platform and input modes. These are separate requests and datasets; this workflow does not promise YouTube coverage or automatic coverage of every topic.
- Inspect each actor’s input schema before constructing requests.
- Keep source URLs or native identifiers for deduplication.
- Do not treat absent records as proof that a topic has disappeared.
Compare observations, not incomparable totals
Store a dated dataset for every run, then aggregate within the same source and sampling method. For example, compare the number of collected posts mentioning a topic within the same subreddit and time window. Sampling limits, deleted content and source changes affect the interpretation.
- Report the time window and number of observations.
- Do not add Reddit scores to TikTok view counts as if they were the same unit.
- Label any AI-written summary as an interpretation and retain links to its supporting records.
Budget and validate before scheduling
A monitoring workflow incurs the charges of its selected actors and the Apify platform. Headline per-result prices do not necessarily include start fees, optional events or all platform costs. Test a small input, inspect actual output and charges, and only then set the schedule.
- No live benchmark or paid collection was performed for this guide.
- Confirm source permissions, platform terms and your data-retention requirements.
- Do not assume an empty run or retry is free.
Sources you can inspect
Product documentation and implementation references. These references describe supported options and billing, not independent proof of performance. Confirm current details before running; examples and estimates are not guarantees.
How to do it
- 1.Define one topic, source, time window and output limit.
- 2.Open the Reddit actor input schema and configure a small collection.
- 3.If social-video data is needed, configure a separate supported actor input.
- 4.Run the requests with your Apify account, then inspect the returned fields and actual charges.
- 5.Save timestamps, inputs and source identifiers; repeat the same method and compare results.
Questions, answered
Does Datapika currently offer the old 30-Day Trend Intel actor?
That actor’s public endpoint returned 404 on September 4, 2026 and is not in the current reviewed catalog. This guide uses separate available actors rather than that endpoint.
Is this a complete measure of online interest?
No. It measures the records returned by your selected queries, sources and limits. It does not measure all platform activity or prove a population-wide trend.
Can an AI agent summarize the results?
An agent can read the datasets through the supported API or MCP workflow. Keep source links, timestamps and sampling limitations with the summary, and verify important conclusions against the underlying records.