Creator Outreach Vetter: Instagram, TikTok & YouTube

Find the creators worth a closer look before you spend an afternoon opening profiles. This Actor combines public profile and post data into a ranked shortlist with engagement, posting activity, contact routes and outreach research notes.

Instagram, TikTok and YouTube use separate engagement benchmarks. The score helps compare a list; it does not verify audience authenticity, predict sales or establish a creator's availability.

See the result first

Leave both Creators to vet and the optional records field empty, then run. Three fictional creators show the output without scraper runs or Actor event charges. Engagement ratios below are decimal values; 0.0474 means 4.74%.

Fictional creator	Score	Band	Engagement ratio	Posts per week
demo.kitchen	92	FIT	0.0474	2.33
demo.runs	52	MAYBE	0.0195	4.81
DemoWoodwork	14	PASS	0.0759	0.15

These are actual outputs from the built-in fictional demo, not customer results. Open the dataset to inspect every reason, then export the rows as JSON, CSV or Excel.

Cost of a real run

$0.02 per creator whose profile returns data. A 50-creator shortlist costs $1.00 in this Actor's events; repeating it weekly for four weeks costs $4.00. Unreadable profiles return an error row without this event charge. Supplied profile records are also charged when scored.

Live reads run separate Instagram, TikTok or YouTube scrapers on your Apify account, with their own charges and plan requirements. Instagram may need an additional Post Scraper run when the profile supplies fewer posts than requested. Supply creators_records to avoid fresh scraper calls. Check the Pricing tab and your Apify plan for current platform or standby usage costs.

What I look at

Seven things feed the score, and they add up to exactly 100 so the total is always readable.

Signal	Points	What earns them
Engagement against our benchmark	30	Likes plus comments per follower on the recent posts, measured against our benchmark for that follower band. 1.5 times the benchmark or better earns all 30.
Recency	15	The newest post went up in the last week. A month of silence drops this to 7, two months to 3.
Posting cadence	15	Three or more posts a week earns 15. Under one a month earns nothing.
Contact route	15	An email in the bio earns 15, a link in the bio earns 8, a bare bio earns 0.
Sponsored share	10	Some paid posts but under a third of them earns 10. None earns 6. Over 60 percent paid earns 0.
Niche fit	10	How many of your niche keywords appear in the bio and the recent captions. No keywords given means a neutral 5.
Audience size	5	10k to 500k followers earns 5. The very small and the very large earn less.

Anything scoring 60 or more is marked FIT. From 35 to 59 is MAYBE, and the rest is PASS. One override sits on top: an account with no post in the last 90 days is marked PASS whatever it scored, because a dead account with a great history is still a dead account. Every row carries the seven reasons that fired with the numbers behind them, so you can see why a creator scored what it did rather than taking the total on trust.

The benchmark

The engagement points compare each creator against our benchmark for its follower band. It is a fixed heuristic table, not a published industry benchmark. Every row names the band it was measured against.

Followers	Instagram	TikTok	YouTube
under 10k	4.0 percent	8.0 percent	30 percent
10k to 100k	2.0 percent	6.0 percent	20 percent
100k to 1M	1.2 percent	4.0 percent	12 percent
over 1M	0.8 percent	3.0 percent	8 percent

Instagram and TikTok rates are likes plus comments per post, over followers. YouTube is different because the channel listing carries no likes, so its rate is the median views of the recent videos over subscribers.

How to run it

Put the handles in, one per line. A profile URL works on its own. A bare handle needs a platform in front of it: ig:handle, tt:handle, yt:handle. Choose how many recent posts to read, add niche keywords if you have them, and run it. Then open the Overview table and sort on the fit score.

Leave the handles empty and you get three demo rows, one per platform, scored by the same rubric I use on real ones, so you can see the shape of the output before you spend anything.

Watching a list over time

Every run saves the follower count of each creator under the watch key you give it. Run the same list again under the same key and each row carries the change since last time, as a number and as a percentage. Use one key per list. The first run under a new key has nothing to compare against, so those fields stay empty.

What comes back

One row per creator. It holds the profile facts and the engagement rate with the benchmark it was measured against. Alongside those sit the cadence and the days since the last post, the sponsored count, the contact route, the niche hits and the follower delta. Underneath sit the rubric reasons in full, plus up to three pitch angles. A pitch angle is one sentence written from that row's own numbers:

Their bio lists hello@demokitchen.example, so the pitch can go by email rather than a DM that may never be opened.

6 of the last 12 posts carry sponsorship markers; check those posts before approaching them.

They publish about 4.5 posts a week, across the recent posts read.

I also write a summary record with how many creators landed in each band, how many carry an email, and the ten highest scores with their first pitch angle.

Honest limits

• The engagement rate describes the posts you asked me to read, not the account's history. Ask for twelve and the number describes those twelve. Pinned posts are skipped when enough unpinned ones remain, because they are usually old best-performers.
• YouTube dates arrive as text like "3 weeks ago", so the recency and cadence for a channel are approximate. The row says so.
• YouTube carries no likes or comments at the channel level, so its engagement is a view rate, and its sponsored share is read from titles only.
• Sponsored posts are found by their markers: #ad, paid partnership, a discount code, the sponsored flag where the platform sets one. A deal with no marker is missed.
• The benchmark is ours. It is a working table for ranking a list, not a published industry figure.
• Nothing here is a judgement about the creator. It describes what the public profile showed on the day I read it.

Use it from an AI agent

Vetting can happen inside an agent conversation, since I made this Actor double as an MCP server. Its endpoint is https://fractionalhqforyou--creator-outreach-vetter.apify.actor/mcp and the one tool, vet_creators, takes the same creator handles and returns the vetted rows. The per-event price is the same as a console run. Clients that only speak to the Apify MCP server can load apify/actors-mcp-server and name fractionalhqforyou/creator-outreach-vetter as the Actor.

Help with your first run

Open an issue on this Actor with the run ID and the result you expected. Remove API keys and private records before sharing. Built by Fractional HQ, independent design and automation.