Instagram & TikTok Engagement Rate - Grade Any Creator List

Give it the dataset of an Instagram or TikTok scraper and it works out each creator's engagement rate and grades it from 0 to 100 (A to F) against creators of the same size. Made for influencer marketers, agencies and brands who have a list of creators and need to know who is worth paying.

No login, no cookies, no scraping: it only does the maths on data you already have.

Quick start (3 copy-paste examples)

1. See it work (empty input): run with {}. You get a labelled demo list of 12 synthetic creators (free).

2. Grade the output of a scraper: run apify/instagram-scraper (profile details) or clockworks/tiktok-scraper, then pick its dataset in the "Creator dataset" field. In JSON:

{ "datasetId": "YOUR_DATASET_ID", "platform": "auto" }

3. Paste your own numbers (CSV or JSON):

{ "rows": "username,followers,likes,comments\n@alpha,12000,540,22\n@alpha,12000,610,30\nbeta,250000,2100,15", "includeCampaignReport": true }

Rows of the same username are combined, so one row per post works as well as one row per creator.

Use it with (pipelines)

• apify/instagram-scraper (profiles, with latest posts) -> this Actor. It reads username, followersCount, followsCount and the latestPosts[] list (likesCount, commentsCount, videoViewCount).
• clockworks/tiktok-scraper (profile or hashtag videos) -> this Actor. It reads authorMeta.name, authorMeta.fans and each video's diggCount, commentCount, shareCount, playCount. Flattened column names such as authorMeta.fans work too.
• Any CSV export from a creator database: name the columns username, followers, likes, comments.

Instagram post-only datasets (for example the Instagram Post Scraper) usually have no follower count. Add a followers column or run the profile scraper as well; creators without followers come back as a type: "error" row and are not charged.

Input guide

Field	What it does
Creator dataset	The dataset of the scraper run (picker).
Or paste rows	JSON array, JSON lines or CSV/TSV. Ignored when a dataset is picked.
Platform	auto detects Instagram or TikTok per row.
Add a campaign report row	One summary row for the whole list (separately priced).
Your own benchmarks	Override the default median rates, e.g. {"instagram":{"micro":2.5}}.
Maximum creators	Safety cap, default 1000.

How the grade is calculated (transparent formula)

1. Interactions per post = likes + comments (+ shares on TikTok). Instagram posts with hidden likes (-1) are left out of the average.
2. Engagement rate = average interactions per post / followers x 100. The median post rate and, when views exist, the rate by views are shown next to it.
3. Tier by followers: nano under 10k, micro 10k-100k, mid 100k-500k, macro 500k-1M, mega over 1M.
4. Benchmark = a rule-of-thumb median rate for that platform and tier (Instagram 3.0 / 2.0 / 1.3 / 1.0 / 0.7 percent, TikTok 10 / 7 / 5 / 3.5 / 2.5 percent). vsBenchmark is your creator's rate divided by it.
5. Benchmark score (0-100) from that ratio: 0.5x = 30, 1x = 60, 1.5x = 78, 2x = 88, 3x = 96, 4x or more = 100.
6. Percentile in list = the share of other creators in your list with a lower ratio (needs 3 or more creators).
7. Grade = 75% benchmark score + 25% list percentile (100% benchmark score when the list has fewer than 3 creators). Letters: A 85+, B 70+, C 55+, D 40+, F below.

The benchmarks are rule-of-thumb medians in the range of public industry reports, not numbers measured by this Actor. Replace them with your own in the "benchmarks" field if you have better ones.

Output

Field	Meaning
creator, platform, profileUrl	Who and where.
followers, following	From the data.
postsAnalyzed, postsHiddenLikes, confidence	Sample size (high at 10+ posts, medium at 5-9, low below 5 or when only totals exist).
avgLikes, avgComments, avgShares, avgViews	Averages over the analysed posts.
engagementRatePct, medianPostErPct, engagementByViewsPct	The rates.
tier, benchmarkPct, vsBenchmark	The comparison.
percentileInList, grade, letter, scoreParts	The 0-100 grade and how it was made.
estPostPriceUsdLow / High	Very rough price range for one sponsored post (a rule of thumb scaled by size and engagement, not a quote).
fakeEngagementHints	Plain-text warnings, only for signals the data can show (see limits).
basis	recent-posts, aggregate-fields (avg columns) or lifetime-average (profile totals only, low confidence).

The last row (type: "campaign-report") holds the median rate, grade distribution, top and weakest creators and a total cost range.

Sample row:

{ "type": "creator", "creator": "demo_fitness_anna", "platform": "instagram", "followers": 8200, "postsAnalyzed": 12,
  "avgLikes": 402, "avgComments": 26.9, "engagementRatePct": 5.23, "tier": "nano", "benchmarkPct": 3, "vsBenchmark": 1.74,
  "percentileInList": 91, "grade": 85, "letter": "A", "estPostPriceUsdLow": 150, "estPostPriceUsdHigh": 300,
  "fakeEngagementHints": [], "basis": "recent-posts", "confidence": "high" }

Price

Pay per event: $0.002 per creator graded and $0.01 for the optional campaign report. Creators without enough data and the free demo are never charged. Example: grading 500 creators with the campaign report = 500 x $0.002 + $0.01 = $1.01.

FAQ

Does it scrape Instagram or TikTok? No. It only reads a dataset you already have, so there is no login, proxy or ban risk.

Which fields does it understand? Instagram scraper profile and post output, TikTok scraper video output, and generic columns (username, followers, likes, comments, views, shares). Field names are matched by name; if your scraper uses other names, rename the columns or paste a CSV.

Is a high rate always good? Not always. A rate far above the tier benchmark is shown as a hint to check for giveaways or engagement pods.

Honest limits

• Fake-engagement hints are heuristics, not proof. They appear only when the data has the needed fields (comments vs likes, spread of likes over 6+ posts, following vs followers, rate far above or below the benchmark).
• The scraper's "latest posts" are a small sample (often 12). One viral post can move a small creator's average; viralOutlierPosts flags this.
• Benchmarks and price ranges are rules of thumb. Instagram saves and reach, and TikTok views-based rates, are not part of the grade.
• Tested here against generated data in the documented scraper shapes and pasted CSV; not yet run against a live scraper export.