Instagram Competitor Content: AI Winning Posts Finder

Instagram posts that beat their account's median, in your inbox weekly — with why they won.

⚡ First 25 winning posts free · 💵 $0.0032 per winning post · 🤖 ≈ $0.012 per winning post with AI (tokens included) · ⏱️ demo in seconds

Try it in 30 seconds. Click Try it — 18 sample posts from 3 profiles, their 6 winners in seconds, nothing charged. Type competitor profiles, or pick a one-click preset (coffee, fitness, skincare, agency), to search live. Your first 25 winning posts are free; the source's scan runs on your Apify account, capped by "Max source spend". → What a run costs · What the AI adds · Run it weekly

⚡ At a glance

What you get	Only the posts that beat their own profile's median likes + comments, ranked, with hook, format, pillar and an angle to copy
You provide	3–10 competitor Instagram usernames (or profile URLs)
Output	A ranked sheet — Excel, CSV or Google Sheets — plus an e-mail digest of what is new each run
Typical run	10 profiles × 30 posts = 300 scanned → ≈ 36 winning posts (≈ 12 %, estimate until a live run is measured)
Cost of that run	$0.42 rules only · $0.73 with AI analysis ($0.30 source on your account + 36 × $0.0032 / 36 × $0.009 here + ≈ $0.11 tokens)
Free tier	First 25 winning posts per account
Keys / setup	🔑 No API key where your Apify plan includes model access — otherwise your own Gemini / Groq key (free key: 1–3 AI verdicts per run)
Which plan	Free demo and your own posts (list mode): any plan · live search: a plan that runs Store Actors — otherwise the run shows the free sample and hands you the list-mode route
Agencies	Preset Agency — 3 client niches: 6 profiles in one run, ≈ $0.18 scan + winners; a clients mode is dated 2026-10-31
Works with	Schedules, webhooks, Sheets, Slack, Teams, Discord, Make / n8n / Zapier, MCP and AI agents

🎯 What this Actor does

You give it Instagram profiles; it keeps only the winners and charges only for those. For every winning post you get:

• A fair winner test. Each post is measured against its own profile's median likes + comments, so a small account's breakout is not buried: "6.9× this profile's median".
• Why it won. Rule tags on every post — format, caption length, hashtags, opening line (question, number, how-to, call to action), weekday and hour — and, with AI, the hook, pillar, call to action and tone, explained.
• Content ideas from competitors. With AI and a one-line brand description, each winner gets a remix brief: how your brand could adapt the idea. Without AI you still get the rule-based angle to copy.
• Pillars across the winners. Winners are grouped into content pillars, so you see which topics keep winning.
• Only new winners. Posts you already received are never delivered or charged twice; dead or private profiles are reported, and we charge nothing for them (the source's scan runs on your account).

Find top performing Instagram posts

Set the outlier multiple (default 2×), minimum likes, post types and the last N days — free filters.

Instagram content ideas from competitors

Describe your brand in "Your brand (for the remix brief)": each winner then carries a remixBrief, an adaptation of the idea.

🚀 How to use it

1. Click Try it — the form is pre-filled with the free demo on sample posts; it works as is.
2. Type the competitor profiles ("Instagram profiles to analyse") and, if you like, the topic you care about — they replace the sample and start the live search on your Apify account. Leave "AI assessment" on: your first 5 winning posts get live AI verdicts on any plan; all 25 free ones get them with Apify model access or a paid own key.
3. Press Start and open the dataset — winners land best-first; export to CSV or Excel, or wire a webhook.
4. Schedule it weekly (Actions → Schedule) with your address in "E-mail me the new winning posts" — only new winners are delivered and charged.

📦 Already scraping Instagram? Find the winners in your Instagram scraper data

Pick your last scraper run's dataset in Dataset — no second scrape, you pay only for the winners (after every scrape).

📚 Learn more: Step-by-step tutorial — find what Instagram content works for competitors

🆚 Why this instead of running the scraper yourself?

The Fast Instagram Profile Posts Scraper returns raw rows; the median per profile, the sorting and the "why" are left to you.

	Fast Instagram Profile Posts Scraper (instagram-scraper/fast-instagram-post-scraper)	This Actor
Price	$0.001 per post + $0.0005 per run start · 1,289 users / 30 days	from $0.0032 per winning post, first 25 free, no start fee
What you pay for	every post scraped	only winning posts — the rest costs $0 here (the scan is billed on your account, capped)
Same 100 winning posts, all in	≈ $0.83 (830 rows), still unsorted	$1.15 rules only · $2.03 with AI (the same $0.83 scan + 100 × $0.0032 / $0.009 + ≈ $0.30 tokens) — ranked, explained, junk costs $0 here
Billing events	4: row, filtered, restricted_profile, run start	2: basic and AI
Your time per 100 winning posts	≈ 10 h sorting by median and reading captions (estimate, not timed)	0 h — ranked, tagged, explained
Repeat runs	the same posts again next week	only new winners are delivered and charged

vs other Instagram tools

	Price	What it gives you	What is missing
This Actor	$0.0032 per winning post (AI $0.009)	winners vs each profile's median, why they won, remix brief	follower growth, ads
davidbenittah/instagram-competitor-content-report	$0.015 per handle report + $0.005 start · 2 users / 30 days	one report per handle: posts per week, format mix, top 5 posts, hooks	one row per winning post with its multiple, AI why, remix brief, cross-run memory
night111/instagram-competitor-content-monitor	$0.0017 per insight + $0.00004 start · 1 user / 30 days	ranked insights from profile or post records you supply	scanning the profiles for you; a per-profile median
quanmatrix/instagram-content-competitive-gap-intelligence	$0.0035 per result · 1 user / 30 days	content gaps and momentum from datasets or public URLs	winners ranked against each profile's own median, per post
feedminer/instagram-competitor-benchmark	$0.0015 per post + $0.80 start per run · 1 user / 30 days	engagement, cadence, top posts for up to 5 handles	a run under $1 (its start fee alone is $0.80)

Rival prices and users are from their Store listings, checked 2026-10-04 — each unit differs (report, insight, result, post).

💵 Pricing — what a run really costs

Pay per result: one event per winning post, nothing else.

Event	When it is charged	Price
free-tier	your first 25 winning posts on this Actor, in any mode	$0.00
qualified-winning-post-basic	AI off — post fields, baseline, outlier multiple, rule tags	$0.0032
qualified-winning-post-ai	AI on — plus why it won, hook, pillar, call to action, tone, cluster, remix brief	$0.009

Never charged by us: posts below the outlier multiple, posts that fail a filter, your exclude list, posts you already received, dead or private profiles. The source's scan is not ours to waive — in live mode the scraper bills your account per post read, capped by "Max source spend".

How that compares — the Store median for this search is $0.00325; $0.0032 is 0.98× it, but only per qualified winning post — one replaces ≈ 8 raw rows: $0.0004 per raw-row equivalent. All in per 100 winners: $1.15 rules only, $2.03 with AI.

Worked example — 300 scanned → ≈ 36 winners (≈ 12 %, estimate): source $0.30 on your account + 36 × $0.0032 = $0.42, about $0.012 per winning post with rules only. With AI: $0.30 + 36 × $0.009 + ≈ $0.11 tokens = $0.73, about $0.020 per winning post. Your account's first 25 winners are free, so its first run costs $0.30 + 11 × $0.0032 = $0.34.

At volume — an agency watching 10 clients × 5 profiles × 30 posts weekly: 1,500 posts ≈ $1.50 source + ≈ 180 winners × $0.0032 ≈ $2.10 a week (≈ $9 a month) with rules, ≈ $3.70 a week with AI.

AI tier, all in: ≈ $0.012 per winning post ($0.009 here + ≈ $0.003 tokens, an estimate from ≈ 600 tokens in / 200 out until a platform run measures it) — 3.8× the basic price; the source scan comes on top, ≈ $0.008 per winning post.

🤖 What the AI tier adds

The same winning post, AI off — facts, baseline and rule tags:

{"name": "@demo_roastery: Why does your espresso taste sour? ☕", "format": "reel", "outlierMultiple": 6.86,
 "flags": ["viral_outlier", "reel_format", "question_hook", "cta_in_caption"], "ruleScore": 89, "label": "hot",
 "whyItWon": null, "remixBrief": null, "pillarCluster": "#coffeetips"}

AI on (qualified-winning-post-ai, $0.009) — the same post, explained and adapted to your brand (a real AI run of the AI example input on the own-key default model, 2026-10-16):

{"name": "@demo_roastery: Why does your espresso taste sour? ☕", "fitScore": 95, "score": 95, "label": "hot",
 "assessedBy": "ai", "aiModel": "google:gemini-3.5-flash-lite",
 "whyItWon": "The post likely succeeded by pairing a compelling question hook about sour espresso with a quick, actionable fix in a Reel, paired with a clear save call to action.",
 "hookType": "question", "contentPillar": "espresso extraction guide", "cta": "save", "tone": "educational",
 "remixBrief": "Create a quick Reel addressing a common daily habit frustration (like why your morning skincare flakes or your home workout feels ineffective). Open with a direct question, give an instant fix, and prompt viewers to save the guide for later.",
 "pillarCluster": "#coffeetips", "analysisSource": "llm"}

The rule tier gives the pattern; only the AI explains the mechanism and writes the adaptation for your brand — about 6 minutes by hand per post, based on our own estimate of reading the caption, comparing it with the median and writing the why and the idea (not yet timed).

⚙️ Input

The full list of targetFlags values — flags and presets — an unknown name stops the run with the valid ones.

📤 Output

One dataset item per winning post — export as JSON, CSV or Excel, or stream it to a webhook. Views: Shortlist (best first), Score breakdown, Why they won, Remix briefs, Baseline maths, Export.

An example row: API reference.

Output fields

Upsert key for a CRM or Sheet: dedupeKey — sc: plus the post's shortcode, so one post is one row however many runs or profile lists find it; itemId is a hash of the same key.

A Slack / e-mail-ready DIGEST (new and hot counts, what changed since the last run — "vs last run: qualified 41 → 55, hot 5 → 9" — and the top 10 with links) and a client-ready REPORT page sit next to OUTPUT after every run that delivered; forward OUTPUT.reportUrl. OUTPUT is the full run record — see the API reference.

⭐ Found it useful? A review on the Store helps others find it.

🔁 Run it weekly

1. Open the saved task and click Schedule after your first real run — the run saves your search as a task on your account and links it in its status (OUTPUT.scheduleTaskUrl; nothing is scheduled until you click). Put your address in "E-mail me the new winning posts": every run with new winners e-mails you the digest and the trend vs the last run.
2. Leave "Never output the same winning post twice" on: the next run delivers and charges only what is new. The memory is kept per source, not per topic wording or profile list — add a profile and the schedule keeps its memory; for several clients give each its own dedupe store name.
3. Send the digest to Slack, Teams, Google Chat or Discord with a webhook URL; the rows reach Sheets or your CRM through a Make / n8n / Zapier "create or update row" step on the webhook (columns as in the tracker template, upsert on dedupeKey) — no weekly import; a native CRM push is dated in the roadmap.

🎯 Try it for your niche

Pick a niche in Start from a ready niche search at the top of the form and press Start, run it once, then schedule it.

🔌 Integrations, automation and API

After every instagram-scraper/fast-instagram-post-scraper run: its Integrations tab → "Run succeeded → start this Actor" with sourceMode: dataset, datasetId = resource.defaultDatasetId, enableAi: false — rules only.

Run outcomes — what your integration sees

Every outcome with what to do, webhook, Sheets, CRM and AI-agent recipes (free first call {} or sourceMode: list; upsert on dedupeKey): API reference.

Use it from Claude, ChatGPT or any MCP client — add Apify's MCP server with this Actor as a tool:

{"mcpServers": {"apify": {"url": "https://mcp.apify.com/?actors=rich_minds/ig-competitor-winning-posts"}}}

Minimal agent input — every other input keeps its default:

{"profiles": ["bluebottle", "intelligentsia"], "outlierMultiple": 2, "recentDays": 7, "dedupeAcrossRuns": false, "maxQualified": 50, "maxDiscoveryChargeUsd": 0.5}
{"sourceMode": "list", "itemsList": [{"shortcode": "…", "like_count": 900, "comment_count": 80, "user": {"username": "…"}}]}

The second line is the any-plan, no-source-run path.

Which call runs what: {} and the untouched form are the free demo, typed profiles start the live search; every run writes OUTPUT.mode — table in the API reference.

👥 Who is it for?

You are…	You run it to…	Start with
Social media manager tracking 5–50 competitor accounts	see what content worked this week and why	the niche preset closest to you, scheduled weekly
Content agency or brand marketer	brief clients with proven angles and remix ideas	preset Agency, then one task per client with its own dedupe store name
Creator researching a niche	find the formats and hooks that break out	5 profiles, post type reel, outlier multiple 3
AI agent or n8n workflow	get typed winners with scores and reasons	the minimal agent input above

🧠 How the AI works

Every number and claim is grounded in the post the model saw; if it fails the run continues on rule scores.

Full detail — grounding rules, pillars, token use, fallback: API reference.

🔒 Data, compliance and limits

• Public Instagram posts only, read by the source scraper without a login — no private accounts. Follow Instagram's terms and applicable law (GDPR, CCPA) for your use.
• Hidden likes and missing view counts stay null, never estimated; a profile needs 2 loaded posts for a median (fewer than 5 is flagged weak_baseline).
• Cover-image links (thumbnailUrl) expire after a few days.

❓ FAQ

What will my first real search cost? 3 profiles × 30 posts ≈ $0.09 on your account (capped at $0.50 in the form) and ≈ 11 winners, all free — your first 25 winning posts are free.

Which Instagram competitor posts are worth copying? Instagram outlier posts — far above their own profile's median. Run their profiles and read the Why they won view: the hook pattern, format, pillar and posting time that repeat across winners are the answer for your niche.

Is this an Instagram profile posts scraper? Not a raw one — it runs the Fast Instagram Profile Posts Scraper (instagram-scraper/fast-instagram-post-scraper) on your account and adds the baseline, filter, analysis and memory.

Will I be charged for the same post twice? No, with "Never output the same winning post twice" on (default): every charged post is remembered by shortcode in a key-value store on your account. Adding or removing a profile keeps that memory; another source and each dedupe store name keep their own. Each set of tag or post-type filters keeps its own list, but a post already paid for under another search of the same store is delivered as already-charged, never charged again. A post is delivered again only after dedupeTtlDays days (0, the default, never).

List-mode rows

🧩 More Actors from the same developer

Not quite your use case? The same pay-per-qualified-result model, closest to this Actor's buyers first:

• YouTube Outlier Videos Finder — The same winner-versus-baseline test for competitors' YouTube videos.
• Facebook Ad Library Winning Ads Finder — Your competitors' longest-running Facebook ads, broken down.
• Facebook Page Post Tracker — The same weekly winners-and-why digest for the Facebook pages you track.
• Website Change Monitor — Every real change on competitors' pages, in your inbox weekly.

🆘 Support

Something missing or wrong? Open an issue on the Actor's page or message the developer through the Apify Console.

📝 Changelog

One line per published build, newest first — what changed for you: output, status messages, prices.

• 0.1.5 (2026-10-16) — One "Send alerts to" box takes your Slack, Discord, Teams, Google Chat, Telegram or webhook links, and webhooks can now be signed, with any posts that failed to arrive named in the run results (webhook keys are now post / posts); A post you already paid for under another profile list or filter set is delivered again but never charged twice, and AI-checked winners now lead their tier with a "Scored by" column; If the source scraper changes its output, the run now stops with a clear reason and charges nothing, instead of finishing empty; and your own Gemini key works with the model field left empty.
• 0.1.4 (2026-10-04) — a plan that blocks the source scraper now shows the free sample winners and the any-plan next step, not 0 rows.
• 0.1.3 (2026-10-04) — shorter page; webhook and "source partial" outcomes spelled out.
• 0.1.2 (2026-10-04) — every row carries all its fields (empty when Instagram does not report them); an Agency preset.
• 0.1.1 (2026-10-04) — initial release: winners versus each profile's median, rule tags, AI analysis with remix brief and pillars, cross-run memory, weekly digest.