Instagram Profile Lead Scraper By Keyword

Instagram Profile Lead Scraper — By Keyword 📱🔍

Give it a keyword. It gives you back Instagram profiles that publish an email address in their bio or captions, ranked by how well they match that keyword.

This is the keyword-driven member of the family. Instead of firing one query and hoping, it expands every keyword into several complementary Instagram-native search angles, pools everything it finds, de-duplicates by profile, and then ranks the pool by keyword relevance before saving the best leads.

"fitness coach"
   ├── exact phrase        →  "fitness coach"
   ├── hashtag             →  "#fitnesscoach", "#fitness"
   ├── bio contact-intent  →  "fitness coach" ("dm for" OR "email for" OR "inquiries" OR "collabs" …)
   ├── profile audience    →  "fitness coach" "followers"
   ├── loose terms         →  "fitness" "coach"
   └── your own extras     →  whatever you add in ➕ Extra Keyword Variants

Each angle surfaces a different slice of the index. In live testing the hashtag angle alone returned nine profiles that the plain-phrase angle never showed.

---

📌 Where the email actually comes from

We think you deserve a straight answer instead of marketing copy.

The actor fetches each profile for real and reads its actual bio. Instagram serves a normal browser a ~620 KB JavaScript shell with no profile data in it — that is the dead end most scrapers hit. Served to a declared crawler user-agent, the same URL returns a ~1.2 MB server-rendered document containing the real profile record: biography, full_name, follower_count, and the account's own recent post captions.

What we tried, live from Apify	Result
instagram.com/<handle>/ with a desktop Chrome UA	~620 KB JS shell — no biography, no counts.
instagram.com/<handle>/ with a crawler UA	~1.2 MB, real profile data. This is what the actor uses.
api/v1/users/web_profile_info/ + x-ig-app-id	HTTP 429 on every proxy group. Dead.
Apify UNBLOCKER group	TLS failure.
Brave / Bing / DuckDuckGo / Mojeek / Startpage / Ecosia	429, 202 block pages, 403, or no site: support. Unusable.

So the pipeline is genuinely two-stage:

1. DISCOVER — the 6-angle keyword expansion finds which accounts are relevant.
2. FETCH — every discovered handle's real profile page is fetched through a residential proxy, and the email is read from the account's own profile.

🔒 Attribution — why a row's email really belongs to that account

The naive version of this scraper reads the email out of a search-result snippet. That is how you end up asserting that @indianyogaassociation's contact address is aymyogaschool@gmail.com — an address that merely appeared in a caption on one of its pages. The row says "this account's email", so a near-miss is a wrong value, not a thin one.

Every row therefore carries an attributionConfidence:

Value	Evidence	emailSource
bio_verified	Read from the account's own bio field on its fetched profile. Strongest.	instagram_profile_bio
own_page_verified	The account repeatedly publishes it in its own captions, or it carries the account's own name.	instagram_own_caption
handle_matched	Opt-in fallback only: a search snippet address that matches the account name.	google_serp_caption_snippet

A one-off address in a single caption is rejected, even when it is the only address on the page — verified case: @dougsyoga's lone caption address was the event registrar's inbox, not his. Addresses that share only a generic trade word (yoga, studio, academy) or the search keyword itself are rejected too, since two unrelated businesses in the same niche always share those.

Default is first-party only (bio_verified + own_page_verified). If that yields fewer leads than you asked for, the actor says so in the log rather than padding the list.

📊 Output fields

Every field below is populated on every row. We do not emit blank columns.

Field	Example	Description
network	instagram.com	Source network.
keyword	yoga teacher	The keyword you asked for.
title	International Yoga Teacher (@yogagise)	Headline of the result that discovered this account.
description	`International Yoga Teacher
Lifestyle, Travel…`	The account's real bio, fetched from its profile.
url	https://www.instagram.com/yogagise/	Canonical profile URL.
email	yogagise@gmail.com	The contact address, attributed to this account.
handle	yogagise	Instagram username.
fullName	International Yoga Teacher	Real display name from the fetched profile.
followerCount	12599	Real follower count from the fetched profile.
matchedKeyword	yoga teacher	Which of your keywords produced this lead.
keywordVariant	hashtag #yogateacher	Which expansion of that keyword found it.
sourceType	profile_page / post_page / reel_page	Which surface discovered the account.
emailSource	instagram_profile_bio	Where the address was actually read from.
handleSource	result_url / title_handle / cite_breadcrumb / source_breadcrumb	How the username was recovered.
attributionConfidence	bio_verified	Strength of the email→account link (see table above).
relevanceScore	13.55	Keyword-relevance score.
keywordRank	1	Position within its keyword, best first.

⚙️ Input

Input	Required	Default	What it does
🔍 Keywords	✅	["fitness coach"]	One or more keywords. Each is expanded into several search angles.
🌐 Platform	—	Instagram	Locked to Instagram today.
📍 Location Filter	—	(empty)	Adds a city/region term to every query.
📧 Email Domains Filter	—	["@gmail.com"]	Keep only leads on these domains. Empty = all major free-mail providers.
📊 Maximum Leads per Keyword	—	20	How many ranked leads to save per keyword.
⚙️ Engine	—	legacy	Processing engine.
#️⃣ Hashtag Discovery	—	true	Also search the hashtag spelling of each keyword.
➕ Extra Keyword Variants	—	[]	Your own extra search phrases, added to the automatic expansions.
👤 First-Party Emails Only	—	true	Keep only emails read from the account’s own profile. Turn OFF to also accept name-matched search-snippet addresses.
📄 Pages per Keyword Variant	—	2	Result pages (20 each) to read per variant.
🏅 Minimum Relevance Score	—	0	Drop leads scoring below this. Good leads typically land at 8–14.
🛡️ Proxy Configuration	—	GOOGLE_SERP	Leave it alone unless you know why you are changing it.

How ranking works

The actor gathers a candidate pool roughly twice the size of your quota, then scores each profile:

• +3 the full keyword phrase appears in the title/bio/handle
• +1 per significant keyword token found
• +2 a keyword token appears inside the username (@lucindabucatfitness for "fitness coach")
• +1.5 per hashtag form matched
• +2.5 the lead came from an actual profile page rather than a post
• up to +2.5 for audience size, when the result exposes a follower count
• +1 the email local part is built from the handle (a real business account)

The top maxEmails leads are saved, best first.

---

🚀 How to use

1. Open the actor on Apify and click Try for free.
2. Type one or more keywords (be specific — "vegan meal prep coach" beats "coach").
3. Optionally set a location and an email domain filter.
4. Set Maximum Leads per Keyword.
5. Start the run, then export the dataset as CSV, JSON or Excel.

💡 If you get fewer leads than you asked for, that is the honest answer: only a fraction of profiles publish an email in a way search engines indexed. Widen the keyword, drop the location filter, raise Pages per Keyword Variant, or clear the Email Domains Filter.

---

⚖️ Compliance

This actor reads publicly indexed search results only. It does not log in, does not bypass authentication, and does not touch private accounts or content behind Instagram's login wall. Contact data you collect is personal data — use it in line with GDPR/CCPA and applicable anti-spam law, and honour opt-outs.

💬 Support and feedback

• Bug reports: open a ticket in the Issues section
• Custom features / enterprise: dev.scraperengine@gmail.com