Instagram similar accounts: build lookalike creator lists
When you find one creator who fits your brand, you usually want twenty more like them. Instagram already knows who is similar: it shows suggested accounts on profiles. Collected in a loop, those suggestions become a map of a niche. This post shows how to turn them into a ranked shortlist.
What the suggestions are
On many profiles, Instagram shows a row of suggested accounts. Instagram picks them; nobody outside Instagram knows the exact signals. Owners can also turn account suggestions off for their profile Instagram Help Center. So some accounts will return an empty list, and that is expected.
Step 1: choose good seeds
Seeds decide everything. Pick 3 to 5 accounts that are clearly in the niche you want, at roughly the size you want. Seeding from a celebrity tends to return other celebrities. Seeding from mid-sized creators returns mid-sized creators.
Step 2: expand one or two levels
Get the similar accounts for each seed. Then get the similar accounts for each result. Two levels from 5 seeds is often several hundred unique accounts. Stop there; a third level drifts out of the niche.
Step 3: score by overlap
An account suggested by four different seeds is more central to the niche than one suggested once. Count how many times each handle appears across all lists. That count is your first ranking signal.
import os, requests
from collections import Counter
API = "https://scrapeio.com"
H = {"x-api-key": os.environ["SCRAPEIO_API_KEY"]}
def similar(handle):
r = requests.get(f"{API}/v1/instagram/user/similar", params={"handle": handle}, headers=H).json()
return [a["handle"] for a in r["data"]["items"]] if r.get("success") else []
seeds = ["seed_one", "seed_two", "seed_three"]
level1 = {s: similar(s) for s in seeds}
hits = Counter(h for lst in level1.values() for h in lst)
for h in list(hits)[:30]:
hits.update(similar(h))
for s in seeds:
hits.pop(s, None)
print(hits.most_common(50))Step 4: enrich and filter
Similar-account lists give handles and names, not stats. Pull a profile for each candidate, and then recent reels for the ones that pass the first filter:
- Size band: keep accounts within your target, for example 10,000 to 200,000 followers.
- Active: at least one post in the last 14 days.
- Public: skip private accounts.
- Reach multiple: median reel plays divided by followers. There is no universal good value; compare within your list.
- Engagement: median (likes + comments) / followers. Very high followers with very low engagement is a warning sign.
Step 5: check fit by hand
Numbers find candidates; people decide fit. Open the top 20, read a few captions, watch two reels, and check the comments for real conversation. Remove anything off-brand. Keep notes on why you rejected each one; after a few rounds those notes become filters you can automate.
A scoring formula
To sort the final list, combine the signals into one score:
score = overlap_count * 2
+ min(reach_multiple, 3)
+ engagement_rate * 10The weights are a starting point. Adjust them until the top of the list matches what your team would pick by hand.
Other ways to find lookalikes
- Keyword search: search accounts by niche terms and treat results as extra seeds.
- Hashtags: authors who post under the niche's tags this month.
- Audio: creators using the same trending sound.
- Influencer platforms: tools with indexed databases and audience demographics are the better choice if you need demographic filters out of the box. See our Modash comparison.
Doing it with ScrapeIO
Use the instagram similar accounts API to expand seeds, the instagram profile API to enrich, and the instagram user reels API for reach and engagement. A 300-candidate run with 60 reel checks costs under 400 credits. The micro-influencer guide shows the same flow with an AI agent and the influencer-discovery skill.
Sources
Checked on 8 November 2026 from each vendor's own pages. Prices and limits change; follow the links before you decide.
100 free credits, no card. Credits never expire.
Get API key