A show flyer on Instagram is a small public record. The venue tags the headliner and the openers. Often the promoter co-publishes it. One flyer covers one night, but a few hundred of them show who works with whom across a city.
We built that map for Austin from public posts on 1 November 2026. We started from 16 venues and pulled their 36 most recent posts each. Then we pulled the 25 artists those venues tagged most, to see whether the artists tag the venues back. Keeping only the last 12 months left 1,011 posts and 5,947 tag and collab links among 1,845 accounts.
What we found
- Live Nation Austin links more venues than any other account. It co-published or was tagged in posts from 6 of the 16 venues: ACL Live, Antone's, Emo's, Mohawk, Scoot Inn and Stubb's. It wasn't linked to any of the 25 artists we pulled.
- Event listings and radio come next. Do512 and When Where What Austin each link 4 venues. KUTX links 3 venues and 6 of the 25 artists, more artists than any other promoter, label or media account.
- Musicians are the biggest group of connectors. 77 accounts link two or more venues. 28 are artists, 13 are media accounts, 7 are promoters or festivals, 2 are labels and 2 are venues outside our list. We don't name the other 25.
- Labels barely show up. Seven label accounts made it into the shared network, and none of them linked more than two venues.
- A few venues hold most of the overlap. The Continental Club and C-Boy's share 43 linked accounts, more than any other pair. Each of the five busiest pairs includes the Continental Club, C-Boy's or the Broken Spoke.
- Elsewhere the city splits into small worlds. Seven of the ten groups we found hold a single venue. Come and Take It Live shares no linked accounts with the other 15 at all.
- Links mostly run both ways. Of 52 venue and artist pairs we saw on venue posts, 34 (65%) also show up on the artist's own posts. The artists' posts added 8 pairs the venues never posted, plus at least 16 venues that weren't on our starting list.
The ring holds the 16 venues. Each dot inside is a named account linked to two or more of them, placed between the venues it links. One side is crowded with musicians and local media. The other side is mostly a single orange dot with lines out to six rooms.
How a tag becomes a link
Two fields in Instagram post data carry the links:
usertags.inlists the accounts tagged in the photo or video.coauthor_producerslists the other accounts on a Collab post. A Collab post is published by two or more accounts together and appears on each one's profile grid.
This is a real Sagebrush post from 16 October 2026, as our posts endpoint returned it, trimmed to the fields that matter. Sagebrush published it with Matt Tedder and a two-step dance night as co-authors:
{
"code": "DekJzbKmLFY",
"taken_at": 1792174491,
"media_type": 8,
"user": { "username": "sagebrushtexas" },
"coauthor_producers": {
"0": {
"username": "matttedder",
"full_name": "Matt Tedder",
"is_verified": true
},
"1": {
"username": "doubleornothingtwostep",
"full_name": "𝗗𝗢𝗨𝗕𝗟𝗘 𝗢𝗥 𝗡𝗢𝗧𝗛𝗜𝗡𝗚 𝗧𝗪𝗢-𝗦𝗧𝗘𝗣 👢👢",
"is_verified": false
}
},
"usertags": {
"in": {
"0": {
"position": { "0": 0.5224766751, "1": 0.9836601298 },
"user": {
"username": "averyisondrums",
"full_name": "Avery Davis",
"is_verified": false
}
},
"1": {
"position": { "0": 0.7285835454, "1": 0.6853265479 },
"user": {
"username": "the.point.music",
"full_name": "THE POINT.",
"is_verified": false
}
},
"2": {
"position": { "0": 0.5, "1": 0.5 },
"user": {
"username": "matttedder",
"full_name": "Matt Tedder",
"is_verified": true
}
},
"3": {
"position": { "0": 0.5, "1": 0.5 },
"user": {
"username": "doubleornothingtwostep",
"full_name": "𝗗𝗢𝗨𝗕𝗟𝗘 𝗢𝗥 𝗡𝗢𝗧𝗛𝗜𝗡𝗚 𝗧𝗪𝗢-𝗦𝗧𝗘𝗣 👢👢",
"is_verified": false
}
}
}
}
}
We turn each post into links with two rules. Co-authors link to each other, and every author links to every tagged account. That post becomes nine links: three among the co-authors, and six from the three co-authors to the two tagged musicians. A caption can also @name people, but that's free text, so we left it out of the graph.
Some details only show up in real responses. We checked these against 612 venue posts on 1 November 2026:
data.items,usertags.in,coauthor_producersandcarousel_mediaall came back as objects with numbered keys, never as arrays. A loop written for arrays finds nothing.- Each page holds 12 posts. Pass
next_max_idto get the next page, and stop whenmore_availableis false. - On carousels, the post-level
usertagsalready lists the tags from every slide. That held for all 398 carousels in our data. Reading the slides as well costs nothing, so the code below does both. - Each tag holds a
positionand auserwith five fields:id,username,full_name,is_verifiedandprofile_pic_url. Nothing in it says whether the account is a business, a band or a person, or whether it's private. - Display names can be styled Unicode, like the bold letters above. Match accounts on
username. - A Collab post appears on every co-author's grid, so the same post can arrive two or three times. Count it once, by its
code.
Collab posts are everywhere in this data. Half the venue posts were Collabs, and 204 of the 570 posts on venue grids (36%) were published by another account, with the venue as a co-author on 196 of them. Across all 1,011 posts, co-authors were nearly always photo-tagged too, 1,099 times out of 1,111. So usertags finds almost everyone on a post, and coauthor_producers tells you who published it. The paid partnership label is another public field on the same posts, which we measured in our paid partnership study.
Who links the most venues
Live Nation Austin is the only account on six venues. Scoot Inn, Stubb's, Emo's and ACL Live co-published posts with it, and Mohawk and Antone's tagged it.
The two event-listing accounts, Do512 and When Where What Austin, reach four venues each. After that comes a 17-way tie at three, and 11 of the 17 are musicians. Several are players who work in more than one band. Rhys John Stygal plays guitar for Nico Little & The North Americans and is linked to Antone's, C-Boy's and the Continental Club.
We counted a venue only when a separate post linked it. Without that rule, one post co-published by several venues hands every tagged account several venue links at once. CAZAYOUX's 29 September post was co-published by Sahara Lounge, the Continental Club and C-Boy's, and it tagged four more accounts. Under our rule, it gives each of them one venue link instead of three.
The clusters
To find groups, we ran a standard community-detection method (Louvain) on the part of the graph we can name. It scored a modularity of 0.65, which means most links stay inside their own group. Ten groups contain at least one venue:
- The big rooms. ACL Live, Emo's, Scoot Inn and Stubb's, with Live Nation Austin and ACL Festival. ACL Live and Stubb's also co-publish with each other.
- Honky-tonks. The Broken Spoke and Sagebrush, with All Hat No Cadillac, Dale Watson, Ruby Dice, three two-step dance nights and other honky-tonks and dance halls outside our list, including Gruene Hall and the White Horse.
- Cheer Up Charlies and Chess Club. They share exactly one account, Dan-Ramone Vivian Chavez, a guitarist and promoter who co-published posts with both venues. One shared account was enough to group them.
- Continental Club, with Jennifer Foster & The Audacity, The Louisiana Surf Department, the Saxon Pub, the Continental Club's Houston location and much of the local press.
- C-Boy's, with Nico Little & The North Americans, two of the band's players, MOTENKO and the duo Uncle Roy & Spice!
- Hotel Vegas, with Los Desechos, Hecho In Austin Records, a booker whose bio lists both Los Desechos and Hotel Vegas, a cumbia festival and a long list of event-listing accounts.
- Sahara Lounge, with CAZAYOUX, Sylla Music, Spaceflight Records and KUTX.
- Antone's, which bills itself as Austin's home of the blues, with Paul Val, Gary Clark Jr, Jimmie Vaughan, Kam Franklin and the Austin Blues Festival.
- Flamingo Cantina, with ska acts like Los Kurados and The Slackers, plus the radio station KAZI 88.7.
- Mohawk, with Resound Presents.
Come and Take It Live sits outside all of them. Its posts tag a metal podcast, a metal collective and the venue's own production company, and none of those turned up at any other venue we pulled.
The Continental Club and C-Boy's land in separate groups even though they share more accounts than any other pair. C-Boy's shares an owner with the Continental Club, and the two co-publish often, but each also has a large circle of its own. Treat the borders between groups as soft.
Promoters, labels and media
Every promoter, festival and media account linked to three or more venues, plus the two labels that reached two, counting separate posts:
| Account | Type | Venues linked | Artists linked (of 25) |
|---|---|---|---|
| Live Nation Austin | Promoter | 6 | 0 |
| When Where What Austin | Event listings | 4 | 4 |
| Do512 | Event listings | 4 | 3 |
| KUTX | Radio | 3 | 6 |
| ATX Artist Spotlight | Media | 3 | 5 |
| Austin Monthly | Magazine | 3 | 3 |
| Austin Independent Music | Media | 3 | 1 |
| ACL Festival | Festival | 3 | 0 |
| Resound Presents | Promoter | 3 | 0 |
| Hecho In Austin Records | Label | 2 | 1 |
| Howlin' Dog Music Group | Label | 2 | 1 |
Two patterns stand out. The big promoter links rooms, while local media link artists: Live Nation Austin reaches six venues and none of our artists, and KUTX reaches six artists. And labels are close to invisible on venue posts. If you want to know which label sits behind a band, the band's own posts are a better bet. Select-O-Hits and Spaceflight Records were each linked to three of our artists and to one venue.
Why pull the artists too
Venue posts alone leave gaps. Two numbers show what the artists added:
- Links in both directions. We saw 52 venue and artist pairs on venue posts. For 34 of them (65%), the artist's own posts show the same link. The artists' posts also added 8 pairs that the venues' recent posts didn't show, such as three Antone's links from Nico Little & The North Americans, Rhys John Stygal and Paul Val.
- Venues we didn't start with. The artists' posts tagged at least 16 venues outside our list. The Saxon Pub came up for 6 of the 25 artists, the Continental Club's Houston location for 4 and the White Horse for 3.
The ranking changed too. With venue posts only, 12 named accounts reached three venues. With the artists' posts added, it was 20. If you only pull the venues you already know, you miss rooms like the Saxon Pub and undercount the musicians who move between stages.
Build the edge list for any city
This script takes any list of starting accounts, pulls their recent posts and writes one CSV row per link. It counts each post once and reads tags from every carousel slide as well as the post. Each page costs 1 credit, so three starting accounts at two pages each cost 6 credits. In JavaScript (Node 18 or later, saved as an .mjs file):
// Usage: SOCIAVAULT_API_KEY=... node build-edges.mjs mohawkaustin hotelvegastexas > edges.csv
const API = "https://api.sociavault.com/v1/scrape";
const HEADERS = { "x-api-key": process.env.SOCIAVAULT_API_KEY };
const PAGES = Math.min(Number(process.env.PAGES ?? 2), 3); // 12 posts per page
// Lists can come back as arrays or as objects with numbered keys
const asList = (x) => (Array.isArray(x) ? x : Object.values(x ?? {}));
const sleep = (ms) => new Promise((resolve) => setTimeout(resolve, ms));
async function postsPage(handle, cursor) {
const params = new URLSearchParams({ handle });
if (cursor) params.set("next_max_id", cursor);
const res = await fetch(`${API}/instagram/posts?${params}`, {
headers: HEADERS,
});
const body = await res.json().catch(() => ({}));
if (!res.ok || body.error)
throw new Error(`${handle}: ${body.error ?? `HTTP ${res.status}`}`);
return body.data ?? {};
}
// Co-authors link to each other; every author links to every tagged account
function linksIn(post) {
const owner = post.user?.username ?? post.owner?.username;
const coauthors = asList(post.coauthor_producers).map((u) => u?.username);
const authors = [...new Set([owner, ...coauthors].filter(Boolean))];
const tagged = new Set();
for (const media of [post, ...asList(post.carousel_media)]) {
for (const tag of asList(media?.usertags?.in)) {
const name = tag?.user?.username;
if (name && !authors.includes(name)) tagged.add(name);
}
}
const links = [];
for (let i = 0; i < authors.length; i++)
for (let j = i + 1; j < authors.length; j++)
links.push([authors[i], authors[j], "collab"]);
for (const a of authors) for (const t of tagged) links.push([a, t, "tag"]);
return links;
}
const seeds = process.argv
.slice(2)
.map((h) => h.replace(/^@/, "").toLowerCase());
const seen = new Set();
const rows = [];
for (const handle of seeds) {
let cursor;
for (let page = 1; page <= PAGES; page++) {
const data = await postsPage(handle, cursor);
for (const post of asList(data.items)) {
// a collab post shows up on every co-author's grid, so count it once
if (!post?.code || seen.has(post.code)) continue;
seen.add(post.code);
const date = new Date((post.taken_at ?? 0) * 1000)
.toISOString()
.slice(0, 10);
for (const [source, target, link] of linksIn(post))
rows.push([source, target, link, post.code, date]);
}
cursor = data.next_max_id;
if (!data.more_available || !cursor) break;
await sleep(1000);
}
}
console.log("source,target,link,post,date");
for (const row of rows) console.log(row.join(","));
// Accounts linked to two or more of your starting accounts: candidates for the next round
const linkedTo = new Map();
for (const [s, t] of rows) {
for (const [a, b] of [
[s, t],
[t, s],
]) {
if (!seeds.includes(b) || seeds.includes(a)) continue;
if (!linkedTo.has(a)) linkedTo.set(a, new Set());
linkedTo.get(a).add(b);
}
}
const next = [...linkedTo]
.filter(([, s]) => s.size >= 2)
.sort((x, y) => y[1].size - x[1].size);
console.error(
`${rows.length} links from ${seen.size} posts. Linked to 2+ starting accounts: ` +
(next.map(([a, s]) => `${a} (${s.size})`).join(", ") || "none"),
);
And in Python:
# Usage: SOCIAVAULT_API_KEY=... python build_edges.py mohawkaustin hotelvegastexas > edges.csv
import csv
import os
import sys
import time
from datetime import datetime, timezone
import requests
API = "https://api.sociavault.com/v1/scrape"
HEADERS = {"x-api-key": os.environ["SOCIAVAULT_API_KEY"]}
PAGES = min(int(os.environ.get("PAGES", "2")), 3) # 12 posts per page
def as_list(x):
# Lists can come back as lists or as dicts with numbered keys
return list(x.values()) if isinstance(x, dict) else (x or [])
def posts_page(handle, cursor=None):
params = {"handle": handle}
if cursor:
params["next_max_id"] = cursor
r = requests.get(f"{API}/instagram/posts", headers=HEADERS, params=params, timeout=90)
body = r.json()
if not r.ok or body.get("error"):
raise RuntimeError(f"{handle}: {body.get('error') or r.status_code}")
return body.get("data") or {}
def links_in(post):
# Co-authors link to each other; every author links to every tagged account
owner = (post.get("user") or post.get("owner") or {}).get("username")
authors = []
for u in [owner] + [(c or {}).get("username") for c in as_list(post.get("coauthor_producers"))]:
if u and u not in authors:
authors.append(u)
tagged = []
for media in [post] + as_list(post.get("carousel_media")):
for tag in as_list(((media or {}).get("usertags") or {}).get("in")):
name = ((tag or {}).get("user") or {}).get("username")
if name and name not in authors and name not in tagged:
tagged.append(name)
links = [(a, b, "collab") for i, a in enumerate(authors) for b in authors[i + 1:]]
links += [(a, t, "tag") for a in authors for t in tagged]
return links
seeds = [h.lstrip("@").lower() for h in sys.argv[1:]]
seen, rows = set(), []
for handle in seeds:
cursor = None
for _ in range(PAGES):
data = posts_page(handle, cursor)
for post in as_list(data.get("items")):
code = post.get("code")
if not code or code in seen:
continue # a collab post shows up on every co-author's grid
seen.add(code)
date = datetime.fromtimestamp(post.get("taken_at") or 0, tz=timezone.utc).strftime("%Y-%m-%d")
rows += [(s, t, link, code, date) for s, t, link in links_in(post)]
cursor = data.get("next_max_id")
if not data.get("more_available") or not cursor:
break
time.sleep(1)
writer = csv.writer(sys.stdout, lineterminator="\n")
writer.writerow(["source", "target", "link", "post", "date"])
writer.writerows(rows)
# Accounts linked to two or more of your starting accounts: candidates for the next round
linked_to = {}
for s, t, *_ in rows:
for a, b in ((s, t), (t, s)):
if b in seeds and a not in seeds:
linked_to.setdefault(a, set()).add(b)
nxt = sorted(((a, len(v)) for a, v in linked_to.items() if len(v) >= 2), key=lambda x: -x[1])
print(f"{len(rows)} links from {len(seen)} posts. Linked to 2+ starting accounts: "
+ (", ".join(f"{a} ({n})" for a, n in nxt) or "none"), file=sys.stderr)
We ran both on Mohawk and Hotel Vegas with PAGES=1. The JavaScript run returned 227 links from 24 posts and flagged Resound Presents as linked to both venues. The Python run a few minutes later returned 222 links from 24 posts, because a newer Hotel Vegas post had appeared at the top of its grid and pushed an older one to page two. Live data moves, so keep the post and date columns with every row.
The last line of output is your next round. Accounts linked to two or more of your starting accounts are the ones worth pulling next. That's how we picked our 25 artists: the artist accounts that the most venues tagged, each checked before we pulled it.
Check accounts before you name them
A tag tells you nothing about who's behind an account, and many tagged accounts belong to people who just happen to be in the photo. Before naming anyone, we looked up the public profile of every account that carried a person's name, unless it was a verified artist account, with our profile endpoint. We kept only professional accounts with a music category or a music bio. This prints both signals:
// Usage: SOCIAVAULT_API_KEY=... node check-profile.mjs resoundpresents kutx
const API = "https://api.sociavault.com/v1/scrape";
const HEADERS = { "x-api-key": process.env.SOCIAVAULT_API_KEY };
const sleep = (ms) => new Promise((resolve) => setTimeout(resolve, ms));
for (const handle of process.argv.slice(2)) {
const params = new URLSearchParams({ handle });
const res = await fetch(`${API}/instagram/profile?${params}`, {
headers: HEADERS,
});
const body = await res.json().catch(() => ({}));
const user = body.data?.data?.user ?? body.data?.user;
if (!user) {
console.log(`${handle}: skip (${body.error ?? `HTTP ${res.status}`})`); // e.g. "Profile is private"
} else {
const pro = Boolean(user.is_professional_account) && !user.is_private;
console.log(
`${handle}: professional=${pro} category=${user.category_name ?? "none"}`,
);
}
await sleep(1000);
}
# Usage: SOCIAVAULT_API_KEY=... python check_profile.py resoundpresents kutx
import os
import sys
import time
import requests
API = "https://api.sociavault.com/v1/scrape"
HEADERS = {"x-api-key": os.environ["SOCIAVAULT_API_KEY"]}
for handle in sys.argv[1:]:
r = requests.get(f"{API}/instagram/profile", headers=HEADERS, params={"handle": handle}, timeout=60)
body = r.json()
data = body.get("data") or {}
user = (data.get("data") or {}).get("user") or data.get("user")
if not user:
print(f"{handle}: skip ({body.get('error') or r.status_code})") # e.g. "Profile is private"
else:
pro = bool(user.get("is_professional_account")) and not user.get("is_private")
print(f"{handle}: professional={pro} category={user.get('category_name') or 'none'}")
time.sleep(1)
We checked 54 accounts. Four were private, which came back as "Profile is private" at no charge, even though all four were tagged on public posts. Eleven were personal accounts. Five more were professional accounts we didn't count as artists, such as a lifestyle creator and a photographer. That left 34 we could name.
The category is less useful than it looks. 25 of the 50 profiles showed no category at all, including 14 professional accounts, because an account can hide it. Resound Presents and KUTX both came back with category=none in the runs above. For accounts like those, the bio decided it.
Everyone else stays out of the charts, this post and the download. That's 1,693 of the 1,804 accounts in the pull other than the 41 we pulled, or 94%. They still count in the totals above, but none of them is named anywhere.
What this map can't tell you
- It's a snapshot of one autumn. The median venue post is from 21 October 2026, and 90% are from 2 October or later. Three pages reached back only 11 days at Sahara Lounge and about eight weeks at Stubb's. A band that played the Mohawk in March isn't in it.
- Tagging habits shape the map. Hotel Vegas tagged someone in all 36 of its posts. Stubb's tagged someone in 10 of 36. A venue that rarely tags looks isolated whether it is or not.
- Tags miss names. 24% of venue posts and 42% of artist posts @named an account in the caption without tagging it. Those accounts aren't in our graph.
- A tag can mean many things. A link can be a sponsor or a friend who happened to be in the photo. The graph records who tagged whom and says nothing about money or contracts.
- Our naming rule undercounts musicians. Some working players use personal accounts. Two guitarists we checked did, so they stay unnamed, along with every link that runs through them.
- One artist is half missing. The posts for THE POINT. returned a retryable error four times in a row, not charged, so that band appears only through venue posts.
- One venue is gone. We dropped Empire Control Room & Garage, which closed over the summer. Its newest post was from May.
- The clusters are one reading. Another community method, or another month of posts, would draw some borders differently.
How we did it
- Starting accounts. 16 Austin venues with public Instagram accounts, from ACL Live and Stubb's down to small rooms like Sahara Lounge and Chess Club. Twelve handles came from the venues' own websites, three from tags on other venues' posts and one from trying the obvious name.
- Pulls. Three pages (36 posts) per venue, then two pages (24 posts) for each of the 25 artist accounts that the most venues tagged, all through
/v1/scrape/instagram/postson 1 November 2026. Three artists post so rarely that two pages reached back to 2014, 2019 and 2023. - Window. We kept posts from 1 November 2025 to 1 November 2026. That dropped 6 older venue posts and 103 artist posts, and left one artist with no posts at all.
- Links. Co-authors link to each other and every author links to every tagged account, with each post counted once. Venue counts use separate posts only.
- Groups. Louvain community detection on the links between named accounts, weighted by the number of posts.
- Cost. 160 credits: 51 for venue pages (including 3 for Empire before we dropped it), 50 for artist pages, 50 for profile checks and 9 for testing the code above. Two venue handles we guessed came back unavailable. Those errors, the four failed tries for THE POINT. and the four private profiles cost nothing.
The CSV download has all 614 links among the 151 named accounts in the graph: source, target, both account types, tag or collab, the number of posts and the first and last post date. If you're building something similar on other platforms, our guide to finding the user ID for any account covers the IDs worth storing next to each handle, and why an API says an account doesn't exist covers the private and renamed accounts you'll hit along the way.
Frequently Asked Questions
Can you get the accounts tagged in an Instagram post through an API?
Yes. The posts endpoint returns them in usertags.in for each post, with the username, display name, verified flag, account ID and profile picture. In our responses usertags.in was an object with numbered keys, so read it with Object.values in JavaScript or .values() in Python.
What is coauthor_producers in Instagram post data?
It lists the co-authors of a Collab post, the accounts that published the post together with its owner. Half of the 570 venue posts in our pull were Collab posts, and 36% of the posts on venue grids were published by another account. Co-authors were photo-tagged as well 99% of the time, so the field matters most for knowing who published a post.
Do I need to read every carousel slide to get all the tags?
In our data, no. The post-level usertags matched the slide tags on all 398 carousels. Reading carousel_media as well costs nothing extra, so our code does both in case a post differs.
How can I tell if a tagged account is a personal account?
The tag won't tell you. It has no private flag and no account type. Look up the profile and check is_professional_account and category_name, then read the bio, because half the profiles we checked showed no category. Private profiles come back with "Profile is private" and no charge.
How many posts do you need to map a city's music scene?
Three pages per venue, 36 posts, covered anywhere from 11 days to about eight weeks in Austin, depending on how often the venue posts. That's enough to see who works together this season. For a longer view, pull more pages from the venues that post most and add the artists' own posts, which reach further back because artists post less often.
Is it okay to publish a map of who tags whom?
Our rule is to name only venues, artists, labels, promoters and media with public profiles, to check any account that looks like a person before naming it, and to keep everyone else as counts. Tags on public posts are visible to anyone, but the people in a photo didn't sign up to be in your dataset. This is our practice, not legal advice.
Want to map a scene you know? Sign up for 50 free credits and run the script on three venues. At two pages each, that's 6 credits.
Found this helpful?
Share it with others who might benefit
Ready to Try SociaVault?
Start extracting social media data with our powerful API. No credit card required.