A Politician on Instagram Is 3x More Likely to Be Verified Than a Photographer

Instagram verifies some professions far more than others. A Public Figure account is over three times as likely to carry the blue check as a Photographer account — same platform, same verification policy, wildly different outcomes. The easy read is that Instagram simply trusts politicians and athletes more than photographers. The real reason is duller and more interesting: it's mostly about how each profession ended up in our index in the first place, not about who Instagram thinks deserves a checkmark.

We built this index the same seeded way we built the X users dataset — cross-referencing public profiles against Wikidata (an independent, external notability signal) and Common Crawl mentions. Instagram carries a field X doesn't: a professional category (Artist, Athlete, Politician, Photographer, and so on) that every business or creator account can set. That field lets us ask a question X's dataset structurally cannot answer — does verification track profession, and if so, why.

Verification swings 3x by profession

Across the ten largest professional categories in our 616,627-profile index, verification rate is nowhere close to flat:

Public figure71.0%n=4,067

Athlete67.5%n=11,974

Actor66.3%n=5,328

Politician58.7%n=1,452

Journalist55.4%n=1,072

Musician/band48.0%n=6,826

Artist45.0%n=17,733

Digital creator40.9%n=5,505

Author38.5%n=1,230

Photographer22.9%n=4,538

Public Figure sits at 71.0%. Photographer sits at 22.9%. Both are legitimate professional categories Instagram itself defines; both are subject to the same verification policy. Whatever is driving the gap, it isn't a documented rule that treats photographers differently.

The reason: how we found them, not who they are

Every profile in our index was discovered one of two main ways — cross-referenced against Wikidata (people with an independent, external notability record) or surfaced through Common Crawl mentions on the open web. That discovery channel, source_tier, isn't a judgment of importance; it's just a record of which door a profile walked in through. But it turns out to predict verification rate almost category-for-category:

Wikidata seed shareVerified rate

Politician98.0% → 58.7%n=1,452

Actor93.7% → 66.3%n=5,328

Athlete93.6% → 67.5%n=11,974

Musician/band71.4% → 48.0%n=6,826

Public figure67.8% → 71.0%n=4,067

Journalist66.1% → 55.4%n=1,072

Artist54.2% → 45.0%n=17,733

Author42.9% → 38.5%n=1,230

Digital creator24.3% → 40.9%n=5,505

Photographer7.5% → 22.9%n=4,538

Politicians are 98.0% Wikidata-seeded. Photographers are 7.5%. That's a 13x spread — sharper than the verification-rate spread itself — and it's why the verification-rate spread exists: a category that's mostly discovered through an independent record of real-world notability is, unsurprisingly, mostly made up of people who were already famous enough to get verified before they ever ran an Instagram account. A category that's mostly discovered through open-web mentions is mostly made up of people who built their following on Instagram itself, with no prior record for the platform to lean on.

The two rankings aren't identical — Photographer is bottom on both, Politician/Athlete/Actor sit at the top of the Wikidata ranking and near the top of the verified ranking — but they track each other closely enough that the pattern isn't coincidence (a rank correlation around 0.8 across the 10 categories; take that as directional, not a formal significance test on a sample this small).

Digital Creator is the interesting exception. Its verified rate (40.9%) is nearly double what its Wikidata share (24.3%) alone would predict. Every other category's verified rate sits within a few points of what its Wikidata share implies; Digital Creator over-verifies relative to its seed composition — plausibly because Instagram runs its own deliberate creator-verification push, a second pathway to a checkmark that doesn't run through "already notable somewhere else."

Here's the same comparison as a table, in case you want the receipts without hovering a chart:

A structural fact worth knowing before you use this field

Before we go further, a gotcha we hit and want to save the next person from re-discovering: Instagram's category field is not a general attribute of every account — it only exists on business or creator accounts. We checked directly: zero of the 408,006 personal (non-business) accounts in our index carry any category value at all, and filtering for a personal account with any category set returns nothing, every time. Setting up a professional Instagram account is what puts a category on a profile in the first place — Meta's own help documentation confirms it's a mandatory step of that flow, not an optional field a personal account can also fill in.

That means every number in this post describes the ~34% of the index that opted into a business or creator account — never the whole platform, and never "Instagram" in general. Worth flagging for anyone else cross-tabbing this dataset by category: you're already looking at the professionalized slice before you filter anything else.

The checkmark isn't the professionalism signal — the business toggle is

If you had to guess which single flag best predicts a "professionalized" bio — a linked website — the blue check is the obvious guess. It's wrong:

Business accounts85.8%

179,093 of 208,621

Personal accounts38.8%

158,340 of 408,006

Verified accounts56.1%

74,794 of 133,200

Non-verified accounts54.3%

262,639 of 483,427

Business-account status is a 2.2x predictor of a linked bio: 85.8% of business accounts carry an external link versus 38.8% of personal accounts. Verification status barely moves the same number at all — 56.1% verified versus 54.3% non-verified, a 1.8-point difference on a metric where business-vs-personal moves it 47 points. The visible status symbol is the checkmark; the thing that actually predicts a professionalized bio is the mundane account-type toggle nobody notices.

What we checked and ruled out

The obvious follow-up question: do spammy, bot-shaped accounts cluster in any particular category — heavy on the "follow everyone, get followed by few" pattern that usually flags automation? We checked, and the honest answer is no, because there's almost nothing there to cluster. Filtering the entire 616,627-record index for a follower-to-following ratio of 0.05 or below returns just 91 records total. Spread across the top categories that show up at all — Artist (15), Digital creator (9), Blogger (9), and a scattering of 1-4 elsewhere — there's no meaningful skew, just noise on a tiny sample. This index is too curated (every profile was found through a deliberate seed channel, not a raw platform crawl) to support a "which categories are spammiest" story, so we're not forcing one.

What this shows, and what it doesn't

What we think holds up: verification rate varies sharply by professional category on Instagram, and the variation is explained almost entirely by which discovery channel dominates that category — Wikidata-seeded categories (Politician, Athlete, Actor) verify far more than open-web-seeded ones (Photographer, Digital Creator), tracking an independent notability signal Instagram never controlled.

What limits it:

• This is a seeded index of category-tagged business/creator accounts, not a census of Instagram — see the structural note above. category only exists on the ~34% of the index that set up a professional account.
• source_tier records how our crawl discovered a profile, not how legitimate or important it is. A high Wikidata share means a category skews toward people independently notable before Instagram; it is not a claim that Common-Crawl-sourced accounts are less real.
• Smaller categories carry thinner samples. Journalist (1,072), Author (1,230) and Politician (1,452) are an order of magnitude smaller than Artist (17,733) or Athlete (11,974) — treat the smaller buckets' exact figures with more caution.
• Wikidata-share tracking verified-rate is a strong correlation across 10 categories, not a causal proof. The plausible mechanism (independent notability overlaps with platform verification criteria) is stated as a mechanism, not asserted as settled fact.
• The follow-ratio finding is a stated negative result on 91 records — evidence that this particular slice of Instagram isn't visibly spam-heavy, not evidence about spam on Instagram generally.
• We publish aggregates only — category-level rates, not individual verification statuses tied to named profiles.

None of this says Instagram is arbitrarily biased toward politicians and athletes. It says something narrower and more measurable: which profession you're in predicts, quite strongly, which door your Instagram account walked through to get discovered — and that door predicts the checkmark almost as well as anything Instagram itself controls.

Sources

Methodology

We built a search index of 616,627 public Instagram profiles, each carrying follower count, following count, professional category (where set), verification flag, business-account flag, bio/external-link presence, and the seed channel it was discovered through (Wikidata or Common Crawl, plus small head-directory and canary tiers). The full corpus is queryable through the Instagram users dataset and reconciled against the dataset's facet counts — every cross-tab in this post reconciles its true/false or tier counts back to each category's flat total.

Category-level verification and Wikidata-seed rates were computed by filtering the search and facets endpoints to each of the ten largest professional categories in turn. "Wikidata-seeded" means the profile was discovered via cross-reference against Wikidata; it is not a claim about the person's real-world fame independent of that record. We publish aggregates only — category-level rates, no individual profile-level verification data.

Want to run your own cut? The Instagram users dataset is queryable by category, verification status, business-account flag and follower band, the how to scrape Instagram guide covers the calls, and pricing has the free tier. See also the companion study: a random developer is 5x more likely to have an X blue check than a Wikipedia-notable person.