GMB Photo Optimisation for Healthcare — What Google Scores
What Google's local ranker reads photos for
Google uses photo signals as inputs into three different systems: local-pack ranking (via profile-completeness and freshness), click-through rate on the profile (via first-view thumbnails), and content classification (via image-recognition on interior and exterior scenes). The three systems overlap but are not identical, which is why photo strategy has to think about all three at once.
Ranking inputs are dominated by category coverage, recency, and geo-tag consistency. Click-through inputs are dominated by cover-photo quality, doctor headshot warmth, and whether the interior imagery matches what a patient will experience walking in. Content classification inputs are dominated by whether the photos actually depict a clinic (as opposed to stock or generic medical imagery).
The six-category upload discipline
| Category | Target count | Purpose |
|---|---|---|
| Exterior | 4 to 8 | Landmark recognition, "is this findable" |
| Interior — reception | 3 to 5 | First-impression, trust |
| Interior — consultation and treatment areas | 6 to 12 | Content classification, service depth |
| Team — doctors and staff | 6 to 15 | Person-entity reinforcement, click-through |
| Equipment and diagnostic infrastructure | 4 to 8 | Service capability signalling |
| Events and community activity | 2 to 4 per month | Freshness |
Total steady-state library: 40 to 90 photos across the categories. The event-and-activity category is the one that keeps the freshness signal alive month over month. Under-invest here and the profile ages regardless of other categories.
The exterior photos — Google's proximity classifier reads these
Google's Street View integration and image-recognition classifier use exterior photos to verify that the profile's registered address matches the physical location a patient would recognise. Weak exterior imagery — no signage visible, no street context, taken at night — can quietly hurt the "prominence" signal that feeds proximity ranking. Strong exterior imagery includes the clinic's signage clearly readable, the street or building context visible, and photos from multiple angles.
Upload discipline for exterior: 4 to 8 photos, taken during daylight, across multiple angles (front approach, side view showing signage, closer signage-only frame, cross-street view). If the clinic is inside a larger complex, include the wayfinding cue — the sign at the complex entrance that patients would follow.
The doctor headshots — click-through's highest lever
Doctor headshots are the single highest-lift photo category for click-through rate on Indian clinic profiles. A clinic with warm, professionally-shot headshots of each consultant sees 18 to 34 percent higher call and message actions on the profile than a matched clinic with generic team photos or no consultant headshots.
The specification we run: neutral background, natural light or diffused studio light, doctor in professional attire (white coat is fine, business formal is fine), warm neutral facial expression (not a full smile, not a stern professional pose), medium framing (chest up), 4:5 or 1:1 aspect ratio. The single most common failure mode: doctor headshots taken on a phone in the corridor with fluorescent overhead lighting. That failure mode drops click-through by roughly 20 percent versus a properly-shot alternative.
The compliance envelope for healthcare photos
Under NMC advertising rules and Google's content policies, four photo categories are off-limits and produce takedowns or profile-level warnings:
- Patient before-and-after visuals. Prohibited under NMC even with consent for advertising purposes.
- Identifiable patient photos. Faces, name tags, medical records in frame, room-number tags — all privacy exposures.
- Photos taken during live procedures without a general contextual disclaimer. Compliance and platform-policy risk.
- Photos with visible pricing that reads as guaranteed outcome. "Full teeth cleanup — 999" tile on the wall is fine as a general price; a poster saying "Guaranteed weight loss 5 kg in 30 days" is not.
Geo-tagging and metadata
Google reads EXIF metadata on uploaded photos. Geo-tagged photos with location metadata that matches the profile's registered address are treated as more trustworthy than metadata-stripped uploads. On iOS and Android, cameras geo-tag by default; the discipline is to not strip metadata during editing. Some batch-editing tools quietly strip EXIF; use a workflow that preserves it.
For clinics that shoot exterior photos across multiple angles from a slight distance (e.g., a landmark shot from across the road), the geo-tag will show the shooter's location, not the clinic's. That is fine as long as the geo-tag is within roughly 200 metres of the registered address. Uploading photos with geo-tags from a different city or from a photographer's home suburb is a signal Google reads as untrustworthy.
The upload cadence — why weekly matters more than volume
A clinic that uploads 40 photos in one batch and then goes quiet for 8 months earns a large one-time completeness bump and then decays. A clinic that uploads 2 to 4 photos per week for 6 months earns a steady freshness signal that compounds. Google's local ranker weighs recency; upload behaviour that reads as "living profile" outperforms one-time bulk uploads even when the bulk upload has higher absolute count.
Angryturtle schedules photo uploads across the week — one team photo Monday, one equipment or interior shot Wednesday, one event or activity shot Friday. The cadence is set at onboarding; the specific photos are pulled from a rolling library the clinic builds monthly.
What "healthy" looks like on the photo layer
- 50+ photos across the six categories, no category below its floor.
- Freshness — at least one photo per week for the last 12 weeks.
- Doctor headshots — every consultant, professionally shot, updated when a consultant joins or leaves.
- Compliance — zero patient-identifying imagery, zero before-and-after, zero guaranteed-outcome overlays.
- Geo-tag consistency — 90 percent+ of uploads carry EXIF geo-tags within 200 metres of the registered address.
Get a photo audit and upload cadence inside Angryturtle
Angryturtle audits your current photo library across the six categories, flags compliance risks, and sets up the weekly upload cadence with a rolling clinic-side capture list. Solo clinics run it themselves; multi-location groups run it inside our Managed track.
Explore Angryturtle WhatsApp ICG