How a single-location dermatology clinic ran a engagement ChatGPT Ads campaign with Rs 40k monthly media
The situation
Picture a single-location dermatology clinic in a large tier-1 city suburb — the kind of practice run by one principal dermatologist with two junior consultants, seeing a mix of medical dermatology (acne, eczema, pigmentation) and elective cosmetic work (laser treatments, chemical peels, anti-ageing procedures). The clinic had never run paid advertising in any real sense before this engagement — its patient acquisition ran almost entirely on referrals and a Google Business Profile listing the principal dermatologist maintained herself. Revenue was healthy but growth had flattened over the past year as two newer clinics opened within a three-kilometre radius, both running visible Google Ads and Meta Ads campaigns the original clinic had no answer to.
The clinic's owner was cautious about paid advertising generally, having heard from peers in the specialty that Google Ads cost per click for dermatology and cosmetic-adjacent keywords in her city had become expensive enough that a small clinic's return on a modest budget was questionable. Her actual budget ceiling was firm — Rs 40,000 a month in media spend, on top of whatever management fee an agency charged, was the absolute limit she was willing to commit without seeing results first. That constraint shaped every decision in this engagement more than any other single factor.
ChatGPT Ads appealed to her for a specific reason once ICG explained the model: because bidding pegs to conversation stage rather than a fixed keyword auction, a small, disciplined budget could still capture meaningful early-exploration volume — patients researching a skin concern or a cosmetic procedure before they've decided where to go — at a cost per conversation the two newer, better-funded competing clinics weren't necessarily out-competing her on, since neither of them had a ChatGPT Ads presence yet either. The whitespace argument, not a promise of dramatic volume, was what moved her to commit to a three-month engagement pilot.
The ChatGPT Ads campaign structure ICG designed
At Rs 40,000 monthly media spend, ICG built a single, tightly disciplined conversation bucket covering the clinic's full service mix rather than splitting medical and cosmetic dermatology into separate buckets. Splitting the budget two ways at this spend level would have left each half too thin to reach a stable bid ladder within a reasonable timeframe, so both service lines ran through one bucket with service-type tagging built into the reporting layer rather than into separate campaign infrastructure.
The bid ladder allocation reflected the budget constraint directly. Roughly 70% of the monthly spend was allocated to early-exploration conversation stages — users asking general questions about a skin concern, or researching what a procedure like laser hair removal or a chemical peel actually involves — where per-conversation cost sits at the floor of the range. The remaining 30% was reserved for mid-to-late intent conversations, capped at a ceiling ICG set conservatively in month one specifically because late-intent conversations in dermatology can bid 6-12x higher than early-stage ones, and a Rs 40,000 monthly budget could exhaust itself on a handful of late-intent conversations if the ceiling wasn't controlled tightly.
Landing infrastructure was a single machine-scannable page covering the clinic's core services — factual, short paragraphs on the medical and cosmetic dermatology offerings, with a single enquiry form rather than separate forms per procedure. This kept production cost proportionate to the account's size; a engagement budget doesn't justify building six separate procedure-specific landing pages the way a larger account might. The page was built to load fast and read cleanly on mobile, since dermatology enquiries — particularly cosmetic ones — skew toward mobile-first research sessions.
Reporting ran monthly on a single dashboard tracking completed conversations, cost per qualified lead, and a simple split between medical-dermatology-tagged and cosmetic-tagged conversations, reconciled by hand each month against the clinic's own appointment book since the clinic's practice-management software wasn't set up for automated integration at this stage. This lean structure — one bucket, one page, one dashboard — is the defining shape of engagement: enough machinery to run a real, measurable campaign, without infrastructure the budget can't yet justify.
The compliance discipline
Dermatology carries a particular compliance pressure point that ICG's review specifically targeted in this account: cosmetic procedure outcome claims. Laser hair removal, chemical peels and anti-ageing treatments are exactly the category where ad copy tends to drift toward implied guarantees — "permanent" results, "flawless" skin, specific before-after framing — and every ad variant referencing a cosmetic procedure was checked against NMC Section 6 and ASCI Chapter III to ensure no phrasing implied a guaranteed or universal outcome. Where the clinic's own marketing materials used language like "permanent hair reduction," ICG's copy review reframed it to reflect what's clinically accurate — a reduction in hair growth over repeated sessions, not a permanent guarantee — because the compliance risk of overstatement in cosmetic dermatology is higher than in most other specialties given how frequently ASCI has flagged this exact category.
Before-after visual and descriptive claims received the same scrutiny as in any dermatology account. The clinic had a modest library of patient before-after photos it had used on its Instagram page with patient consent, and there was an early instinct to bring similar descriptive language into ChatGPT Ads copy. ICG's compliance stance was that even consented before-after material needs to be framed as illustrative rather than representative or promised — language describing a typical treatment course and its general expected direction of change, without implying any specific patient's result as a guarantee for a new patient considering the same procedure.
Medical dermatology copy — acne, eczema, pigmentation concerns — carried a lighter compliance load than the cosmetic side but still ran through the same NMC Section 6 baseline: no claims of guaranteed cure, appropriate framing around chronic conditions like eczema that don't have a single definitive resolution, and no comparative language against the two newer competing clinics the owner was implicitly worried about. DPDP 2023 governed the clinic's single enquiry form from day one — explicit consent language, clear statement of how contact information would be used, and no default sharing with any third party, which mattered particularly given the clinic's small size meant it had no dedicated data-protection officer and ICG's form design had to carry that compliance weight on its own.
The 90-day outcome pattern
Illustrative numbers for this hypothetical: month one produced roughly 12-15 completed conversations a week, reflecting both the budget ceiling and the standard data-gathering window every new ChatGPT Ads account runs through before bid ladders stabilise. By month two, volume rose modestly to around 18-22 conversations a week as the 70/30 early-to-late-intent split began producing a cleaner signal on which conversation shapes were converting. By month three, weekly volume settled around 20-25 completed conversations, a plateau consistent with what a Rs 40,000 monthly budget can sustain once the bid ladder has learned the account's conversion patterns.
Cost per qualified lead ran high in week one, as expected, and settled by month two into a range the clinic owner judged competitive with — and in the cosmetic-procedure segment specifically, better than — what she'd heard from peers about Google Ads cost per lead for similar procedures in her city. The cosmetic-tagged conversations, despite representing the smaller share of the bid ladder's budget allocation, produced a disproportionately higher conversion-to-consultation rate than the medical-dermatology-tagged conversations — illustrative figures put cosmetic conversions converting to booked consultations at roughly 35-40%, against roughly 20-25% for medical dermatology, plausibly reflecting that cosmetic-procedure researchers tend to arrive closer to a purchase decision than someone researching a chronic skin condition.
In GA4, this traffic aggregated under the AI Assistant channel, consistent with the elevated key-event rate ICG tracks across its healthcare book generally — well above the clinic's modest organic search baseline and its negligible direct traffic. By month three, the clinic owner's own read on the pilot was that ChatGPT Ads had produced a small but genuinely incremental stream of new patients that her referral network and Google Business Profile listing weren't reaching — enough to convince her to renew the engagement engagement for a further quarter, though not yet enough volume to consider a engagement upgrade, since her patient intake capacity at a single-consultant-heavy clinic was itself a constraint on how much volume made sense to chase.
One pattern worth noting for any similarly sized clinic weighing this channel: the service-type tagging built into the reporting layer, even without separate campaign infrastructure, let ICG's account team spot that a specific sub-segment of cosmetic conversations — users asking specifically about pigmentation-focused treatments rather than anti-ageing or hair-removal procedures — converted at a noticeably higher rate than the cosmetic category average. That granularity, gathered cheaply within the single bucket rather than through a dedicated pigmentation-treatment landing page the budget couldn't yet justify, became the basis for a month-four content adjustment to the shared landing page, adding a dedicated section addressing pigmentation treatment specifically. The change lifted the page's conversation-completion rate for pigmentation-tagged traffic without requiring any additional media spend, illustrating a broader point about engagement accounts: the constraint is usually landing-page and campaign infrastructure, not the underlying signal, which a well-tagged single bucket can still surface with enough clarity to act on.
The clinic's two junior consultants, who between them handled the bulk of medical dermatology consultations, also reported a qualitative shift worth recording even though it doesn't show up cleanly in the conversion numbers — patients arriving from ChatGPT Ads conversations tended to ask more specific, better-informed questions during their first consultation than patients arriving through referral or walk-in, consistent with the idea that a completed conversation with an AI assistant beforehand already resolves a good deal of the basic informational back-and-forth a first consultation would otherwise need to cover. Neither junior consultant could say definitively that this translated into a shorter average consultation time, but both felt the first-visit conversations were more productive, a soft signal the clinic owner factored into her decision to renew rather than treating the quarter purely as a cost-per-lead exercise.
What we'd do differently next time
The conservative late-intent bid ceiling ICG set in month one, while the right cautious starting point given the tight Rs 40,000 budget, likely cost the clinic a handful of genuinely high-value late-intent conversations in the first few weeks — users who were close to a booking decision but whose conversation-completion bid the account wasn't yet willing to match. Setting that ceiling marginally higher from week one, even at the cost of exhausting budget slightly faster in the early data-gathering period, would probably have captured a few of those conversations without materially changing the account's overall learning curve.
The single-bucket structure covering both medical and cosmetic dermatology worked reasonably well at this budget level, but the manual monthly reconciliation against the clinic's appointment book introduced a lag that made mid-month bid adjustments harder to justify with confidence. A lightweight, low-cost tracking integration — even something as simple as a shared spreadsheet the front desk updated weekly rather than monthly — would have given ICG's account team a faster feedback loop without requiring the clinic to invest in full practice-management software integration it wasn't ready for.
Finally, the before-after language rewrite for cosmetic procedures happened during the first month of live review rather than before launch, because the clinic's existing Instagram-derived marketing language wasn't fully audited during onboarding. As with any dermatology account carrying a meaningful cosmetic-procedure component, front-loading that specific compliance review before the campaign goes live — rather than catching it in the first live review cycle — is now standard practice ICG applies earlier in subsequent single-location dermatology engagements.
How this maps to your own clinic
If you run a single-location dermatology or cosmetic-adjacent practice on a comparable budget, this scenario's structure — one shared conversation bucket covering your full service mix, a conservative early-to-late-intent budget split, one lean landing page, simple monthly reporting — is the realistic engagement shape to expect, not a scaled-down version of a multi-branch or multi-specialty build. The volume ceiling a tight budget imposes is real, and the honest expectation to set going in is incremental, complementary lead flow alongside your existing referral and Google Business Profile channels, not a replacement for them.
If your practice runs a heavier cosmetic-procedure mix than this scenario's clinic, expect the compliance review on outcome and before-after language to be proportionately more involved — NMC Section 6 and ASCI Chapter III apply with particular weight to exactly the procedures (laser treatments, peels, anti-ageing work) that make up a larger share of cosmetic dermatology revenue.
