Med Spa Marketing: The System I Run For Clinics

1. Score the 4 chokepoints before you spend anything

Pull one number per stage for last month. Attraction: new people who reached the clinic, and from where. Conversion: how many became a lead with details in the CRM. Closing: how many booked and paid a deposit. Fulfillment: how many showed, and how many came back.

The biggest drop between 2 numbers is your chokepoint. It's the only thing you work on this month. Why: most clinics assume the problem is Attraction and are leaking at Closing, so they buy more traffic for a leaking pipe.

2. Build the offer before you touch the ads

Nobody wakes up wanting micro-needling. They wake up hating something about their face. The strongest practitioner in this space, in his 2025/26 update, weights the offer far above creative, copy and targeting. His argument is the strongest in the space, though it's one voice, so I use the ranking and leave his percentages out. Treatments are ingredients, and ingredients are commodities. Sell the cake, not the flour.

Build it: start with the patient's problem. Name the package like a makeover. Include an assessment, but don't call it a consult. Add one entry-level treatment as the carrot and 2 or 3 bonuses that are cheap to deliver. Total the honest retail value, then price at 10 to 50 percent of it. Get every claim approved by the clinician.

Never discount one treatment 80 percent. It erodes the price you charge for it every day. The offer builder prompt below does the math.

3. One CRM, and every lead lands in it with a source

Before any ad spends, send a test lead through every door: ad lead form to CRM webhook to pipeline stage to notification. Instagram gets a comment keyword that opens a DM and lands in the same CRM. The website form or quiz posts there too. Every lead carries a source and a tag, or you can't score chokepoint 1 again. Here's how to build the GoHighLevel workflows.

On ad settings, the sources I read agree on one thing: keep interest targeting broad. Stacked interests starve the algorithm on a local budget. The specific budget, radius and age rules you'll read are one practitioner's starting point, not a platform minimum. Here's how I use AI to run the Meta side.

# test the front door before spending a dollar (CRM inbound webhook)
curl -s -X POST "$CRM_INBOUND_WEBHOOK_URL" \
  -H 'Content-Type: application/json' \
  -d '{"first_name":"Test","last_name":"Lead","phone":"+15550000000",
       "email":"test@example.com","source":"meta-leadform",
       "tags":["new-lead","test"]}'

# then confirm all four, by hand, once:
#   1. contact exists    2. pipeline stage set
#   3. notification fired 4. no medical field anywhere on the form

4. The setter: reply in minutes, qualify, book, then deposit

Speed is the cheapest lever clinics never pull. The speed-to-lead figures quoted in this space rarely trace back to a primary study, so I don't repeat them. The direction is consistent across every source I read: minutes, not hours.

Illustrative math, not a benchmark: 100 leads a month, 10 percent book at a $2,000 lifetime value is $20k. Reply fast enough that 20 percent book and it's $40k. A part-time setter costs a fraction of that gap.

Book first, deposit after. Someone who already picked a slot pays far more often. In the setter I run for a client, the deposit sequence is: link with a 2 hour hold, reminder at 1 hour, last chance at 30 minutes, then the slot releases. The Instagram DM automation guide covers the plumbing.

5. Put handover rules in code, and train the front desk

Rules in code, not the prompt. 3 rejected slots or a refused deposit: handover tag, bot stops. Stop is terminal. One owner per conversation: the front desk takes over with a tag the bot checks before every send. A booking counts only when the CRM calendar says so.

The front desk converts calls. Answer by the second ring. Lead the call: "this is Kelly, how can I help you today," not "ABC Med Spa, this is Kelly." Use their name. Treat a price call as an opening. Never hold twice. Don't prejudge what someone can afford. Between yes and no there are several stops: two-pay, financing, fewer sessions, one session, product only, a follow-up visit, a follow-up call. If most calls don't convert, fix the training, not the ads.

6. Engineer the show rate

People no-show for 2 reasons: fear outweighs desire, and no skin in the game. Answer the fears in your confirmations: being pressured to buy, a sketchy clinic, looking fake, pain, lasers burning. A named package beats a bare consult, and your Google reviews get read before the visit.

Then take the deposit as a choice. Not "would you like to pay by card?" but "would you like to pay over the phone or through text?" That swap is the highest-leverage detail here.

CONFIRMATION SEQUENCE (after the slot is booked)

1. On the call/chat  : get them to say the date, time and location out loud
2. Immediately       : text with the full address + a photo of the building
3. Immediately       : calendar invite they can accept
4. 24h before        : NOT "are you still coming?"
                       instead: "do you want bonus A or B?
                       A is better for dry skin, B for fine lines"
5. Deposit hold      : link (2h) -> reminder (1h) -> last chance (30m) -> release

Why 4 works: a yes/no question invites no. A choice invites engagement.

7. Run 2 loops: reactivation and referral

The cheapest patient is one you already have. The next cheapest is one they bring. And your email list is the only channel you own.

Reactivation is its own route with its own offer, consent check and outcome, never mixed into new leads. A segment that hasn't visited in a set window, a specific reason to return, the same booking flow, and a hard exclusion for opt-outs. The owner approves the offer and copy once. My cadence is how I set it up, not a benchmark.

Referral: work out what a new patient costs all-in (ad spend plus discount, divided by patients who bought). You can likely afford a serious incentive. One practitioner reports that a symmetric give-and-get at a serious amount beat smaller splits in his tests. Treat that as one voice and test your own number. Ask at checkout after a good result, not in an email 6 weeks later, and get the case study at the same moment (with consent, see step 8).

8. Hold the compliance line

Across the 9 videos I read, compliance content totaled a few SMS tips and one line about Google certification. That gap is the most dangerous thing in this category, so here's the version an MD writes.

Lead forms and chats never collect medical details. No conditions, medications, photos of the area, or "what are you being treated for." Intake happens on your approved path after booking. A DM thread isn't a medical record and isn't access-controlled. Don't paste patient medical details into ChatGPT or any other AI chat either. Share operational counts only, de-identified.

Before-and-afters need documented consent for that specific use, not a general clinic waiver. For HIPAA-covered practices, HHS marketing guidance generally requires the patient's written authorization before their information is used in marketing. Confirm the specifics with your counsel.

Medical claims in ad copy get clinician approval before they run. Every time, including the Canva static at 11pm.

SMS needs a real opt-in at collection, plus a clear opt-out and sender identification in the first message, per the carrier industry's CTIA Messaging Principles, which also expect STOP to work. Some advice in this space routes people to "reply NO THANKS" instead of STOP so carriers don't flag the number. Don't. Deliverability isn't a reason to hide an opt-out.

Google restricts ads for prescription drugs and prescription drug services, and several categories require certification. Botox and weight-loss drugs are prescription products. Check the current Healthcare and medicines policy before you write the campaign.

Refuse the rest of the playbook, whoever runs your marketing: discounts on treatments you sell full price, bots written to pass as human, and medical questions in a lead form to improve lead quality. The offer, the pricing and every medical claim stay with the clinician, even with an agency.

9. One weekly decision and one daily page

Weekly, on a fixed day, every ad gets a dated line: scale, hold, refresh or kill, with the number. My rule, not Meta's: raise a strong ad 20 to 30 percent, never double overnight. Meta gives no percentage. It says large or frequent budget changes can send an ad set back into the learning phase, and my step size is how I stay clear of that. Run a Meta ads audit before you scale anything. Falling click-through or rising frequency means refresh the creative before adding budget. Never let one campaign carry the clinic.

Daily, one page for the owner: new leads by source, conversations waiting on a human, calendar-verified bookings, deposits, no-shows, pipeline bugs with a day counter. Every conversation with no owner gets a named person that day.

The owner keeps 4 things: the offer, pricing, medical claims, and approving the setter's first 10 sends one at a time.