Case study · Healthcare and clinics
Vein Clinic · Chicago
A clinic spending roughly $20,000 a month on Google Ads did not have a traffic problem. It had a dependency problem, and the fix was to earn the visibility it was renting.
How does a vein clinic reduce its ad spend without losing patient enquiries?
By replacing the part of the ad budget that was buying visibility the clinic could have earned. A vein clinic that ranks organically and appears in map results for its treatment terms does not need to bid on those same searches at full rate.
In this engagement, monthly ad spend came down from about $20,000 to about $7,000 while qualified enquiries kept arriving. That took a year of organic work, tighter targeting on what remained in the paid account, and conversion work on traffic the clinic was already paying for.
Renting every patient, every month
The clinic was spending roughly $20,000 a month on Google Ads and wanted more qualified leads at a lower cost.
That is a sentence worth sitting with, because it describes a position a lot of specialist clinics end up in without ever deciding to. Paid search works. It works from the first day. So the budget goes up, the phone rings, and nobody builds anything underneath it. Two years later the practice has a functioning patient pipeline that costs five figures a month and stops the moment the card is declined.
Read the full breakdown: Renting every patient, every monthHide the full breakdown: Renting every patient, every month
Vein care makes this worse than most specialties for three reasons. First, the patient journey is long: people live with heaviness, aching and visible veins for years before they search. Second, the searches split into two completely different intents, a symptom search and a treatment search, and they behave nothing alike. Third, the category attracts aggressive bidding from med spas and multi-state groups who will happily pay more for a click than a single-site clinic can justify.
So the clinic was in an auction it could not win on price, buying clicks from people who were often years away from booking, and had nothing organic to fall back on. The ask was not really for cheaper clicks. It was for a patient source that did not switch off.
What we looked at before touching the budget
Cutting spend on a working account is easy and usually stupid. The first job was to find out which part of the $20,000 was doing work and which part was buying searches the clinic should have owned for free.
The search terms report, not the keyword list
The keyword list tells you what someone intended to buy. The search terms report tells you what Google actually charged for. We read it in full, at the query level, and sorted every term into three buckets: people describing a symptom, people naming a treatment, and people who were never going to be patients at all.
- Symptom searches: high volume, long consideration, poor immediate conversion
- Treatment searches: lower volume, far higher intent, much higher competition
- Irrelevant traffic: research, imagery, cosmetics, out-of-area, job seekers
Which of those searches the clinic already ranked for
We cross-referenced the paid search terms against organic position. Wherever the clinic was already on page one organically and still bidding, it was paying twice for the same visit. That overlap is the cheapest saving in any mature account and almost nobody audits for it.
What happened after the click
We walked the landing experience the way a patient does: on a phone, on a mid-range connection, with a specific worry in mind. Where does the page answer the worry? How far down is the phone number? Is the form asking for insurance details before it has earned the right to?
See the remaining steps: What we looked at before touching the budgetHide the remaining steps: What we looked at before touching the budget
Whether anyone knew which ads produced patients
Most clinic accounts count form submissions and call clicks and stop there. That measures the website, not the practice. We checked whether a booked consultation could be traced back to the campaign that produced it, and in this case the trail ran out at the front desk.
What the clinic looked like outside Google Ads
Organic footprint, map presence, the profile, the citations, and what an AI assistant said when asked where to go for vein treatment in the area. That last one is not a novelty any more. It is increasingly the first thing a patient sees.
The picture that came back was consistent: a well-run paid account carrying the entire weight of a practice that had no other visibility to stand on.
Where the money was going, and where it went instead
This is a description of the change in approach, not a performance table. The only figures published anywhere on this page are the ones the engagement actually produced.
| Item | Before | After the work |
|---|---|---|
| Source of enquiries | Paid search, almost exclusively | Paid search, organic, map results and AI answers |
| Bidding on terms the clinic ranked for | Yes | No |
| Symptom-stage searches | Bought at full price in the auction | Answered with content that earns the visit |
| Landing experience | General service pages | Pages matched to the specific search |
| Enquiry handling | Manual, during office hours | Captured and qualified automatically, at any hour |
| What happens if the budget stops | The pipeline stops | The earned visibility keeps working |
The plan, in the order we ran it
The sequence matters more than the list. Reducing spend before the organic side can carry any load is how clinics end up with a quiet quarter and a lost year.
Build the earned layer first, and change nothing about the budget
SEO, AEO and GEO went in together: pages that answer the symptom-stage questions in plain language, structured so that a search engine and an AI assistant can both lift a clean answer, and marked up so the clinic reads as a specific medical business in a specific place. For the first stretch the ad budget was left exactly where it was. Nothing was cut on a promise.
Fix the conversion rate on traffic already being paid for
CRO came next because it is the only lever that improves paid and organic at the same time. Every enquiry the existing traffic was failing to produce was a patient the clinic had already bought. Tightening the path from arrival to enquiry made every subsequent decision cheaper.
- One clear next action per page, visible without scrolling on a phone
- Forms that ask only what is needed to book, and nothing else
- The phone number treated as a primary conversion, not an afterthought
Tighten the paid account toward high intent
With the earned layer starting to carry symptom-stage demand, the Google Ads and Meta Ads accounts were narrowed onto searches where someone was choosing a provider rather than researching a condition. Negative keyword work removed the traffic that was never going to convert. Budget followed intent instead of volume.
See the remaining steps: The plan, in the order we ran itHide the remaining steps: The plan, in the order we ran it
Automate capture and qualification
Enquiries were routed into a CRM that captured every contact, asked the qualifying questions consistently, and made sure nothing sat in an inbox overnight. The clinical team stopped spending time on people who were never going to be suitable candidates, which is a real cost even when it does not appear on an invoice.
Only then, bring the spend down
The budget came down in steps, with the earned channels watched at each stage. That is the entire discipline: spend is reduced as a consequence of coverage being replaced, never as a target of its own.
Counting patients, not clicks
A clinic can halve its ad spend and quietly halve its patient volume at the same time. The only way to know the difference is to measure the thing that matters.
We tracked qualified enquiries as the primary measure, with channel attached, and treated impressions and clicks as diagnostics rather than outcomes. An impression is not a patient. A click is not a patient. A person who asks for a consultation and meets the clinical criteria is close enough to count.
That choice is what made the spend reduction defensible. Because every enquiry carried its source, it was possible to see organic and map results absorbing symptom-stage demand as the paid account narrowed, rather than guessing at it from a monthly total that blends everything together.
Google publishes what its conversion tracking can and cannot see, and the honest answer is that no attribution model captures everything. Some patients see an ad, read three articles, ask an assistant, and then walk in. We measure what is measurable and say plainly where the gaps are.
What changed
Monthly ad spend came down from about $20,000 to about $7,000. Qualified leads kept coming in while the paid budget was cut, and the qualification step was automated so the team spent its time on people who could actually be treated. Those are the three outcomes this engagement reports, and they are the three it is willing to stand behind.
The figure worth understanding is not the 65% reduction. It is the change in what the practice owns. Before, the entire patient pipeline was rented on a monthly basis from an auction the clinic did not control. After, a meaningful share of it came from pages, profiles and answers the clinic owns outright, with paid search doing the job it is genuinely good at: catching people at the moment they choose a provider.
The remaining $7,000 a month is not a failure to reach zero. Paid search at the bottom of the funnel is usually worth buying. What is not worth buying is the top of the funnel, indefinitely, at auction prices, when content can do the same job once and keep doing it.
About these results
The spend figures are the clinic's own account figures, stated as approximate on the original case page and kept approximate here. The 65% reduction is arithmetic on those two numbers and is not a separate finding.
Qualified leads were counted using the clinic's definition of a suitable candidate, not an industry standard. Lead quality definitions are not comparable between practices, so this number should not be benchmarked against anyone else's.
This case does not show what revenue changed, because that was not measured. A clinic that reduces spend and holds enquiry volume has improved its cost per enquiry, which is a real result, but it is not the same as showing profit moved.
Results are from a specific client engagement and vary by market, budget, competition and other factors. They are examples of past outcomes, not a guarantee of future results.
What was actually delivered
The full list, so the outcome can be read against the work rather than against a promise.
- SEO across the treatment and symptom-stage terms the clinic could realistically win
- Answer engine optimisation, so pages resolve a question cleanly enough to be quoted
- Generative engine optimisation, aimed at being a source AI assistants cite for the category
- Google Ads restructured and narrowed toward provider-choice searches
- Meta Ads run alongside, for demand that does not show up in search
See the full checklist: What was actually deliveredHide the full checklist: What was actually delivered
- Conversion rate optimisation on the existing traffic and enquiry path
- Brand visibility work through Branding Signal
- Automated lead capture and qualification through the Dash Funnels CRM
Why the earned layer was worth building
SparkToro's 2026 analysis of Similarweb clickstream data measured how many US Google searches end without anyone clicking through to a website.
SourceSparkToro with Similarweb clickstream data, 2026
For a clinic, the practical consequence is that a page which answers the question inside the results, and a profile that carries the right information, now do work that a click used to do. That is the coverage the paid budget was replaced with.
If you run a specialist clinic on paid search
The transferable lesson is about sequence, not tactics. Almost every clinic in this position is told to cut the ad budget and invest in SEO. That advice is correct and, delivered in that order, it is also dangerous, because organic visibility takes months to arrive and the gap in between is a quarter of missed patients nobody budgeted for.
The order that works is: build first, convert better second, narrow the paid account third, reduce spend last. Each step funds the next. By the time the budget moves, the coverage has already been replaced, so the reduction is a consequence rather than a bet.
Read the full breakdown: If you run a specialist clinic on paid searchHide the full breakdown: If you run a specialist clinic on paid search
The second transferable point is about intent splitting. If your category has a symptom-stage search and a treatment-stage search, they are two different marketing problems wearing the same costume. Buying both through the same channel at the same bid is the single most common way a specialist clinic overspends. Content is the cheaper answer to the first. Paid search is usually the right answer to the second.
The third point is the least comfortable. If reducing your ad spend would end your patient flow within a month, you do not have a marketing programme. You have a subscription. That is worth knowing before a competitor with deeper pockets decides to find out what you will pay.
Spending more on ads than you would like?
Tell us what the account looks like and we will tell you honestly which part of it you could be earning instead.
Straight answers.
How long did it take before the ad budget could be reduced?
The organic side of this engagement was built across roughly a year before the spend came down in steps. The original case page reports the spend outcome without publishing a month-by-month timeline, so we will not invent one.
As a general rule, we do not recommend reducing a working paid budget until earned visibility is demonstrably absorbing part of the same demand. That is usually several months, not several weeks.
Does cutting ad spend always mean fewer patients?
It does if nothing replaces the coverage. Paid search is a tap: turn it down and the flow through that tap drops immediately.
It does not if the searches you stop bidding on are searches you now rank for organically or appear for in map results. In that case you are removing a duplicate cost, not a source of patients. Knowing which of the two you are doing requires the search terms report and organic position data side by side.
What counts as a qualified lead for a vein clinic?
It varies by practice, which is exactly why the number on this page should not be compared to anyone else's. Most vein practices qualify on symptom presentation, insurance situation and whether the person is in the service area at all.
The important thing is that the definition is written down and applied consistently before you start measuring, so that a change in the number reflects a change in reality rather than a change in who was counting.
Is it worth optimising for AI assistants as well as Google?
For a clinic, increasingly yes. A growing share of patients ask an assistant for a recommendation before they ever open a search results page, and assistants tend to draw on the same signals as search: clear pages, consistent business information and sources that actually answer a question.
The practical upside is that the work overlaps heavily with good SEO. The practical caution is that visibility in AI answers is harder to measure and much less stable than a ranking, so it should be treated as an additional surface rather than a replacement channel.
Can you do this without cutting ads at all?
Yes, and for a clinic with capacity to fill that is often the better choice. The same work applied without a budget reduction produces more total enquiries rather than the same enquiries at lower cost.
Which of the two you want is a business decision about capacity, not a marketing one. A practice with an eight-week wait should be reducing spend. A practice with open chairs should not.
Do you publish patient reviews or testimonials for clinics?
We do not publish review or rating data on our own pages, and we do not build review markup out of third-party ratings. Where a client's own rating changed, we say so in plain text and leave it at that.
For clinics we also keep a hard line on patient privacy and on how reviews may be requested. Incentives, gating and anything that filters unhappy patients out of the request are off the table.
Where this comes from.
Primary documentation and published research behind the guidance on this page.
- Google Ads Help: Find out what searches triggered your ad (opens in a new tab)The search terms report, which is where wasted spend is found.
- Google Ads Help: About Quality Score (opens in a new tab)Expected click-through rate, ad relevance and landing page experience, defined by Google.
- Google Ads Help: About conversion tracking (opens in a new tab)The mechanism that connects a click to a booked job.
- eCFR: 45 CFR Part 164, HIPAA Security and Privacy Rules (opens in a new tab)The regulation itself, which governs what a clinic may publish or track.
- FTC: Health products compliance guidance (opens in a new tab)The substantiation standard for health and outcome claims.
Talk to the team
A short call, a look at how the business currently shows up, and a straight answer on what we would do first.