Dr. Suganya’s first question, when I walked her through the project scope, was about the ad budget.
“Will I need to keep paying for Google Ads alongside this?” she asked. She’d been running Google Ads campaigns with a prior agency and had gotten some appointment inquiries but nothing consistent.
I told her the system we were building would place her content in front of patients through organic search rankings, not paid placement. She looked at me the way people look at someone who’s just said something that sounds too good to be true.
“Let me show you what the results look like after five weeks,” I said. “Then you can decide whether it works.”
This is the story of those five weeks.
What Fertilia Health Needed
Dr. Suganya runs Fertilia Health, a women’s health and fertility practice. She sees 20-30 patients a day, which means she has deep clinical expertise and almost no time to put it online. Patients searching for her specialty, PCOS, hormonal imbalances, fertility concerns, were finding generic health sites, not her practice, because she had no Google presence.
She’d tried a content agency the previous year. They published a handful of articles over three months and generated zero appointment requests from organic search. She wasn’t sure the problem was solvable without sustained ad spend.
The brief I got was specific: generate consistent organic patient inquiries, don’t require Dr. Suganya to write anything, and don’t depend on ads to keep working.
Week 1: Infrastructure and the Calibration Problem
The first week was setup. Configuring the keyword research pipeline, calibrating the content generation prompts, building the review queue. We published nothing publicly. From Dr. Suganya’s view, nothing visible was happening, and I had a mildly uncomfortable call on Thursday where she asked if we were on track.
We were. But the first batch of content drafts was not good.
The initial drafts were structured like patient information leaflets: technically accurate, formatted in a way no actual doctor would write. Dr. Suganya reviewed seven of the first twelve and flagged them as “not how I’d say it.” We spent three days rewriting the prompt logic with her input, using her phrasing from prior patient handouts and clinical notes as the calibration source.
This is the part that doesn’t show up in clean case study summaries. Week 1 is almost always about finding the voice. If you’re evaluating a content engine for your business, the calibration week isn’t overhead. It’s the most important work in the project. I now budget it explicitly and tell clients not to judge anything until week 2 is half done.
Week 2: First 40 Posts, First Impressions
By end of week 2, we had 40 posts published. Google Search Console showed 217 total impressions for the week, spread across roughly 30 different queries. No clicks yet.
Dr. Suganya checked the report I sent on Friday. “217 doesn’t sound like much,” she said.
I told her 217 impressions in week 2 for a brand-new domain publishing medical content was ahead of projection. Fresh domains typically sit at zero impressions for the first 10-14 days while Google’s crawlers work through the sitemap. We were four days ahead of the expected indexing timeline.
She accepted this. She didn’t look reassured, which is normal at week 2, and I didn’t try to over-promise what week 3 would look like.
Week 3: First Clicks, First Calls
Impressions for week 3 jumped to 1,840. More importantly, nine patients clicked through to the site from Google. Two of them booked consultations.
The post that drove the first click was targeting a query cluster around signs of hormonal imbalance, with about 880 monthly searches from Indian patients. Not a post we’d planned specifically, but one the keyword system flagged as low-competition with strong patient intent.
Dr. Suganya called me that week. She’d searched a PCOS-related term and found one of her posts on the first page. “I didn’t even remember approving that one,” she said. “I reviewed it and moved on. I didn’t think Google would surface it this quickly.”
That call is one I think about when clients ask whether this kind of system actually works. Before week 3, Dr. Suganya was tolerating the process. After week 3, she was invested in it.
Week 4: The Review Bottleneck
We’d planned for Dr. Suganya to review 5-7 posts per day. The reality: on a full clinic day, she could manage 3-4, and sometimes zero. The queue backed up.
I redesigned the review routing mid-sprint. Posts now get categorized based on clinical complexity. A post about lifestyle changes for PCOS goes through a light review, quick approval, fewer clinical claims to verify. A post about treatment protocols or medication interactions goes through a full review. This halved the average review time without reducing the quality gate for high-stakes content.
We finished week 4 at 82 published posts, slightly behind the target of 90. Impressions for the week came in at 3,600.
Redesigning mid-sprint is something I avoid when I can. But the alternative was Dr. Suganya missing review days because the queue felt overwhelming, which would have slowed the output more than the redesign did. Sometimes the right call is to change the system rather than push the person through a process that doesn’t fit their actual capacity.
Week 5: 5,000 Reached, #2 Ranking
The final week closed at 5,200 weekly impressions. 102 posts published in total. For a locally-targeted fertility search query, Fertilia Health ranked at position 2, below one large hospital system and above every generic health information site.
I sent Dr. Suganya the Search Console data on the Friday call. She looked at the impression graph, the ranking positions, the weekly trend.
“And this keeps running without ads?” she asked.
“It’s running now,” I told her. “We keep publishing and reviewing. The rankings compound.”
What Happened After Week 5
The more useful story is what happened in weeks 6-12, because that’s where the system proved it was durable and not just fast.
Total consultation clicks from organic search reached 109 by week 8. The site now gets 40+ monthly visits from ChatGPT, which means Fertilia’s content is being cited by AI tools when patients ask about fertility symptoms and treatment options. That’s a compounding effect that didn’t exist in the brief and wasn’t part of the original pitch. It happened because well-structured, specific content is what AI search tools cite: patient-intent language, clinical specificity, consistent publishing cadence.
Dr. Suganya now spends about an hour per week on review. The keyword research runs automatically. New posts go live daily. The only thing that changed from week 5 is that her weekly report check takes less time because she knows what to look for.
The full technical breakdown of the three-layer system, the keyword pipeline, the review queue design, and what the content strategy looked like at the query level, is in the SEO for Doctors case study.
For the 5 content workflow patterns that turn a publishing system into a qualified pipeline (intent clustering, ICP stage mapping, closed-loop attribution), see AI Content Marketing: 5 Workflows That Drive Pipeline.
What Transfers to Non-Healthcare Businesses
The Fertilia case comes up in conversations with founders from other industries often enough that I want to address what actually transfers.
The calibration week is not optional for any business. The difference between a content engine that sounds generic and one that sounds like your business is in week 1 prompt work, the feedback loop with whoever owns the voice, and the editing on early drafts. This is as true for a SaaS company as for a medical practice.
Week 2 looks like nothing. It isn’t. The indexing lag is real. If you evaluate the system in week 2, you’ll cancel something that’s working. Set expectations before you start: impressions typically come in weeks 2-3, clicks in weeks 3-4, qualified conversions in weeks 4-6.
A human review step isn’t always required, but when it is, build around real capacity. Dr. Suganya can review four posts on a full clinic day. Not seven. Designing for seven guaranteed a bottleneck. If your team has a subject matter expert who needs to sign off on content, figure out their actual daily ceiling before you set a publishing cadence.
Attribution is the part most teams skip. The 109 consultation clicks come through from organic search. The ChatGPT citations show up as referral traffic from chatgpt.com in PostHog. Neither of these requires paid attribution infrastructure, but you need PostHog or GA4 properly set up before the system goes live, not after. We built this into the Fertilia project from day one.
FAQ
How much does an AI content engine like this cost?
Our AI Content Engine starts at $500/month for the first three months, then $2,000/month. That covers the full system: keyword research, content generation, quality gates, publishing, and the SEO feedback loop. No ad spend is required because the system generates organic traffic, not paid traffic. The first three months at $500 is intentionally discounted: it’s the calibration period, where we’re building and tuning the system alongside you, not just running a production pipeline.
How long before we see results like Fertilia’s?
Fertilia’s 5-week timeline was at the faster end of what we see. An established domain with some existing Google authority typically gets meaningful impressions in weeks 3-4. A brand-new domain takes weeks 5-7 for first impressions. Clicks and conversions follow 2-4 weeks after impressions stabilize. I tell clients: don’t judge the system before week 8. Everything before that is setup and warm-up.
What’s the minimum time commitment from our team?
For Fertilia, Dr. Suganya spends about one hour per week on post review. For non-medical or non-regulated content, you can reduce that further by letting quality gates handle more of the review automatically. The minimum we recommend for any business is a weekly 30-minute check on the Search Console report and the content queue. Less than that and you lose visibility into what’s ranking, which is the signal that tells you what to produce more of.
Does this work for industries other than healthcare?
Yes. The Fertilia case gets referenced because the results are specific and the client approved being named publicly. The same architecture runs our own blog at Kalvium Labs, two posts per day in the AI development space. It works for professional services, SaaS, ecommerce, and any business where content SEO is a viable acquisition channel. Healthcare is actually harder to run than most industries because of the physician review requirement. Most other businesses can automate more of the pipeline.
Why does “zero ad spend” matter for the ROI case?
Ads stop working when you stop paying. Content assets compound: a post that ranks today still drives traffic 18 months from now, with no ongoing spend. The 5,200 weekly impressions from Fertilia’s week 5 are still active impressions now, generating consultation clicks without a monthly fee attached. That’s the compounding property of organic search that paid placement doesn’t have. If you’re evaluating content versus ads as a primary channel, the right question isn’t which one works faster. It’s which one builds equity.
If you’re running a practice or business where patients or clients need to find you online, the system we built for Fertilia is now available to other businesses. Book a 30-minute call and we’ll tell you what the first five weeks would look like for your specific situation.
