Doctorate-level credentials behind a front door that doesn't open.

A market scoring 4 out of 100, and a website nobody can reach

So Acupuncture has a credentialed practitioner, a genuine second-generation Corona legacy, and a home market that is close to undefended. What it does not have is a website search engines can read — or, for a meaningful share of visitors, a website that loads at all.

That is not a figure of speech. https://soacupuncture.com — the domain in the email address, the one a patient would type — returns HTTP 522 after a 19.5-second hang, on every attempt. The plain-HTTP version returns 404. Only the www version resolves. Anyone who types the domain without www waits twenty seconds and receives an error page.

[ Data verified ] Live HTTP inspection of soacupuncture.com (headers, redirect chains, raw HTML), NPPES NPI Registry API, Search Atlas keyword API, Yelp listing data, competitor sites fetched individually — pulled July 19, 2026.

Beneath the domain failure, the site runs on Google Sites, which ships with zero structured data, zero meta description, zero Open Graph tags, zero canonical tag and zero H1 elements. The page title reads "Dr. Jennifer So, DAOM, MSAOM, LAc." — no city, no service. Nobody in Corona searches that string.

The seven questions, answered

A second-generation Corona practice, invisible by platform

Verified through the federal NPI registry: Dr. Jennifer So, DAOM, L.Ac.; NPI 1083466825, active; taxonomy 171100000X (Acupuncturist); California license 19465; 591 N McKinley St Ste 104, Corona CA 92879; (951) 268-0208. The address and phone on the federal record match the website exactly — a clean starting point most practices do not have.

The legacy asset nobody can see

Dr. So's site states her father founded Corona Acupuncture and describes it as the first licensed acupuncture clinic in the area. The predecessor business and its closure are independently confirmed. The "first licensed" claim is self-reported.Needs confirmation

This is a real differentiator in a market where the two highest-review competitors are chiropractic offices that added acupuncture as a service line. A second-generation, doctorate-level, acupuncture-first practice is a distinct story. It currently appears in no title tag, no meta description and no schema — because the platform cannot carry any of them.

Credentials — an important distinction

The DAOM, MSAOM and L.Ac. credentials and California license 19465 are corroborated through the federal registry. Those are safe to lead with anywhere.

Separately, the site states Dr. So is a trained Corona CERT disaster volunteer who has served with Acupuncturists Without Borders, the Integrative Healers Action Network and World Central Kitchen, and holds her degrees from South Baylo University with honors. All of these are self-reported on her own site; no third-party page on any of those organizations names her.Needs confirmation They are excellent trust material — but they need to be documented, not merely asserted, before they carry weight with a patient or a search engine.

The home market is undefended — and Riverside is four times the size

All figures below are live keyword data pulled July 19, 2026. Difficulty runs 0–100; lower is easier to rank.

[ Data verified ] Search Atlas keyword API, July 19, 2026. Volumes are US monthly averages.

What the table actually says

• The home market is nearly undefended. A difficulty score of 4 on a commercial term carrying a $4.25 cost-per-click is rare. That $4.25 is the market stating what a Corona acupuncture click is worth — advertisers are paying it today. Ranking organically means acquiring the same click at no per-click cost.
• Riverside is four times the volume and still rated Easy. The stated goal of reaching patients "in or close by my city" is well-founded by the data. Riverside is the prize.
• Norco is a trap, and the data shows why. Generic "acupuncture norco ca" draws 12 searches a month — but the branded query "zen acupuncture norco" draws 70. One competitor owns that town's demand by name. Serve Norco with a city page for completeness; do not treat it as a growth channel.
• The national-versus-local split is the strategic insight. "Acupuncture for back pain" nationally scores 86 — Super Hard. The same intent attached to a city name is wide open. The play is never national condition content; it is condition content welded to geography.

What that demand is worth

Assumptions stated: average initial visit $135 and follow-up $95 — the only published pricing in this market, from the one competitor who publishes rates. Average course assumed at six follow-ups after the initial, giving $705 per patient episode.Needs confirmation Capture applied to combined Corona + Riverside + Eastvale volume of roughly 948 searches per month, at maturity.

These scenarios move only on the assumptions stated above, and they assume the technical foundation is repaired first. At present the capture rate is constrained by a front door that does not open.

A local field with reviews but no pages worth ranking

Review benchmark — the starkest number in this assessment

On the So Acupuncture figure: the Yelp listing exists and shows 25 photos, but its page title renders without the "& N Reviews" segment that all seven competitors display — which is how Yelp renders a listing with zero recommended reviews. Treated here as at or near zero, pending a manual check.Needs confirmation

What competitors have and have not built

The finding that reframes the whole market

For "acupuncture for fertility Riverside CA," the top-ranking true condition page is riverside-acupuncture.com/infertility-and-ivf. A direct fetch of that clinic's own location page places it at 453 Cherry Street, Columbia, Pennsylvania.

A Pennsylvania clinic outranks every acupuncturist in Riverside County for fertility acupuncture in Riverside — on the strength of a matching domain name and one condition page.

The pattern repeats: Irvine and Newport Beach clinics rank for Corona migraine and anxiety queries, and a Glendale practice runs programmatic city pages at /acupuncture/corona and /acupuncture/eastvale into this market from roughly forty miles away. Out-of-market operators are harvesting Corona and Riverside patients with basic page structure, and the local field has not responded because the local field has not built anything.

The honest balance sheet

Two problems, both severity one

Problem two — the platform cannot do SEO. The site is built on Google Sites. Direct inspection of the served HTML found:

Google Sites does not expose controls for most of these. This is not a configuration oversight to be corrected — it is a ceiling imposed by the platform. No amount of on-page work raises it. The site has to move.

Targeting one search: her own name

The title tag "Dr. Jennifer So, DAOM, MSAOM, LAc." targets exactly one query — a search for her by name, which only existing patients perform. It omits "acupuncture," omits "Corona," and omits every condition she treats.

The consequence is precise and measurable: the 183 monthly searches for "acupuncture corona ca" cannot find this page, and the difficulty-4 rating means those searches are being lost on a term almost nobody is defending.

• Metadata: no meta description exists on any page, so search snippets are auto-generated from whatever text can be scraped — the value proposition is not controlled at the exact point where a patient decides which listing to click.
• Schema: none present. For a local medical practice this is the single largest structural gap — see §8.
• Content architecture: no dedicated service pages, no condition pages, no city pages. The market opening documented in §4 requires all three.
• Measurement: Google Analytics 4 is installed, which is a real head start. Search Console verification, conversion tracking and call tracking still need to be wired up so that the work can be measured.

Described to patients in words she doesn't control

Ask an AI engine "who is the best acupuncturist in Corona, CA" and it has to answer from something. Our sampling of that exact query found the answer assembled almost entirely from directory listings — eight of ten cited sources were Yelp, Healthgrades, WebMD, HealthProfs, Thervo, Thumbtack or ThreeBestRated.

The clinics named in those answers were described in third-party words, drawn from directory listicles rather than from the clinics' own sites — because none of them have published anything a machine can read.

Why schema is the fix. For a local medical practice, MedicalBusiness and Physician schema is how a search engine learns — in machine-readable form — the practice name, address, phone, hours, practitioner credentials, license, accepted insurance and service area. With zero schema present, every one of those facts has to be inferred from a JavaScript-rendered page that has no H1.

This matters more every month. AI answer engines resolve local questions by reading structured data and citing sources. There is a real first-mover position available here, because no clinic in this market has claimed it.

Present on five directories, absent from the ones that matter

The pattern: every owned social profile was claimed and then never used. Zero posts, zero videos, zero followers. These are not dormant accounts waiting to be revived — as they stand, they signal abandonment to anyone who finds them. Better to run one channel properly than four badly.

Name and address consistency

• The practice name appears in five different forms across her own properties — "So Acupuncture," "SO ACUPUNCTURE," "SO Acupuncture," and two credential-led variants. No canonical entity name is in consistent use anywhere.
• A conflicting suite number is in circulation. The contact page adds "inside Phenix Suites (Office #127)" while the address of record is Ste 104. Two suite numbers on one record is a genuine citation risk — aggregators can and do ingest the wrong one.
• The practice address is a shared salon-suite location. 591 N McKinley St is Phenix Salon Suites, confirmed on the practice's own contact page. Shared-suite and co-working addresses sit in a difficult area of Google's location guidelines — they carry real profile eligibility and re-verification exposure, and they tend to suppress local ranking quietly rather than visibly. This deserves a direct answer before any local campaign is built on top of it.
• Do not inherit the predecessor's citations. Corona Acupuncture is a different business at a different address, now closed. Conflating the two would corrupt the citation profile.

Every claim verifiable in one click — that's the goal state

The stated goal was conversion. Start where the funnel breaks.

Dr. So's own words were "increase conversion." The analysis has to begin at the top of the funnel, because that is where it is severed.

No on-page conversion work matters for the share of visitors who type the domain without "www" and receive a twenty-second timeout. That is the first conversion fix, and it is an infrastructure fix rather than a design change.

Below that, ranked:

1. No structured data — the practice cannot surface with hours, phone or rating in the rich formats that drive local click-through.
2. No H1 and no meta description — the value proposition is not controlled at the moment a patient chooses which search result to click.
3. Insurance status is unaddressed — and it is the highest-converting filter in this category. The one competitor who publishes a full carrier list names Aetna, Blue Shield, UnitedHealthcare, Cigna, VA/TriWest, workers' compensation and three personal-injury carriers. A patient's first question is "do you take my insurance." A site that does not answer it loses the visit.
4. Zero visible reviews — the strongest conversion asset in local healthcare is simply absent. A five-star practice with no reviews converts worse than a 4.3-star practice with sixty.

Online booking vendor, whether the phone is a tappable link, and form destinations could not be fully enumerated — Google Sites renders those elements client-side.Needs confirmation None of them change the priority order above.

Three referral channels, claimed by nobody

Personal injury and letters of protection. Medical liens let accident victims receive care with no upfront payment, repaid at settlement, and licensed acupuncture for pain management is accepted on lien in this region. Corona has dense personal-injury attorney infrastructure — one firm's Corona page explicitly states it helps clients obtain acupuncture after injury. No acupuncture clinic in this market mentions liens or letters of protection at all. Note that these arrangements are contractual rather than statutory, and terms vary; the size of this channel is not quantified and no number should be attached to it.Needs confirmation

California workers' compensation. Acupuncture is covered under state law, capped at 24 visits per industrial injury. One competitor mentions workers' comp; none market to it.

Veterans and TriWest. Two competitors mention TriWest; one built veteran-first positioning around it — on a site with no HTTPS, last updated in 2018. Corona has a substantial veteran population and the channel is demonstrably active locally.

Where automation earns its place

• Review requests triggered automatically after each visit — the single highest-return automation for this practice.
• New-patient intake and follow-up — instant confirmation, pre-visit paperwork, and a rebooking nudge at the end of a treatment course.
• Insurance pre-verification — answer the highest-converting question before the patient has to call.
• Recall campaigns for lapsed patients, segmented by the condition originally treated.
• Content repurposing — each condition page becomes a set of social posts, in a market where nobody has published since 2016.

A low score for a reason that is unusually good news

The current score is low, and it is low for a reason that works in Dr. So's favour: almost all of it traces to platform and configuration rather than to the practice itself. A broken domain record, a page builder that cannot emit schema, and an unused review pipeline are all fixable inputs. The credentials, the clinical differentiation and the family legacy — the parts that are genuinely hard to build — already exist and are simply unexpressed.

A 24 caused by a bad platform is a considerably better position than a 45 caused by a saturated market. This market is not saturated. It scores 4 out of 100.

The first five dimensions score current state (higher = stronger today). The last three score headroom (higher = more opportunity available to capture). Scores are directional, built from the verified findings in §6–§12.

From unreachable to ranked and cited

What to do first — ranked by impact

The first five moves

• Repair the domain — the front door does not open, and every day it stays shut costs patients.
• Get off Google Sites — no ceiling can be raised from underneath.
• Ship schema and a real title tag — become legible to Google and to AI answer engines.
• Start collecting reviews this week — the gap against the field compounds daily.
• Publish the insurance page — answer the question every patient asks first.