What the data says
before we build

Almost none of it reaches the website we are rebuilding

Pregnancy Choices runs three separate web properties. Nearly all measured traffic goes to the paid landing pages, which are managed by the ad agency and sit outside this project.

Almost every recorded conversion is someone tapping a phone number

Analytics records around 18,000 conversions across the two years. Listing them by name shows what they are.

The device data agrees. On the paid ultrasound page, conversions run at 65% on phones and 1.3% on desktops — a fiftyfold gap on the same page. Phones can dial a number; desktops largely cannot. On the three pages driven by forms rather than phone calls, mobile and desktop rates are roughly equal, which is what a form looks like.

What changed in June 2025

Forms: the number that never broke

Form submissions ran steadily throughout at 1,068 across all three properties. Separating the confirmation pages by property shows where they landed.

Two smaller faults worth fixing

“Get directions” has fired four times in two years. For a walk-in clinic whose main invitation is to come in, that is a broken tag rather than a low number. It is almost certainly missing a trigger on the map and the directions link.

All three properties use a confirmation page called /thank-you, distinguishable only by a trailing slash. It is an easy trap when reading reports. The new site should use a separate address for each form — for example /thank-you/appointment/ and /thank-you/contact/ — so that enquiries can be traced back to the page that earned them.

Paid search brings almost everyone; nearly all of them are on a phone

The mobile-first commitment in the proposal is well supported. In practice this means design review should happen on a phone before a desktop, and that tap-to-call and tap-to-text belong on every page rather than only on the contact page.

Light on content, and competing with itself

A well-developed service page in this sector runs 800 to 1,200 words. The median page here is 208. This is the clearest single reason organic traffic sits where it does: there is not enough on the page for Google to rank, and not enough for an anxious visitor to feel informed.

Twenty-seven English pages overlap with another page on the same topic — five separate pages for men and fathers, three content hubs, two ultrasound pages, two scheduling pages. This shows up in search results: one query returns four different Pregnancy Choices pages at positions 18, 34, 58 and 78, splitting a signal that should belong to one page.

Technical findings

Three different answers, none of them written down

Before we build the location pages, we need one agreed answer to a simple question: which communities does Pregnancy Choices serve? Right now the website gives three.

The Twin Cities metro is a seven-county region of roughly three million people. As a statement of welcome that is generous and true. As a search target for one walk-in clinic it is too broad to rank for, and it competes directly against organisations with a physical location much closer to the person searching.

Please correct anything below that is wrong. This is our reading of the data, not a decision — and the client records will beat our guess every time.

What the geography actually looks like

Five of the eight cities sit within six miles of the clinic. Bloomington is eight. St. Paul and Minneapolis are sixteen and seventeen — and both have two First Care locations inside them. A person in Minneapolis searching “pregnancy centre near me” will be shown the centre two miles away, not the one seventeen miles south. That is not a content problem we can write our way out of.

One question that would settle this

Rather than deciding from search data, the better source is the client’s own intake records. Which cities do clients actually travel from? A year of intake by home city or postcode would rank these communities by real demand and settle the question properly.

It is also worth checking whether any state or county funding carries a defined service territory. If it does, that answer should govern.

Two local listing issues

There are currently three different phone numbers in circulation: 952.234.9850 across the website, 952.234.9854 on the “Text” button in the menu, and 952-295-3287 on the Google Business Profile. Some of this is call-tracking software working as intended, but search engines use consistent name, address and phone details as a trust signal for local ranking. This is worth reconciling deliberately.

The second is review volume, which is one of the strongest factors in whether a clinic appears in the local map results.

The rating of 4.3 is in line with everyone else. The count is not — 25 against 57, 74 and 48. For an organisation with this many grateful clients, a simple, consistent review request at the end of a positive visit is among the cheapest available improvements to local visibility.

Strong on the brand name, close to invisible on everything else

The 1,016 keywords sitting below position 20 are not failures. They are topics the site is already associated with and could compete for with stronger pages.

Where competitors are winning

All eight target city pages already exist on the site, which is a genuine head start. They need depth rather than creation.

52 terms, weighted to the South Metro service area

Tracking starts now rather than at launch, so there is a genuine before-and-after picture rather than a clock that starts on day one.

Set-up: one campaign for mypregnancychoices.com with the location set to Apple Valley, Minnesota, device set to mobile, and a second campaign in Spanish. Local targeting matters here — “near me” terms only return meaningful results when the tracker is anchored to a real place.

Search volumes are Semrush estimates and consistently run well above measured traffic. They are useful for comparing terms against each other, not as forecasts. Blank volumes are terms with no estimate available — usually a sign the term is genuinely local, which is not the same as no demand.

Every page, its search target, and what folds into it

This takes the draft structure and attaches a search term to each page, so the build, the copywriting and the redirect plan all work from one document.

Revised structure

Page-to-keyword map

Each page gets one primary term it is written to win, plus a supporting cluster. The third column is the redirect plan: every existing address is accounted for, so nothing that currently earns visibility is lost at launch.

Volumes shown are Semrush estimates and run well above measured traffic; they indicate relative priority, not forecasts. Terms without a figure are hyper-local and carry no estimate — which reflects the limits of the tool rather than absence of demand. Spanish equivalents mirror this structure at /es/.

What we do next

Before launch

• Confirm with the advertising agency when the call-tracking software was installed, to settle the June 2025 question
• Restore call and text tracking, which worked until June 2025, and connect the call-tracking software to analytics so it cannot silently break again
• Repair the “get directions” tracking
• Give each form its own confirmation address so conversions are traceable to a page
• Consolidate the 27 overlapping pages into single, stronger pages
• Correct the Spanish addresses in the sitemap
• Map every existing address to its new home so no visibility is lost
• Rebuild and expand the Minnesota abortion law page
• Audit the 50 embedded code blocks that carry tracking and scheduling tools

At launch

• Sticky tap-to-call and tap-to-text on every page
• Online scheduling that works without a phone call
• Complete metadata and structured data across all pages

After launch

• Deepen the eight city pages to compete on “near me” searches
• Build content in the priority clusters, targeting depth over volume
• Pursue community and directory links, which competitors already hold
• Consider professional translation of the highest-value Spanish pages
• Introduce a consistent review request at the end of a positive visit

Decisions we need from you