Beautiful Curly Me

Announcement bar: “FREE DOMESTIC SHIPPING for orders $60+”
Promo bar, same page: “Get free shipping when you spend $75 or more!”
Homepage meta and PDP both render both.

A shipping threshold is an AOV instrument. Two of them, fifteen dollars apart, on one screen, means the instrument is broken and the shopper is being asked to work out which number is true.

Every basket between $60 and $75 is a coin flip on whether the customer feels cheated at checkout. This is a one-hour fix with a measurable effect on both AOV and support tickets.

Hero asset: Labor_Day_Sale-3.jpg — 240 KB, 1920px wide. Headline, subhead and a pink “Shop Now” pill are baked into the pixels. A second banner ships at 343 KB uncropped.

Nothing in that hero is text. It can’t reflow on mobile, can’t be read by a screen reader or a search engine, can’t be translated by Shopify Markets, and can’t be updated without opening Photoshop. The typo it carries — “No code needed.Savings at checkout” — has been shipped at 1920px because fixing it means re-exporting an image.

Every promotional cycle costs a design round-trip instead of a text edit, and the brand’s largest above-the-fold statement is invisible to search.

Bella PDP: 4.9 ★ · 191 reviews (confirmed in Product JSON-LD).
/collections/all: 0 rating elements on 40 product cards. Homepage featured products: 0.

The store’s strongest conversion asset is locked one click deeper than the decision. Shoppers comparing four dolls in a grid see four prices and no proof.

Star ratings on collection cards are one of the most reliably positive changes in ecommerce. The data already exists and is already paid for.

Loading together on the PDP: loox.io (widget iframe + sticky side tab) and cdn.fera.ai (9 DOM nodes). Both are paid subscriptions. Both claim to own the rating.

Two review systems means two review inboxes, two request emails a customer might receive for the same order, two sets of star markup competing for the rich result, and two monthly invoices.

Duplicate spend, split social proof, and a genuine risk of conflicting aggregate ratings reaching Google.

Third-party hosts on the storefront: loox.io · cdn.fera.ai · static.klaviyo.com · static-tracking.klaviyo.com · config.gorgias.chat · config.gorgias.help · connect.facebook.net · googletagmanager.com · cdn.codeblackbelt.com · content.9gtb.com · cdnjs.cloudflare.com · shop.app + 4 CDN endpoints.
Theme app extensions: sales-countup, zipifypages, ocu-in-checkout. Plus a SimplyInsurance shipping-protection SKU sitting in the catalogue.

Two review apps, two Klaviyo endpoints, two Gorgias endpoints, a page builder (Zipify Pages) that renders landing pages outside the theme’s design system, and an 80 KB country-flag sprite loaded purely for the currency picker.

6.8 MB and 437 requests on a product page. On a 4G phone in a Target parking lot, that is the difference between a sale and a back button.

Homepage <script type="application/ld+json"> blocks: 0.
No Organization, no WebSite + SearchAction, no LocalBusiness, no sitelinks search box.

PDPs are correctly marked up, so this is not a theme-wide failure — it’s a specific omission at the one URL that should be telling Google who this brand is, where it’s based, who founded it, and which social profiles are official.

Knowledge-panel eligibility, brand-name SERP presentation, and the sitelinks search box are all left on the table for a brand with national press coverage feeding branded search.

Bella meta description: “… Gift for Gils ages 3+”
Homepage title: “Beautiful Curly Me| Empowering…” (no space)
Collection title: “All- Beautiful Curly Me…”

These are the lines that appear in Google results and in every link preview shared on social. For a brand whose product is confidence and whose buyer is a careful parent, a misspelling of girls in the search snippet is a brand problem before it is an SEO problem.

Small, but it’s free to fix and it’s the first thing a new customer reads.

Meta description: “14-year-old girl CEO”
Our Story page: “13-year-old CEO”
Homepage brand block: “Founded by a 7-year-old girl” — without saying that was her age at founding.

Each number is defensible on its own; together they read as three brands. A visitor skimming from Google result to homepage to Our Story collects 14, 7 and 13 in under a minute. And because the age advances every year, any hard-coded number is a maintenance liability by design.

The origin story is the brand’s single biggest differentiator against Target shelf competition, and it currently has no canonical version.

One homepage row, four products: a loose doll on white · a doll inside retail packaging · two dolls in a box · a flat-lay bundle. Four crops, four scales, three backgrounds, three lighting setups.
Badges in the same row: Bestseller (pink, top-left), available at Target (dark, top-centre), Limited Edition (pink, top-left).

The catalogue mixes studio cutouts, packaging shots, event photography and screenshots. Nothing establishes a repeatable shot list: no consistent hero angle, no fixed scale relationship between a 10-inch and an 18-inch doll, no styling grammar, no lifestyle-to-silo ratio.

The grid reads as a marketplace listing rather than a brand. At $65–$95 price points, photography is the perceived-value argument.

Published theme name: “July 2026 copy of Live Updated Impulse ||…”
Base: Impulse 9.2.0 (Archetype, Theme Store ID 857).

The duplicate-to-edit workflow with no version control, no dev theme and no staging. Every customisation drifts further from the Impulse base, which makes each vendor update more expensive to take, and there is no way to diff, review or roll back a change.

Change velocity gets slower every quarter, and one bad Saturday edit hits live traffic with no undo.

“Inspire Confidence in Her” row — four tiles: Beautiful Black Dolls · Empowering Books · Inspiring Puzzles · Gift Sets & Bundles. Small white caps, bottom-left, no scrim or gradient, over high-detail photography.

Contrast depends entirely on whatever happens to be behind the words. On the puzzle tile the label sits over a multicoloured illustration; at mobile widths the type gets smaller while the photo stays busy.

Fails WCAG contrast in several tiles and, more practically, the four labels are the main navigation into the catalogue.

Bella: highlights, narrative, care instructions, materials, packaging note and a 9-question FAQ all rendered as one continuous scroll. Paragraphs nested inside <ul>; two items (“Great for ages 3+”, “Encourages self-love…”) lost their bullet markers entirely. Bold headings sit mid-list.

The content itself is good — it answers washability, safety, age, contents and returns. It is simply unstructured: no accordions, no spec table, no care panel, no size reference near the buy box.

On mobile the buyer scrolls through nine screens of unformatted text to find whether the hair is washable. Most won’t.

375×812 viewport, Bella PDP: the Gorgias chat bubble sits on top of the sticky ADD TO CART bar; the Loox “Reviews” tab covers the left edge of the product image; the announcement bar wraps to two lines and clips its own “Shop Now” link.

The gallery also runs 12 pagination dots with no thumbnails, and the sticky bar carries a button with no price and no product thumbnail.

The single most valuable pixel region on mobile commerce is partially covered by a support widget.

/collections/all — H1: “ALL”. Description, and therefore the CollectionPage schema description: “All discount codes will be applied at checkout. Bundles are already discounted as reflected.”

A logistics note is doing the job of category copy, and it’s being fed to search engines as the page’s description. The Dolls collection does have real copy, so the pattern is inconsistent rather than absent.

Collection pages are the natural ranking targets for “black dolls with curly hair”, “natural hair dolls for girls”, “black girl puzzles” — the highest-intent non-branded terms in this category.

Bella PDP “You may also like”: features Bella.
Sitemap: 49 products. Published to /collections/all: 40. Nine SKUs unaccounted for.

Recommendations aren’t configured, and the catalogue has drift — products that exist, are indexable, but appear in no browsable collection.

Six bundle SKUs exist specifically to raise AOV and the PDP isn’t using them. The natural Bella upsell — the $95 Curl Confidence Bundle — is one row away and isn’t shown.

Single column, no headings, no logo, no mission line, no social icons, no newsletter. Order runs: Shop, Book Zoe to Speak, Join our Email List, Find us in Store, Wholesale, Search, FAQs, Contact Us, Returns, Shipping, About Us, Join our Team, Earn $$$, Customer Reviews, Terms of Service, Blog.

The footer is where a values-led brand restates what it stands for and where a wholesale buyer, a school partner and a journalist all find their door. Here, “Book Zoe to Speak” — a genuine revenue line — sits second in an undifferentiated list.

Three distinct audiences (retail buyers, educators, press) share one flat list with the returns policy.

Wholesale page: name / email / message contact form. No minimums, no tiers, no price list, no application gate, no B2B catalogue.
Confident Girls Lead course: vendor Thinkific, $59, fulfilled off-platform.

With product on Target shelves nationwide, inbound retail interest is a real pipeline and it is currently handled by inbox. Shopify’s B2B tooling would let approved retailers log in to a price-tiered catalogue against the same inventory.

Retail partner acquisition is manual, and course customers leave the brand’s own environment at the point of highest engagement.

Repeated “Just a moment… / Your connection needs to be verified” Cloudflare interstitials on product URLs during automated browsing, from an Asia-Pacific IP.

Triggering this with automation is expected behaviour, so this is not itself a defect. It is worth confirming that the same rules aren’t challenging Googlebot, Bingbot, Shopping feed crawlers, or real shoppers on VPNs and corporate networks.

If crawlers are being challenged, it would silently suppress indexation and Merchant Center feeds. Fifteen minutes in Search Console settles it.