Every claim below was verified against the actual code by three independent AI reviewers — Kimi K3 (read all four files + the audit), Codex (adversarial code review, 100k+ tokens of checking), and Sakana Fugu Ultra (business strategy). Where they found problems in OUR concepts, we fixed them and logged the fix. This is the receipts page.
We scanned every page of the live site and documented 16 concrete problems. Kimi and Codex then independently checked our concepts' markup to verify each one is actually fixed — not assumed fixed. Verdicts below are theirs, updated after our fix round.
| Problem on the live site today | A | B | C | D |
|---|---|---|---|---|
| Two competing Login/Register menus (custom pages + wp-login) | FIXED | FIXED | FIXED | FIXED |
| Broken phone link — href="http://(727)-472-5456" is an invalid URL | FIXED | FIXED | FIXED | FIXED |
| Footer year missing in the pre-JavaScript markup (appears once scripts run — crawlers see the bare "Copyright ©") | FIXED | FIXED | FIXED | FIXED |
| Stat counters read "0 Pounds of Meat / 0 Happy Customers" until JavaScript runs — the zero-state is what Google, crawlers, and slow connections see. (Verified live: in a full browser they do animate up to 6.4K / 1,028 / 7 / 82 — so we kept his real numbers and dropped the zero-risk.) | FIXED | FIXED | FIXED | FIXED |
| Events in the nav opens a completely empty page | PARTIAL | FIXED | PARTIAL | PARTIAL |
| Contact page leaks WordPress blog sidebar (Recent Posts, Archives, honeypot text) | FIXED | FIXED | FIXED | FIXED |
| Price text duplicated: "$22.00 – $150.00 Price range: $22.00 through $150.00" | FIXED | FIXED | FIXED | FIXED |
| Raw variation tables; no grade help or reviews at the point of choice | PARTIAL | PARTIAL | PARTIAL | PARTIAL |
| Flat 38-product wall mixing seafood/chicken into a beef brand | FIXED | FIXED | PARTIAL | FIXED |
| Three rotating hero taglines — no single message | FIXED | FIXED | FIXED | FIXED |
| No "How it works"; local delivery vs 48-state bulk never explained | FIXED | FIXED | PARTIAL | FIXED |
| VIP ($7.77/mo) and Bootlegger Bucks buried in menus | FIXED | FIXED | FIXED | FIXED |
| No social links anywhere despite active IG/TikTok/FB | FIXED | FIXED | FIXED | FIXED |
| About page repeats counters + a second testimonial block, empty headings | FIXED | FIXED | FIXED | FIXED |
| Two competing "reviews" experiences, both uncurated walls | FIXED | FIXED | FIXED | FIXED |
| Trust signals scattered — no guarantee, no delivery-zone story | PARTIAL | PARTIAL | PARTIAL | PARTIAL |
Honest reading of the PARTIALs: these are homepage concepts, not full sites. "Point-of-choice grade help" needs real product pages, "Events" needs a real events page with dates, and "trust" wants a delivery-zone map — all scoped for the production build, impossible to fake honestly in a one-page concept. The current site fixes NONE of the 16.
And the animations question, directly: his animations are a 3-slide tagline rotator, count-up counters, and stock fade-ins riding on jQuery + 20 script files (4.4MB page). The rotator IS the "no single message" problem, and the counters ARE the zero-state SEO problem — the two flashiest animations on the site are the two causing defects. Ours animate with intent (entrance reveals, tilt, and D's draggable 3D vault), weigh a fraction as much, and every one respects the reduced-motion accessibility setting, which stock jQuery counters and sliders do not.
Drey's UI-STANDARD is the design rulebook every client build gets judged by. Here's the old site vs the concepts on its core rules, scored /10:
Old site: 3 rotating taglines, decorative "shape" PNGs, identical card grids — the swap-the-logo test fails instantly. Each concept has one message, one accent color, a distinctive layout move (dot-leader menu / occasion picker / member table / 3D vault), and passes the litmus: swap the logo and it still looks like nobody else's site.
Old site's only "motion" is the tagline rotator and counters that visibly break. Every concept ships intentional entrances, hovers, and one signature moment — and every animation is wrapped in a reduced-motion check, including smooth-scroll (Codex caught that we'd missed smooth-scroll; fixed in all four).
Concepts: layered shadows on cards that should feel like objects, SVG stars instead of icon-font spam, focus-visible outlines on every interactive element, skip-to-content links, per-page tab titles + meta descriptions, semantic header/nav/main/footer. Old site: flat sections, default WooCommerce chrome, honeypot text visible to humans.
Kimi traced the first-purchase path for a cold visitor landing from Instagram (where Stu's traffic actually comes from):
| Site | Steps to a buyable product | What kills the sale along the way |
|---|---|---|
| Current site | 5+ steps | Shop nav → scroll a 38-item wall across 2 pages → open a product → decode a raw variation table with duplicated price text → add to cart. No grade help, zero reviews shown, delivery area unexplained. |
| A — Speakeasy | 2 steps | Gorgeous, but the menu is a price list — production build must make rows clickable. |
| B — Butcher's Counter | 2 steps | Occasion chip → 3 recommended cards with price + add-to-cart. Kimi's pick for cold traffic: answers "what is this / can I trust it / what do I buy" in the first viewport. |
| C — Meat Club | 2 steps | Leads with a $250 box + membership — a big first ask 10 seconds off a reel. |
| D — The Vault | 2 steps | The 3D vault is the product browser — spin, stop on a cut, price is right there. Needs the same cart wiring as the others. |
Fugu Ultra ranked all ten shared fixes by expected revenue impact for THIS business — a relationship-driven local brand where one recovered bulk order is worth $1,550–$5,700:
Fugu Ultra ran Stu's own numbers: if Bootlegger Bucks (a 9.09% effective discount) stacks on top of VIP's 10%, a $100 product effectively sells for ~$81.82 — an 18.2% discount before the membership fee. And a VIP buying a $5,700 whole beef takes a $570 discount against $7.77 of monthly revenue. Before ANY redesign promotes these offers harder, Stu needs stacking rules, bulk exclusions, and caps in writing. We already removed every "stacks with VIP" claim from the concepts. This one insight may be worth more than the website.
We didn't just grade the old site — we had all three models attack the new concepts. Every legitimate hit got fixed the same hour:
| Catch (who) | Status |
|---|---|
| Mobile nav was deleted below 900px — nav AND phone number vanished on the exact devices his Instagram traffic uses (Kimi + Codex) | FIXED — phone number stays visible on mobile in all 4 concepts; full hamburger menu specced for production |
| "NEVER FROZEN" hero sticker contradicted the frozen bulk-shipping program (Kimi + Codex, independently) | FIXED — now "CUT TO ORDER" |
| "No antibiotics/hormones" was documented for the Grass line only, but we'd claimed it brand-wide (Codex) | FIXED — scoped to "Grass-Fed Line" in every concept |
| Invented offer terms: Bucks "never expire", "stack with VIP", "cancel anytime"; "unmarked" deliveries; "aged" beef (Codex) | FIXED — every undocumented claim removed or softened to what reviews actually say |
| Contrast failures in C: guest-column text 2.65:1, red kickers 3.82:1, white-on-red buttons 4.31:1 (Codex, measured) | FIXED — guest text lightened, kickers/footers moved to darker red or light coral, buttons darkened to pass 4.5:1 |
| Smooth-scroll not gated by prefers-reduced-motion; no skip links; C's comparison table had no overflow wrapper or label (Codex) | FIXED in all 4 — gated scroll, skip-to-content links, scrollable labeled table |
| Occasion chips lacked aria-controls (Codex) | FIXED |
| Reveal animations left the page blank if JavaScript ever failed (found in our own testing) | FIXED — content is visible by default; animation only activates when JS confirms it's running |
| "My grill started clapping — Julian B." flagged as invented (Codex) | FALSE POSITIVE — it's a real Google review from the live-site scrape; Codex only saw the audit summary. Kept, and flagged for quote-verification with Stu anyway. |
| Full-size 2100px images used in card-size slots; no width/height attributes; no srcset (Codex) | ACCEPTED for local concept files (instant off disk) — resized/responsive images are line 1 of the production checklist |
| No cart, search, account, or ZIP-check — nothing is actually purchasable (Kimi) | BY DESIGN — these are direction concepts. The real build (Shopify) brings cart/checkout/accounts natively; ZIP-gate is specced in the README. |
Fugu Ultra's guardrail, followed to the letter: "Move product evidence, not the purchasing interface." And its success metric if Stu picks D: non-refunded revenue per visitor, A/B tested against a static version — spin-time and scroll depth are diagnostics, not proof. Honest risk (also Fugu): D is the most expensive concept to maintain and the easiest to get wrong on cheap phones — it should A/B its way in, not launch on faith.
Concept B converts cold Instagram traffic best. "It answers the three cold-traffic questions in the first viewport... its skeleton is the shortest honest path from curiosity to checkout."
Concept B is the strongest first-client demo — "occasion-based discovery, grade comparison, delivery split, and restrained voice make the business easiest to understand." A has the cleanest performance + brand fit; C was weakest pre-fixes.
B is the safest commercial base for families AND corporate hosts — with the warning that clarity can commoditize: keep Stu, hand-delivery, and Tampa identity central, or B becomes any online butcher.
Build on B's skeleton, dressed in A/D's skin, with C's membership table as its own page. All three reviewers independently picked B's structure as the best seller — and all three warned its plainness is the risk. So: B's occasion picker, grade table, and delivery split as the layout; the Speakeasy/Vault dark-premium look and voice as the brand layer; the member-vs-guest table from C on a dedicated VIP page. The Vault carousel from D becomes the signature "wow" module on the homepage — A/B tested, per Fugu. Confidence: high on the structure call (3-for-3 reviewer consensus), medium on the exact visual blend until Stu reacts to A vs D.
Kimi K3's page architecture for the Shopify build — the concepts are homepage directions, the real site is six pages: Home (hybrid homepage: stats, occasion picker, grade table, delivery split, Google reviews, Stu, VIP teaser) · Shop (full 82-product catalog with grade/cut/occasion filters) · VIP / Meat Club (C's member-vs-guest table as a revenue page) · Bulk Beef (quarter/half/whole with per-pound math) · Our Story (Stu + the 7 farms + the lore) · Contact / FAQ (delivery zones + how-it-works). Two of his current tabs die honestly: Events (an empty page today) and standalone Reviews (absorbed into Home and product pages). The redline for Stu: current — 7 tabs, one empty, VIP buried. New — 6 pages, zero dead ends.
Prepared 2026-07-31 · Sources: full scan of beefbootlegger.com (all pages, 38 products, 122+ reviews, socials) · research/site-audit.md · UI-STANDARD.md · Reviews by Kimi K3, Codex (gpt-5.6-sol, read-only sandbox), and Sakana Fugu Ultra v1.1 · All four concepts: open the comparison page