James & Ella: Funnel & PDP CRO Audit

Marcin Mleczko · 2026-06-04 Audit scope: ella.co PDP (freeze-dried raw, grass-fed beef), homepage, reviews page. Traffic context: visitors arrive via the James & Ella advertorial (listicle). They are already Solution Aware to Product Aware before hitting the PDP. The audit focus is congruence and friction removal, not cold traffic readiness.

How to read this

Findings are grouped by where they live in the funnel. P0 = fix before running paid traffic. P1 = include in the next dev sprint. P2 = phase 2, hold until test reads come back.

Traffic context: visitors arrive pre-sold by the advertorial. The PDP's only job is to (a) immediately mirror the promises the advertorial made and (b) remove every barrier between "I want this" and "card charged." Any friction that doesn't serve those two goals is a conversion leak.

Conversion score: B-

The PDP reinforces the advertorial's core claims adequately but breaks on three points that will cost post-click conversion: the guarantee signal is missing, the social proof count the advertorial cited isn't visible ATF, and the founder story that anchors the advertorial's credibility is reduced to one short quote at position 11. For warm traffic the bones are workable. Fix those three gaps and the test can run cleanly.

Top 3 fixes (the 80/20)

Written after the full section walkthrough. If the dev agency only has time for three things before the first traffic test, these.

1. Add a satisfaction guarantee to the PDP

The advertorial uses risk-removal framing ("if she doesn't eat it" type language) but the PDP has no explicit guarantee anywhere. A visitor who clicked through trusting that framing arrives and finds only "cancel anytime" -- which covers subscription commitment, not the real fear: wasted money on food a picky dog ignores. Fix: "If your dog doesn't finish the trial, email us within 14 days for a full refund. No questions asked." Place it under the calculator CTA, inside the Why Subscribe section, and as a fourth FAQ answer. Note: verify whether the guarantee language actually appeared in the advertorial or was implicit framing -- either way the PDP needs to state it explicitly.

2. Surface the 4.9 / 933 rating above the fold

The advertorial cites "4.9 stars across 900+ Trustpilot reviews" as a core proof point. A visitor who clicks through expecting to see that number confirmed arrives at a PDP where the Trustpilot link is present but no star count or review number appears in the first screen. Fix: "★★★★★ 4.9 from 933 verified reviews" as one line directly under the product title.

3. Clarify the subscription type options

The subscription type selector offers four options: Replacement / Trial / One-time / Subscription. "Replacement" is unexplained and creates confusion before the calculator is filled. Fix: rename or remove "Replacement" for new visitors and surface "14-day trial" as the pre-selected subscription path with a clear "order once" alternative link.

PDP findings (the highest-leverage page)

Above the fold

P0 Social proof count missing above the fold - Current: Trustpilot link present, but no star rating or review number visible in the first screen - Issue: The advertorial explicitly cites "4.9 stars across 900+ reviews." Visitors arrive expecting to see that confirmed immediately. They don't. Congruence break. - Fix: One line under the product title: "★★★★★ 4.9 from 933 verified reviews." Static hardcoded is fine, update quarterly.

P0 No guarantee signal above the fold - Current: Nothing. "Cancel anytime" appears inside the calculator section but addresses subscription commitment, not product fit. - Issue: The advertorial implies a risk-free first try. The PDP's silence on refunds creates a trust gap at the most critical moment -- right before the calculator is filled. - Fix: Add a one-line confidence badge beneath the price: "14-day trial · if your dog doesn't eat it, we'll refund you."

P1 Eight benefit badges are doing little conversion work - Current: 100% raw, human grade, gluten & grain free, all natural, delicious natural flavour, made by experts, soft crumbly texture, no prep or special storage - Issue: Eight undifferentiated icons is cognitive noise. None of them addresses the specific anxieties of a visitor who just read the advertorial (will my dog eat it, is it actually worth £40, how is this different from what I've tried before). - Fix: Cut to three. The three that land hardest for this audience: "Made by canine nutritionists · British-made in Norfolk · 933 five-star reviews." Move the rest to a "Why James & Ella" expandable below the calculator.

P1 No time-to-result claim above the fold - Current: Missing entirely ATF. Exists implicitly in testimonials lower on the page. - Issue: The advertorial primes "bowl was empty before I'd put her water down" as a near-immediate result. The PDP doesn't reinforce a timeline. Adding "most owners notice changes in coat and energy within 10--14 days" near the ATF trust strip converts the open-ended trial into a defined experiment, which reduces perceived risk.

P2 Hero image format is tall -- switch to square to recover ATF space - Current: Portrait-format product shot dominates the first mobile viewport, pushing the H1, badges, and calculator start below the fold - Issue: On a 390px screen, vertical image height is conversion real estate. A 1:1 square crop shows the product clearly while allowing the social proof strip and calculator entry to appear before the visitor scrolls. - Fix: Set the hero image to a square aspect ratio. Check that lifestyle carousel images (founder shots, dogs eating) crop cleanly at 1:1 before committing -- some may need reframing.

P2 Benefit-led subhead under the product title is missing - Current: Product title ("Freeze-Dried Raw Dog Food") followed directly by feature badges - Note: The product title itself is the product name -- that stays. The gap is a benefit-led hook directly beneath it before the badge row. Something like "Real raw nutrition. No freezer. No prep." between the title and the badges would do the conversion work without touching the product name.

Variant + calculator flow

P0 Subscription type options are confusing for new visitors

The four options -- Replacement / Trial / One-time / Subscription -- include "Replacement" which is unexplained and ambiguous before the calculator is filled. A new visitor doesn't know what they'd be replacing. For a first-time buyer the path should be obvious: trial vs one-time order. Everything else belongs in account management.

Fix: pre-select "Trial" and rename to "Start my 14-day trial." Offer a plain "Order once" link as the alternative. Remove Replacement and Subscription as visible choices for visitors who haven't bought before.

P1 Calculator output doesn't frame savings visually - Current: Shows cost/day, grams/day, delivery, discount, total -- functional but low-impact - The output already shows the trial price (£40 for small dogs, more for large dogs). The design opportunity is making the savings line more prominent. A callout style "You save £X vs one-time" with a visual treatment would anchor the subscription value better than a plain discount row.

P1 Large-dog price needs contextual framing - A 23kg dog calculates to £160 for a 17-day trial. Without context, this reads as a high price. A comparison anchor -- "that's £9.41/day, roughly the same as one bag of premium kibble a week" -- would reframe the output for larger dog owners before the sticker shock lands.

P1 No "what's in the box" preview before submitting - Fix: A small line near the CTA: "Your trial contains: [X]g of food portioned for [dog name], a feeding scoop, and a transition guide." Tangible items increase perceived value.

Trust, reviews, risk reversal

P0 Money-back guarantee missing - See Top 3 fix #1 above. The fix is copy, placement, and (if needed) a formal policy write-up.

P1 Four cherry-picked testimonials for a brand with 933 reviews - Current: Michelle (fussy Cockapoo), Grace (Sprocker Spaniel), Dhelia & Molly (Flo), Victoria (Romanian rescue) - Issue: Strong testimonials individually, but 4 out of 933 undersells the depth. Visitors can't see "do dogs like mine have this result?" - Fix: Add a Trustpilot widget embed with live count and a scrolling carousel of recent reviews. If the widget is slow, pull 8--10 Freeze-Dried-specific testimonials and display them with star ratings and dates. Priority segments: fussy eaters, skin/coat transformation, rescue dogs.

P1 Founder quote is one sentence at position 11 - Current: James Middleton quote at position 11 of 13 sections, short - Issue: The advertorial's strongest conversion asset is the James Middleton story -- the memoir, Ella, the therapy dog, the depression recovery. A visitor pre-sold by that story arrives at the PDP and finds it reduced to a single functional quote at the bottom of a very long page. - Fix: Move a 3--4 sentence founder excerpt to position 4 or 5, right after the benefits section. It's the brand's primary differentiator against every category competitor.

Section order below the fold

Current order: Highlights → Calculator → FAQs → Ingredients (×3 flavours) → Nutritionals → Why subscribe → Benefits → Testimonials → Press logos → Reviews → Founder → All FAQs link

P1 FAQs positioned before proof - "How does a trial work?" appearing before the ingredient and benefits sections reads as defensive. Move FAQs below testimonials + press.

P1 Three ingredient sections stacked without progressive disclosure

• Current: Three complete breakdowns (12--13 ingredient cards each) displayed simultaneously regardless of flavour selection
• Issue: The longest and heaviest section on the page, and 65% of it is irrelevant to the visitor's selected flavour. Page length correlates directly with mobile abandonment.
• Fix: Tabbed or accordion UI -- show only the selected flavour by default. "Compare flavours" toggle for side-by-side view. Ideally auto-updates when the calculator flavour changes, making the page feel personalised to the specific dog being set up.

P1 Founder story buried at position 11 - See Trust section above. Same finding, different angle: the section order compounds the positioning problem.

Homepage findings

The homepage is doing brand-positioning work but very little conversion work.

P0 No trial offer or price visible - Current: "Raw Food, Made Easy" hero, CTA "Shop the range" -- no price, no trial mention - Issue: Visitors who land on the homepage first can't establish a price anchor before deciding whether to explore further. They bounce. - Fix: Add below the hero headline: "New here? Start the £40 14-day trial. Calculator-portioned to your dog. Cancel anytime." A secondary CTA "Build my dog's plan" pointing to the PDP would also capture intent earlier.

P0 No social proof count visible anywhere on homepage - Fix: "★★★★★ 4.9 from 933 verified reviews" in the hero strapline or immediately below it. One line.

P1 Hero subhead trying to cover three product lines at once - Current: "Support your dog's health with the power of Freeze-Dried Raw, the innovation of Cold-Pressed, and the balance of Kibble + Raw..." - Issue: Three product lines in one sentence. New visitors get whiplash. - Fix: Lead with the hero product (Freeze-Dried Raw), surface the other ranges through a product selector below.

P1 Six or more competing CTAs - Current: Shop the range / Find out more / How it's made / How it works / Our ingredients / Read all / Shop our products - Fix: One primary CTA ("Start my dog's trial") and one secondary ("How it works"). Everything else moves to navigation or footer.

P2 No product range guidance for multi-SKU visitors - A one-question quiz "Currently feeding kibble / raw / a mix?" routing to the most relevant product would reduce decision paralysis for visitors who don't already know they want Freeze-Dried Raw.

Reviews page findings

This page is the brand's most under-used conversion asset.

P0 No average star rating or total count shown - Current: "Join Thousands Of Happy Customers" -- no number, no aggregate rating - Fix: The page hero should be "933 reviews · 4.9 stars average." That's the only number that matters here.

P1 No filtering by outcome, breed, or product - The highest-value use case for a reviews page is "show me reviews from owners of fussy eaters / dogs with itchy skin / Cockers." Currently impossible. - Fix: Three filter tabs minimum: Dog type · Health concern · Product. Tags drawn from existing review content or manually curated.

P1 No dates visible on any review - Undated reviews lose credibility. Add dates or relative timestamps ("8 months ago").

P2 No video reviews - Video reviews convert at 3--5x text reviews. A simple program (free bag in exchange for a 60-second video) would build this asset over time.

P2 No pagination / load more - Showing ~12 reviews when 933 exist understates the social proof depth.

Funnel congruence, what the advertorial promises that the PDP doesn't reinforce

This is the core audit for a presell-first funnel. Every congruence break below is a moment where the trust built in the advertorial leaks out before checkout.

Fix: The PDP's above-the-fold trust strip should explicitly mirror these five lines: rating + count, founder + nutritionists, time-to-result, guarantee, British-made. Five words each, one line per claim. That's the entire trust scaffold a click-through visitor needs to confirm before they fill the calculator.

Quick wins for the sprint

Priority order for the first dev sprint:

1. ATF trust strip -- star rating + count + guarantee signal + time-to-result, directly under the product title.
2. Satisfaction guarantee -- write policy, place in calculator zone + Why Subscribe + FAQ. Requires internal sign-off on the policy before dev can ship.
3. Surface Trustpilot widget with live count on homepage and PDP.
4. Calculator simplification -- pre-select Trial subscription type, rename or remove the Replacement option.
5. Move founder story up the PDP section order -- position 4 or 5, not 11.
6. Add date stamps to reviews on /our-reviews.
7. Cut ATF badges from 8 to 3. The three that do conversion work for this audience: British-made · Made by canine nutritionists · 933 five-star reviews.

Structural items (Phase 2 after test reads)

Hold these until data comes back from the sprint above. Running everything simultaneously dilutes the test signal.

• Calculator full redesign -- progressive two-step flow (dog data first, flavour second) rather than stacked selectors
• Homepage trial entry point -- a homepage-level dog-profile widget that captures name/breed/age before routing to a flavour decision
• Reviews page filtering -- by dog type, health concern, product. Requires dev + content tagging work.
• Video reviews program -- operational and content lift, takes 4--6 weeks to build inventory
• Ingredient section progressive disclosure -- tab/accordion UI tied to calculator flavour selection
• Cart abandonment flow -- requires email + SMS infrastructure

Awareness + angle layer

Traffic context recap: Visitors arrive from the advertorial already Solution Aware to Product Aware. They know raw feeding is better than kibble. They know James & Ella exists. They've seen proof. The PDP's only job is to confirm what the advertorial promised and remove the last objections before checkout.

Belief sequence gaps: The PDP does adequately on Beliefs 3 and 4 (the mechanism exists, this product delivers it). It fails on Belief 5 (the risk of trying is low). The guarantee gap is not a copy problem -- it's a missing belief. Until it's filled, a meaningful fraction of warm click-throughs will read the whole page and still not add to cart.

Angle variants the current PDP can't serve alone:

The advertorial was built for a single angle. But three high-volume audience segments would each convert better with their own ATF section or dedicated landing page: - Fussy eater segment -- Michelle's testimonial is the hook. "Finally a food she goes mad for" should be the above-fold headline for this angle, not buried in position 4 of the testimonial carousel. - Health / skin / digestion segment -- Grace's testimonial covers five outcome dimensions. This audience needs outcome specificity ATF, not general "raw is better" positioning. - First-time raw feeder -- transition anxiety is high. "How should I introduce freeze-dried raw?" as an FAQ answer doesn't address it. A "switching from kibble?" callout with a 3-step transition plan near the ATF would convert this segment.

Ready for advertorial traffic?

Run it, but fix three things first. The funnel structure is sound -- listicle to PDP to checkout is a valid path and the PDP won't actively break the trust the advertorial builds. But three congruence gaps (guarantee signal, star count ATF, time-to-result claim) mean the PDP is currently giving back some of the conversion work the advertorial did. Fix those in one sprint, then run. Don't wait for the full structural phase to test.

One philosophical note

The advertorial does a lot of heavy lifting -- it establishes the problem, builds the mechanism, earns the trust, makes the promise. The PDP then has a relatively simple job: confirm the promise and get out of the way. The frustrating thing is that the PDP fails at that simple job in the exact four places the advertorial was most specific. It cited 933 reviews -- the PDP doesn't show the number. It implied a guarantee -- the PDP doesn't say one. It named James Middleton and his nutritionist team -- the PDP buries both. The fix isn't a redesign. It's the PDP learning what the advertorial already knows about what this audience needs to see.

Audit prepared: 14:05, 2026-06-04 · Good Advertorials Pages audited: ella.co/product/freeze-dried-raw-dog-food, ella.co, ella.co/our-reviews Presell context: James & Ella advertorial (listicle). PDP visitors are warm -- Solution Aware to Product Aware.