Bloom Nutrition is a direct-to-consumer supplements brand Megan Pritchard launched out of Cardiff in early 2026 with a thirty-SKU starter range covering vitamins, mineral complexes and protein powders. The product positioning was distinctive — sourced ingredients with single-source manufacturer transparency, third-party tested batches, and an editorial layer covering the actual science behind each formulation. The conventional advice for a new e-commerce founder is Shopify; the maths for Bloom did not support it.
The maths Megan ran
Unit margins on supplements in this segment run thin — Megan was targeting 38% gross margin on the average SKU, with the cheapest unit (a £14 vitamin C tub) landing at 31% margin. At Shopify Basic's 2.0% + 25p per-transaction fee, the £14 SKU was paying Shopify £0.53 — 1.5 percentage points of margin gone before plan fees, app fees or marketing costs. At anticipated order volume (1,000 orders / month within twelve months), the per-order fees alone would run £4,400 / year. The Stripe-direct alternative ran 0.7% + 5p per transaction lower, plus Megan could built in the editorial layer (heavy on long-form articles) without paying Shopify's headless-commerce premium.
The brief
A custom Next.js + Stripe storefront with thirty product pages, a Stripe-Tax-handled UK VAT layer (Megan was below the threshold at launch but expected to cross it inside six months), Klarna as a payment option, abandoned-cart recovery, customer accounts with order history, and a content layer of nine launch articles plus the editorial CMS to add more without a developer. Mobile-first, schema-tuned, Core-Web-Vitals-clean. Eight-day build window since this was Pro tier rather than same-day.
The content layer
Nine articles drafted from Megan's research notes covered: the bioavailability difference between magnesium glycinate and magnesium oxide; the actual evidence for vitamin D supplementation in the UK climate; what the third-party testing protocols on each batch actually measure; the sourcing story for the marine collagen line; the difference between whey isolate, concentrate and hydrolysate; the regulatory landscape around the supplements category in the UK (a particularly tangled topic); a comparison piece on plant-versus-animal protein for satiety outcomes; the case for cyclical (rather than daily) supplementation in healthy adults; and a long-form FAQ piece that lived alongside the customer-service page but was indexable in its own right. Each article ran 1,800-2,800 words, was technically reviewed by an external nutritional scientist Megan had retained on a contractor basis, and carried citations to primary sources.
What we built across the eight-day window
Days 1-2: catalogue architecture, type system for the SKU shape (including the unusual fields supplement products need — serving size, doses per container, suggested use, allergen flags, batch testing certificate URL). Days 3-4: storefront UI — homepage, category pages, individual product pages with the editorial-quality detail sections, cart, checkout. Day 5: Stripe integration with Setup Intent flow for the future Klarna upgrade, customer-account login, order-history retrieval, abandoned-cart recovery via a server-side trigger. Day 6: content layer — the nine articles loaded into the CMS, the editorial templates, the related-article cross-linking. Day 7: schema layer (Product, Offer, Article, FAQPage), Search Console submission, GA4 conversion goals, the Stripe Tax configuration, the abandoned-cart recovery email sequence. Day 8: final QA, payment flow testing against real cards, accessibility audit, mobile testing across five viewport sizes, launch.
The schema and SEO layer in detail
Each of the thirty product pages emitted Product schema with Offer (price, currency, availability, priceValidUntil), an AggregateRating placeholder that filled in as real reviews came through Trustpilot, the brand entity, the manufacturer, and a connected PropertyValue for each declared specification. Each of the nine articles emitted Article schema with a named Person author (the external scientist whose review backed the technical accuracy), publishedTime, modifiedTime, wordCount, and a citations field listing the primary sources. The site indexed inside 48 hours; the long-form articles began ranking for their target queries inside three weeks.
The launch and first month
Soft launch on a Friday with a 200-person friends-and-family list and a 1,400-person Instagram audience Megan had been building since the brand concept stage. First sale at 6:48 PM the same evening — the £14 vitamin C tub. By midnight: 11 orders, £312 revenue. By the end of week one: 22 orders, £680 revenue. Month one closed at £4,500 across 78 orders with an average order value of £58 — comfortably ahead of Megan's conservative £2,800 month-one target. Crucially, 17% of the month-one orders came through Klarna, which Megan had not been sure would matter for the supplements category but turned out to be the deciding factor on the higher-AOV bundle orders.
The content engine working as designed
By month three, organic search was driving 31% of all traffic — a faster ramp than Megan had projected. The nine articles ranked for a long-tail set covering the technical questions Megan had been answering one-by-one in DMs for the previous year. The bioavailability article ranked first-page for "magnesium glycinate vs oxide UK"; the vitamin D piece ranked for "vitamin d UK climate evidence"; the cyclical-supplementation article ranked for several related queries that brought a noticeably higher-intent visitor than paid social. Conversion rate on organic traffic ran roughly 1.7× the conversion rate on paid social traffic — typical of content-driven brands but visible in Bloom's analytics from month two.
What the Stripe-direct stack actually paid back
On 78 orders averaging £58, month one revenue of £4,524 carried Stripe fees of £74 (1.4% + 20p × 78). On Shopify Basic the equivalent fee would have been £110 (2.0% + 25p × 78) plus the £19 monthly plan and an estimated £80 of app fees for equivalents to the cart-recovery and reviews features built in to Bloom's stack. Net month-one platform-cost saving versus Shopify Basic: roughly £135. Annualised at the projected order volume, the saving runs to roughly £4,000-£5,000 — comfortably above the original build cost premium Bloom paid for the Pro tier.
What the next twelve months look like
Megan crossed the UK VAT threshold at month four, which the Stripe Tax integration handled transparently. The Klarna integration moved from optional to actively pushed in the upsell flow once data confirmed it lifted AOV by 22%. The content engine has been running at one article per fortnight since launch — Megan now uses a freelance writer briefed against the same editorial standards as the original launch nine. The site sits at 42% organic traffic share at the end of year one. Bloom has not migrated to Shopify, and Megan reckons she never will — the unit economics that drove the original decision compound rather than reverse over time.
The technical-review checklist for the supplements category
Supplements as a category sits inside Google's Your Money or Your Life (YMYL) classification, which means health-and-safety claims attract substantially stricter scrutiny than for a non-regulated category. The launch sequence included an additional review step: every article and product page was reviewed against MHRA advertising rules, the ASA CAP code for medicinal claims, and the Food Standards Agency rules on supplement labelling. Three product pages had to be reworded before launch (the original copy implied therapeutic effects that the MHRA does not allow for unlicensed supplements). The technical review took two of the eight build days but prevented an inbound from the regulator, which would have cost materially more than the review time saved. We have since added the YMYL review step to every health-adjacent build.
Why the content depth mattered more than expected
A pre-launch hypothesis was that the long-form articles would mostly serve as a trust signal — landing pages that visitors would read but not search-find. The data contradicted that. By month three, the articles were collectively driving 31% of all traffic, and the average visitor who landed on an article converted at 1.7x the rate of a visitor who landed on a product page from paid social. The mechanism appears to be: the articles match the technical-question search intent that the highest-margin customer segment uses (people researching specific nutrients, sourcing protocols, bioavailability differences), and that segment converts at a higher rate because the editorial layer pre-qualifies them. For the supplements category specifically, content depth is closer to a primary marketing investment than a secondary trust layer.
What we got wrong on launch
Two structural misses. First, the abandoned-cart recovery sequence was timed against US/Shopify default templates — first email at 1 hour after abandonment, second at 24 hours. Bloom's data showed UK supplements customers responded better to first-email-at-4-hours, second-at-48-hours; we adjusted in week three after the cohort data was visible. Second, the Klarna integration was off by default at launch — visible only if the customer expanded the checkout payment options. Moving it to a primary checkout button at the £80+ cart threshold lifted Klarna usage from 7% to the 17% figure cited above. Both were fixed inside the standard 30-day support window at no additional cost.
If you have a similar business
If you are launching a DTC consumer brand with thin unit margins (sub-40% gross), an offer that benefits from editorial depth, and an expected order volume that crosses 200/month in the first six months, the Bloom template applies. Pro-tier custom Next.js + Stripe build (£1,499), content layer at launch (six to ten substantial articles), Stripe Tax pre-configured, Klarna for AOV uplift, abandoned-cart recovery with UK-tuned timings, and per-product Product + Offer schema with PriceSpecification. Pre-condition: you can either write the editorial layer in-house or budget for a freelance technical writer at roughly £150-£300 per article. Expected outcome: month-one revenue 30-60% ahead of Shopify-baseline equivalent, faster organic traffic compounding, lower lifetime advertising cost ratio.