55%
gross margin, unchanged
from £200k to £6.3m
E-commerce & Growth Director
£200k
to £6.3m
in ten years.
Annual turnover at Bareback Footwear, 2016 to 2026.
I build the growth systems consumer brands actually run on: paid media, CRM, product and the automation underneath. Fourteen years in consumer retail, ten of them here. Gross margin held at 55% the whole way up.
Annual turnover · Bareback Footwear
31.5×
2016 · £0.2m
2021
2026 · £6.3m
Worcester, UK. Open to Director roles and a small number of consultancy projects.
£570k
annual media spend,
built from zero
10:1
lifetime value to acquisition cost,
£180 against £18
100k+
CRM subscribers,
grown from an empty list
Meta AdsGoogle AdsKlaviyoShopifyTikTok ShopZendeskSEO & AEOCreative strategyClaudeAWIN
Acquisition
£570k a year, bought
at a 10× blended return.
Meta and Google carry the acquisition between them. Meta takes roughly seventy percent of the budget and has never dropped below 6×, even through the weak seasons.
Google Ads
Per year
16×
return on ad spend
- Share of revenue
- ~30% of £6.3m
- Campaign types run
- PMax · Search · Shopping · Demand Gen · YouTube
The creative
Briefed against the ad account,
not a mood board.
Product film for the Kentucky range: the boot in use, the fit, the ground it is built for. Videographers, photographers and influencers were commissioned to answer what the account said was working, and the footage was cut for the feed before it was cut for anything else.
376%
new-customer growth,
year on year
~£18
new-customer acquisition cost,
held while spend grew
39.96%
of customers return to buy again,
averaging 1.7 orders each
−34%
product returns,
32% down to 21%
Bareback Footwear · 2016 – 2026
Channel split modelled from the annual budget at a 10.3× blended MER. Platform exports on request.
Growth that costs margin is not growth.
Fifty-five percent gross margin at £200k, and fifty-five at £6.3m. Every decision below was made with that number in the room.
The growth factory
Five stages, and none
of them stand alone.
What a customer tells us at the end of this loop decides what we buy, photograph and advertise at the start of the next one.
Acquisition
Bought at £18 a customer
- Meta at roughly 70% of budget, held to 10× lifetime and never below 6×.
- Google across PMax, Search, Shopping, Demand Gen and YouTube.
- TikTok Shop and GMV Max, plus AWIN affiliates.
- £570k a year of media, built from nothing.
The shop
Built by hand, not by app store
- Self-built Shopify themes in HTML and CSS, no page-builder tax.
- Product finders and customer navigation written from scratch.
- Technical SEO and content automation running unattended.
- Every conversion test shipped by the person who ran the ads.
Retention
A quarter of revenue, owned outright
- 100,000+ Klaviyo subscribers grown from zero.
- Thirteen live flows — welcome, browse, checkout, post-purchase, win-back, back-in-stock.
- A custom loyalty scheme with its own referral engine.
- 39.96% of customers come back, averaging 1.7 orders.
Product
Best-sellers grown into ranges
- Quebec, Hudson and Kentucky extended into ankle, mid, zip and long.
- Care products, insoles and socks sourced and priced in-house at ~80% margin.
- Bundles and checkout upsells lifting order value.
- One proven boot, expanded into a range, reaches more price points on the same spend.
Fit & returns
A third fewer boots came back
- 360° scans coded into the product page.
- A 1–5 fit level on the size row of every core boot.
- Charts, imagery and size-led advice before the add-to-cart.
- Returns fell from 32% to 21% in four months, and held.
What customers say about fit and use decides the next range, the next photograph and the next ad
£126k
Recovered from imperfect stock through a direct-to-subscriber seconds programme.
~80%
Margin on care products, insoles and socks, sourced and priced in-house.
34
Listings across four price points, grown from three proven boots.
Product development
Advertise the boot,
not the category.
Then grow that boot into a range.
A category has to be explained before it can be sold. A single boot only has to be good, and you know inside a week whether it is. So we advertised one boot, waited for it to prove itself, then built a range around it — heights, closures, linings — until the same advertising reached four price points instead of one.
Quebec
The proven boot
Quebec Waterproof, Brown
4.86from 554 reviews
extended into
Ankle bootshigh instep
Zip-up ankleeasy entry
Zip-up mid-calfwide calf
Wool-lined zipwinter
Hudson
The proven boot
Hudson Mid Waterproof
£234one colourway, three heights
extended into
Hudson Ankle£194
Hudson Mid£234
Hudson Storm, long£269
One ladder, £75 of headroom+39%
Kentucky
The proven boot
Kentucky Jodhpur
£75the entry point into the brand
extended into
Kentucky Stormwaterproof, 8 colours
Kentucky Long Riding£279
Half-chapscompanion sale
£75 boot to £279 boot3.7×
3proven boots, grown into 34 listings across four price points
The range is the brand journey. Someone arrives on a £75 jodhpur and finds there is somewhere to go next: a waterproof version, a taller shaft, a warmer lining. The advertising never got wider. The catalogue did, and that is what turns a first order into a second. Nearly 40% of customers come back.
Returns reduction
A boot that fits
doesn't come back.
Returns are a photography and information problem long before they are a logistics one. Four changes to the product page, four months, and returns fell from 32% to 21%. Read the case study
Returns rate · last four months to date
−34%
Jun · 32%
Jul
Aug
Sep · 21%
The four months to date. Vertical axis runs 18–34% so the change is legible — eleven points of returns rate, taken out and held there.
360° scans, coded into the page
Fifteen frames per boot, drag to spin. No more guessing what the back of the heel looks like.
A 1–5 fit level on the size row
Runs small to runs large, published per boot, not buried in a generic size guide.
Advice that changes as you pick a size
Calf and instep guidance surfaced at the moment of the decision, not after it.
Then we ask, after it arrives
A post-purchase email asks how the fit is, and the answers rewrite the fit levels.
360 VIEW
DRAG TO SPIN
DRAG TO SPIN
Hudson Mid Waterproof Boots
£234.00
Size
FIT
Runs slightly large
EU 42 — UK 8
Runs slightly large.
Your usual size should be right. If you’re between sizes, go down.
Men: stick to your typical size, or size up if you’re between sizes.
Calf
37.5cm
Measured at mid-calf height
Height
26cm
Sole to top of boot
Fit scale
Runs small
Slightly small
True to size
Slightly large
Runs large
Post-purchase, split on order history
Both arms send at day six
Order placed
First order
Exchange Inquiry
Never bought before, so the exchange promise leads.
Orders > 1
Fit Check
Knows the brand, so the fit question leads.
“How do they feel?”
A quick fit check now your boots have arrived.
Six days after the order, once the boots have actually been worn. It asks three things — heel, toes, calf and instep — then offers a free exchange for sixty days. The answers feed back into the fit level published on the product page, so the next customer buys on this one’s experience.
Both arms carry the same message today. The split is already built, so the copy can diverge without touching the flow — that is the next iteration.
One of nine emails the flows send. See them all
Retention
Thirteen flows, and
one of them is a business.
A quarter of annual revenue arrives through a channel that costs nothing per send. Built from an empty list, and rebuilt from scratch in 2026 when the old flows stopped earning their keep. Read the case study
Acquisition & recovery
Welcome SeriesAdded to list
Abandoned CheckoutMetric
Browse AbandonmentMetric
Back in StockMetric
After the order
Post-Purchase, splitting on repeat vs firstMetric
Fit check, day sixBranch
Win-backMetric
The loyalty engine
Voucher claimedReward redeemed
Friend's referral codeMetric
Referral creditMetric
Birthday creditDate based
Fit & service
International orderAdded to list
Wide-fit guidanceAdded to list
- Built, not bought
- A loyalty scheme written against Shopify and Klaviyo rather than licensed off the shelf, so the rules match how boots are actually sold and returned.
- Referral, not just points
- Four flows carry it: a code issued to a friend, credit released once they buy, credit on redemption, and a birthday credit. The referral arm waits five days, then splits on whether the friend has ordered in the last thirty.
- What it earns
- 25% of annual revenue arrives through email and SMS.
The emails, as they arrive
Nine of the messages the flows send, in the order a customer meets them.
Rendered from the live Klaviyo templates with a sample name. Tap any one to read it in full.
Leadership
Eight people, thirty events,
one set of numbers.
Most senior marketing and digital decision-maker at Bareback, reporting directly to the owners. The team was small enough that everyone could see what their work did to the P&L, and it was built that way deliberately.
8
at peak, across marketing,
warehouse and events
1
direct report: social media
and marketing manager
30
trade events a year,
organised and operated
2
owners, reported to
directly
Who I hired and directed
A social media and marketing manager as a direct report, plus commissioned videographers, photographers and influencers. Creative was briefed against what the ad account said was working, not against a mood board.
Giving everyone the numbers
Automated daily reporting for owners, directors and staff, so sales, media and operating performance were visible to the people doing the work rather than locked in a monthly deck.
Knowing when to shrink something
Designed a lower-risk wholesale operating model that kept the channel and its reach while moving weight and working capital to the faster-growing online business. Not every decision is a growth decision.
The first ninety days
What I'd do first,
in a new brand.
The pattern is the same whatever the product. Find where the money leaks, stop the leaks, then spend behind what has proved itself.
Days 1 to 30
Listen and measure
Read the ad accounts, the CRM, the returns and the margin by product before touching anything. Sit with the people who answer the customer emails. Agree one set of numbers that everyone sees, every day.
Days 31 to 60
Stop the leaks
Fix the product page before buying more traffic. Put fit and sizing where the decision is made. Rebuild the flows that have stopped earning. Brief creative against what the account says is working.
Days 61 to 90
Scale what is proven
Put the spend behind the product that has proved itself, grow it into a range, and hold the margin line while doing it. Growth that costs margin is not growth.
John Spilsbury Digital
The same work,
for other people's brands.
Eight consumer, local and online businesses since 2018, alongside the day job. Paid media, Shopify build, CRO and the commercial thinking that decides which of those to do first.
Leather goods, UK
Meta strategy, website improvements and growth planning for an established British bag brand.
Agency partner
Meta, CRO and Shopify recommendations feeding into larger client engagements.
Jewellery, two brands
Meta and Google Ads managed end to end: commercial goals translated into campaign structure, then into practical site changes.
Local & independent
Focused local campaigns for smaller operators, where the whole budget has to work the first time.
Client names and references on request.
Contact
Fourteen years of this,
and still curious.
If you are hiring someone to own digital growth, or you have a brand that has stalled somewhere between £1m and £10m — I would like to hear about it.
- Emailj.e.t.spilsbury@gmail.com
- LinkedInjohn-spilsbury
- BasedWorcester, UK
- Open toDirector roles & select consultancy
- CVTwo pages, PDF
- EducationBA (Hons) Human Geography, Edge Hill
- ElsewhereMountain ultrarunner · Lakeland 50 finisher