Pott Candles · Welcome Flow: Current vs. Recommended

The synthesis

The merged flow: what to build

The recommended sequence, one job per email. The live renders below show exactly what each change looks like; the open point to align on is at the end.

Email 1 Immediately

Deliver the promise + plant the seed

The £5 code, one in-home hero, and a single refill hook. One CTA to Shop Starter Packs, nothing else.

"What is Pott, and where's my offer?"

Email 2 +2 days

Why it's worth it

The full refill story: handmade, refillable forever, craft proof. Framed as value, no price named.

"Why is this worth it?"

Email 3 +2 days

Reduce choice anxiety

Core scents and the scent wheel (rebuilt as live text), plus what's inside a Starter Pack.

"Which one should I choose?"

Email 4 +2 days

Sell the dream

The world of Pott: seasonal scents and future discovery. Seeds subscription without ever naming it.

"What makes it special long-term?"

Email 5 +1 day

A branded last-call: "£5 ends tonight," a real expiry, and your single best review.

"Why buy now?"

SMS 1 (≈1 min after E1): welcome + £5 offer + code, the quiet winner (33% click), leave it. · SMS 2 (after E5): expiry reminder, fix the "coce" → "code" typo (live, 0 conversions).

Execution layer (applies to every email): re-export photos as JPG (<1 MB/email); one primary CTA each; no text baked into images; add real customer/UGC + visible proof; explicit colours for dark-mode safety; keep all product prices off the emails.

To align & review together before building

Where the New Flow and this audit diverged. The first three are resolved; the last is the one open point to confirm together before the flow is built.

Decision · Refill differentiator

Both E1 and E2. A short hook in E1 (most people open E1 just to grab the code, so don't rely on it landing) and the full deep-dive in E2 (the safety net that catches everyone who skimmed E1).

Decision · Subscription

Not named in the flow. E4 "sells the dream" (seasonal discovery / the world of Pott), seeding the recurring relationship emotionally without an upsell. Preferred over naming subscription outright.

Decision · E5 close

Simple, but branded. Lean and direct as proposed, but with a Pott header + explicit background/text colours so it doesn't break in dark mode like the current one.

⚠️ Pricing · to align before building

No prices in the flow, value only. The recommendation is to build value so price isn't a friction point by the time they reach the PDP. The New Flow proposal suggested naming "£55" in E2. This is the one open point to confirm together before the flow is built.

Email by email

Live render · what's in it now · what I'd change

"You're in 💫"

Immediate · Preview: "Open me, your welcome treat is inside!"

£7,537 · 74.5% open

Carries the thesis

Live mobile render

What's there now

Generic warm welcome: "early access, exclusive offers, latest updates."
£5 off + code Starter-DV46XFF; "explore our range."
Strong design: hero, product grid, Instagram strip, press logos (Good Housekeeping, House Beautiful).
Says nothing that distinguishes Pott; "refill" appears only in the nav.
Offer wording clashes: hero "£5 off first order" vs body "first Starter Pack."
Starter Pack link → the -old-copy PDP (the CRO-flagged page).
Two competing CTAs: GET STARTED + SHOP NOW.

What I'd change

Lead with the one reason Pott is different: buy the Pott once, refill forever: cheaper over time, less waste. This is the price-justifier AND the differentiator.
Keep the beautiful design; re-point the copy, not the layout.
Collapse to a single primary CTA → Shop Starter Packs.
Add one short customer review (you have press logos; add a human voice).
One line of expectation-setting ("here's what to expect from us") to slow unsub bleed.
Fix offer wording + repoint link off -old-copy: safe to do now.

Why here: 74.5% of people open this and almost no one reads past it. If the differentiator isn't in E1, it effectively isn't in the flow.

🎨 Design & imagery

Now: hero is thin white script over a busy product cluster (the £5 line reads small) and leads with product, not a room; the "Explore the range" tiles are labelled and clickable but use inconsistent crops; the press logos are legible but buried in the footer. Fix: high-contrast headline + offer over a styled in-home hero; one consistent crop across the grid (keep price out, value before price); lift a press/review strip higher.

"Discover our Story 💗"

+2 days · Preview: "Find out more about us and meet our founder…"

£120 · 2.0% click

Becomes the proof email

Live mobile render

What's there now

Lovely founder story: Ailis, lockdown origin, 360M UK candles/yr, handmade UK artisans.
"Small changes, big impact." Signed "Ailis xx." READ OUR STORY CTA.
Genuinely strong brand-building: design and voice are right.
Doing the differentiation job that E1 should already have done.

What I'd change

Re-cast as the proof & price-justification email now that E1 carries the USP.
Lead with craft: handmade, UK artisans, "each Pott is unique": this is what the £ buys.
Add collaborations (e.g. IOW Tomatoes) as credibility + brand depth, a natural fit here.
Keep the founder voice; add a soft Starter Pack CTA + code reminder.

Why here: the "why does it cost this?" objection is best answered with craft + collaborations once belief is already seeded in E1.

🎨 Design & imagery

Now: the strongest imagery in the flow: the founder shot + hands-making / pouring shots are the craft proof that justifies the price. It's all brand-side, though. Fix: lean into these even harder; add one genuine customer-in-home photo for warmth and proof.

"How do Pott refills work?"

+2 days · Preview: "Get one Pott with endless refills…"

£169 · 1.4% click

Best asset · home of the savings ladder

Live mobile render

What's there now

Clean step-by-step refill guide + "One Pott. Many Refills." SHOP NOW.
The single best-designed asset in the whole flow.
Stranded at position 3 with a 1.4% click: teaching the USP where nobody's looking.

What I'd change

Promote its essence into E1; keep the full visual guide here as reinforcement for slow deciders.
Make the savings ladder obvious: this is awareness of a price benefit, not a pitch to subscribe:
→ a new candle every time = the priciest, most wasteful option
→ refills = much cheaper (you already own the Pott)
→ subscribe = cheaper still, and your favourite scent always arrives: never run out, never reorder
Frame it as "spend the least and never think about it again," not "now subscribe."
Add social proof (a review or "X,000 Potts refilled").

Why here: subscription isn't a separate upsell; it's the cheapest rung of the refill story, so it belongs right where refills are explained. The customer only needs to hear the saving and the convenience; the loyalty/LTV upside (subscribers repeat 86.6% vs 36%) follows on its own.

🎨 Design & imagery

Now: the best-working imagery in the flow: the Step 1–6 guide genuinely teaches and even shows the refill going into the Pott. One catch: it's a single tall image with the step text baked in, so the copy gets small on mobile. Fix: keep the layout, but rebuild it as live text + images (not one flattened graphic) so it stays legible and dark-mode-safe.

"Find your perfect scent 🌸"

+2 days · Preview: "With the help of our scent wheel…"

£316 · best back-half · 0.33% unsub

Mostly keep

Live mobile render

What's there now

Scent Wheel + core scents; DISCOVER OUR SCENTS / SHOP NOW.
Well-placed "now go choose" nudge: best slot of the back half (lowest unsub).
Several duplicate CTAs stacked (SHOP NOW + DISCOVER repeated).

What I'd change

Keep the role, light touch only.
Bridge scent → Starter Pack type: which pack (Petite / Standard / Grand) suits them, so the nudge lands on a product, not a page.
Trim duplicate CTAs to one clear primary.

Why here: it already converts best of the back half; don't over-engineer, just point it at the packs.

🎨 Design & imagery

Now: appetising, on-brand botanical scent imagery (figs, citrus, greenery). But the Scent Wheel renders far too small to read on mobile, and the DISCOVER buttons repeat. Fix: make the wheel a larger tappable / simplified-for-mobile graphic; trim to a single CTA.

"Your discount is about to expire!"

+1 day · Preview: "Be quick, you wouldn't want to miss out…"

£174 · 3.1% click

Bare · breaks in dark mode

Light mode: bare & unbranded, but legible

Dark mode: inverts to black, looks broken

What's there now

Short founder note: "expires in a few hours," code, "treat yourself or a loved one."
Bare plain text: no header, logo, button or images; renders on a blank background.
Urgency may not be real: the code shows no visible expiry.

What I'd change

Rebuild as a properly branded last-call (Pott header + a real button), or commit fully to a deliberate clean "note from the founder" style, not an accident.
Verify the code actually expires. If it doesn't, make it real or soften the claim: false urgency trains people to ignore deadlines.
Add the single best review as a final nudge.

Why here: it's the last paid-attention moment before they go cold; right now the format undercuts the urgency.

🎨 Design & imagery

Now: bare text with no header, logo, image or button. In light mode (left) it's legible but completely unbranded; in dark mode (right) it inverts to black because no background or text colours are set, and looks broken. Fix: rebuild with a Pott header + hero + a real button (or a deliberately minimal but branded founder note), and set explicit background + text colours so it can't invert in dark mode.

✆

The two SMS

Welcome SMS (~1 min after E1) · Expiry SMS (after E5)

Welcome: £501 · 5.0% CVR

Expiry: live typo · £0

What's there now

Welcome SMS: "Enjoy £5 off your first Starter Pack…" 33.3% click, 5.0% CVR, £501. A quiet winner.
Expiry SMS: live typo: "Use the coce Starter-DV46XFF…", 0 conversions, going to real subscribers now.

What I'd change

Leave the welcome SMS alone, it's pulling its weight.
Fix "coce" → "code" immediately. One-word edit, best effort-to-impact in the whole audit.
Reassess whether the expiry SMS is redundant vs E5 / poorly timed (0% conv suggests one or the other).

Going deeper

Imagery & design: the full review

What's already working: the foundation to build on

Across E1–E4 the design system is genuinely premium and cohesive: the cream / blush palette, the elegant serif "Pott · The Candle Refill Co." lockup, and generous whitespace. You also own a strong photo library most candle brands don't have: founder portraits, artisan hands-making shots, scent botanicals, and the step-by-step guide. Keep the photography and the ideas. One nuance: a few of these assets flatten their text into the image; that's the execution fix in point 1 below, not a knock on the art or the concept.

E2 craft shot — E2: founder & craft (the price-justifier)

E3 step guide — E3: the step guide's teaching is excellent (the concept)

Stop baking text into images

E4 scent wheel — E4 wheel: labels turn to mush on mobile

E3 steps — E3: great guide, but the text is flattened in

The hero headline + offer, the Scent Wheel, and the refill step guide are baked inside image files, so the words shrink with the picture and become unreadable on a 375px phone, where most of these are opened. (The photography and the step guide's concept are strong; see "what's working" above; this is purely about how the text is delivered.)

Keep critical words (headline, offer, scent labels, step copy) as live HTML text over or beside the photo; reserve image files for the photography itself.

Seen in: E1 hero, E4 scent wheel, E3 steps

Fix the hero: contrast + aspiration

It's thin white script over a busy product cluster ("Welcome to Pott" reads, but the "£5 off" line is small and low-contrast) and it leads with product on a plinth, not a room. (It also says "first order," feeding the offer-wording mismatch flagged in the copy.)

Give the headline/offer a high-contrast treatment, and lead with a styled in-home moment (a lit Pott on a shelf, bath or bedside) so the first thing they feel is atmosphere, then the offer.

Seen in: E1 (also E3/E4 banner headers)

One image recipe, held across the flow

Even within E1's four-tile grid the treatments clash (lifestyle, lit-product, dark boxed, product-on-pastel) with different lighting and crops, and it sits beside E2's warm lifestyle look. The flow reads like several brands stitched together.

Pick one treatment (I'd go lifestyle-forward for this brand) and apply it consistently so the five emails feel like a single set.

Seen in: E1 grid vs E2 / E4

Make the range a clean, clickable gateway

Credit where due: the tiles are labelled (Petite / Standard / Grand Starter Pack, Join the Refill Club). The only real issue is visual: the four images clash in background, lighting and crop, so the set reads like a moodboard rather than a considered gateway.

Standardise the image treatment + crop and make each tile an obvious tap-through. Deliberately keep price out of the email: value before price is the whole strategy here, so let the grid earn the click and let the PDP introduce price once the value has landed.

Seen in: E1

Bring in real customer / UGC photography

studio imagery — Current imagery is all studio / brand-made

Every photo is studio / brand-made (lovely, but staged). There isn't a single real customer's Pott in a real home anywhere in the flow.

Add genuine UGC: customers' Potts in situ. It warms the brand, supplies the social proof the audit flagged as missing, and this product photographs beautifully in homes.

Seen in: all emails

Make the proof actually visible

E1 press logos — Legible logos, but stranded in the footer

The press logos (Good Housekeeping, House & Garden, House Beautiful) are perfectly legible, but they're buried in the footer, below the social icons, so the credibility lands late and weakly. And there are no customer reviews shown anywhere in the flow.

Pull a press strip higher (E1, above the fold), and add a short star/review quote to E1 and one nurture email.

Seen in: E1, E2, E5

Dark-mode safety

E5 dark — E5 in dark mode: inverts, looks broken

E5 proves the risk: identical email, but with no explicit colours it inverts to light-text-on-black in a dark-mode client and looks broken. The other emails could invert unpredictably too.

Set explicit background and text colours on every email so dark-mode clients don't re-colour them.

Seen in: E5 (live, both modes shown) · risk across all

CTA discipline: one primary action

DISCOVER OUR SCENTS button repeated — …the same button again, lower in E4

E4 runs the identical "DISCOVER OUR SCENTS" button twice (and E3 stacks SHOP NOW + GET STARTED), which dilutes the single action you want and adds visual noise.

One clear primary CTA per email; demote or remove the rest.

Seen in: E1, E3, E4

⚠️ Image weight & load time: the biggest technical issue

The photos are uploaded as PNG (a format meant for graphics/transparency) instead of JPEG or WebP (made for photographs). The result: emails that are 3.6–7.2 MB each. On mobile data that means images load slowly, or don't render before the reader bounces. This is almost certainly the same issue you see in the campaign emails; it's an asset-handling habit, not flow-specific. (Measured from the live Klaviyo/CloudFront image URLs, deduped per email.)

E1	3.6 MB	15 images
E2	7.2 MB	worst: 3 photos = 6 MB
E3	4.2 MB	step graphic = 2.2 MB
E4	4.4 MB	7 heavy photos
E5	0 MB	text only

Best-practice ceiling for a marketing email is roughly ≤ 1 MB total, with individual photos under ~200–250 KB. Every image-bearing email here is 3–7× over.

Recommendation: re-export the photographs as JPG instead of PNG, typically 80–90% smaller with no visible quality loss (e.g. E2's 2.55 MB craft PNG → ~200 KB). Keep widths to ~1080px max (plenty for retina). Target: every email under ~1 MB, individual photos under ~250 KB. Bonus: converting the flattened text-graphics (hero, scent wheel, step guide) to live text + lighter photos cuts weight and fixes the mobile legibility issue in point 1.

How to fix it: exporting from Canva

The photos just need re-exporting in the right format and re-uploading to Klaviyo. No redesign required.

Open the design (or photo) in Canva. To export a single image rather than a whole page, select that element first.
Click Share (top-right) → Download.
Under File type, choose JPG: this is the single biggest fix (JPG is 80–90% smaller than PNG for photos). Only keep PNG if the image genuinely needs a transparent background (e.g. a logo over a colour).
Size / dimensions: aim for about 2× the size the image displays at in the email, roughly 1080–1200px wide for a full-width image. That keeps it crisp on retina screens while staying light. (Pott's images are already ~1080px, so for these you don't need to enlarge, just re-export. The 2× rule is for when you're starting from a smaller canvas. Don't go to 4× / ~2160px, it adds weight for no visible gain.)
If you have Canva Pro: tick "Compress file (lower quality)." Exporting a touch larger and compressing harder actually looks cleaner: the email shrinks the image back down, and that downscaling hides any compression artifacts.
Click Download, then in Klaviyo replace the old PNG with this new JPG in the image block.
Sanity check: each photo should land under ~250 KB (a few hundred KB, not MB). If one is still heavy, compress a bit more or drop toward 1.5× display size.

Rule of thumb going forward: photos → JPG; logos/icons/anything with transparency → PNG; size ≈ 2× display, then compress. Applying this to the campaign emails too should fix the slow-loading issue across the board, not just this flow.

If you only fix three for design: (1) rebuild E5 so it's branded and not broken; (2) fix the E1 hero (contrast + a real in-home image) and standardise the range grid's crops; (3) stop baking headlines and the scent wheel into image files so mobile can read them.

The overall play

Overall strategy

The strategy you described is the right one. The problem isn't the arc; it's that the arc is spread thin across emails almost nobody opens. Keep the goal (convert a new customer on a Starter Pack), but compress the payload into where attention actually is, and let the back half reinforce and seed subscription.

Front-load the "why"

The refill story is the differentiator, the price-justifier, and the subscription seed in one. It belongs in E1, where 74.5% of attention lives, not stranded at E3.

Handle the objection with craft + collabs

"Why does it cost this?" is answered by handmade UK artisans and real collaborations. That's E2's new job: proof, not just a nice story.

Frame subscription as the cheapest rung, not an upsell

Don't chase the subscribe. Make the price benefit obvious: refilling already beats buying a new candle every time, and subscribing is cheaper still: favourite scent always on the way, nothing to reorder. Awareness of a saving, not a grab.

Quick wins: safe to action right away

Fix the "coce" → "code" SMS typo do now
Repoint E1's Starter Pack link off the -old-copy PDP, and align "first order" vs "first Starter Pack" wording safe
Verify whether Starter-DV46XFF actually expires (gates the credibility of E5 + the expiry SMS)

The honest summary: offer → differentiate → justify cost → introduce scent & packs → social proof → convert is a strong sequence. The single highest-leverage move is moving the refill USP and one piece of social proof into E1, because everything downstream is read by a small and shrinking audience. Everything else is reinforcement.

Welcome Flow: What's There Now vs. What I'd Do

Where the money, and the attention, actually is

The same arc, re-pointed at where people are paying attention

CURRENTLY: a linear journey that assumes people walk all 7 days

RECOMMENDED: payload front-loaded, back half reinforces & seeds subscription

The merged flow: what to build

Decision · Refill differentiator

Decision · Subscription

Decision · E5 close

⚠️ Pricing · to align before building

Live render · what's in it now · what I'd change

What's there now

What I'd change

What's there now

What I'd change

What's there now

What I'd change

What's there now

What I'd change

What's there now

What I'd change

What's there now

What I'd change

Imagery & design: the full review

What's already working: the foundation to build on

Stop baking text into images

Fix the hero: contrast + aspiration

One image recipe, held across the flow

Make the range a clean, clickable gateway

Bring in real customer / UGC photography

Make the proof actually visible

Dark-mode safety

CTA discipline: one primary action

⚠️ Image weight & load time: the biggest technical issue

How to fix it: exporting from Canva

Overall strategy

Front-load the "why"

Handle the objection with craft + collabs

Frame subscription as the cheapest rung, not an upsell

Quick wins: safe to action right away