Email Marketing Automation Case Study in Pakistan

The Client

A Faisalabad-based D2C professional haircare brand — keratin treatment kits, bond-repair serums, color-protection shampoos and conditioners — founded by a formulation chemist who had spent a decade supplying salons before launching the retail line. The brand sold through its own Shopify storefront to two very different audiences at once: home users managing color-treated and chemically processed hair, and more than 300 beauty parlors and freelance stylists across Punjab who bought salon-size tubs and restocked every few weeks.

Online revenue ran at roughly PKR 9.8M a month. Paid acquisition — mostly Meta, with some Google Shopping — carried the growth at a blended 3.2x return, but acquisition costs had climbed 19% year on year and the founder could see the ceiling approaching. Meanwhile the brand’s 22,400-subscriber email list, collected through pop-ups, checkout opt-ins, and a genuinely popular haircare quiz, was being used as a noticeboard: one broadcast a fortnight, product announcements, the occasional sale.

The strange part was that they were already paying for a full Klaviyo plan. A previous agency had set up the account, switched on its two default flows — abandoned checkout and a half-finished welcome email — and moved on. The platform sat there collecting order data it was doing nothing with. When the founder approached WeProms Digital, the brief was specific: stop paying for retention infrastructure that retains no one, and build email marketing automation lifecycle flows that treat home users and salon professionals as the two different businesses they actually are.

The Problem

Five issues were draining revenue every month, and none of them were about the platform itself:

1. A paid platform running on defaults. Two flows were live. The welcome email had been erroring silently for months — a broken quiz integration meant the trigger event never fired for 61% of new subscribers, so thousands of sign-ups never received a first message. The abandoned-checkout flow used the stock template: no cash-on-delivery objection handling, no courier timeline, nothing that answered why Pakistani buyers actually abandon.

2. Two audiences, one message. Salon professionals — roughly 1,900 identifiable accounts contributing about 34% of revenue — received consumer discount codes, consumer-size product suggestions, and “treat yourself” copy. They order in tubs, ask for invoices, and restock on a four-week rhythm; none of that was recognized anywhere. Home users, meanwhile, received no aftercare education at all, in a category where using the product wrong visibly damages results.

3. No replenishment logic. A keratin kit lasts a home user roughly 45 days. A salon tub lasts about four weeks at twenty-plus applications. Shampoos run out in five. Nothing in the programme knew any of this — the single highest-value behavior in the category, restocking on schedule, was left entirely to the customer’s memory.

4. Deliverability quietly failing. Sixty-two percent of the list had not opened or clicked anything in 90+ days, and the account kept mailing all of it. Seed-list tests put inbox placement at 82%, with Gmail spam and promotions placement concentrated on the largest sends — which is exactly why the fortnightly broadcast looked fine on paper and did almost nothing.

5. An education gap that cost real money. Keratin aftercare has rules — no sulfate shampoos, no washing for 72 hours, specific pH ranges — and support was handling roughly 190 usage tickets a month. A meaningful share of complaints and refund requests traced back to customers simply not knowing how to use what they had bought.

The net effect: email contributed about PKR 0.6M of a PKR 9.8M month — 6% — from a list that should have been the brand’s cheapest revenue channel.

Phase 1 — Diagnosis, Hygiene, and Data Repair (Weeks 1-2)

Before building anything, we needed to know what the list could actually do and who was on it. The first two weeks were forensic.

Engagement audit and sunset. We tiered the 22,400-subscriber file: about 8,500 genuinely active (opened or clicked in 90 days), 4,100 warming (91–180 days), and 9,800 cold. Rather than deleting the cold tier outright, we ran a three-email sunset programme with a final incentive — it revived 640 subscribers, a useful reminder that “inactive” is not always permanent — and suppressed the remaining 9,160 from promotional sending. Suppression is the deliverability move that pays for every flow you build afterwards.

Sender repair. SPF, DKIM, and DMARC were either missing or misconfigured on the sending domain. We authenticated everything properly, moved bulk sending to a dedicated subdomain so the primary domain’s reputation stayed clean, and rebuilt the account as a properly instrumented Klaviyo email marketing setup. Seed-list inbox placement moved from 82% to 92% by the end of week two.

Data repair — the part that made everything else possible. We classified the file into home users (17,600), salon professionals (1,900), and unknown (2,900) using order-notes mining (pros leave telling traces: bulk quantities, invoice requests, salon names in shipping fields), quiz answers, and a two-question post-signup profile. We verified that viewed-product, added-to-cart, started-checkout, and ordered events fired on every template — including the quiz landing page the previous agency had never instrumented, which is why the welcome flow had been silently broken.

Diagnostic	Before	After cleanup (end of week 2)
Active subscribers (90-day)	8,500 (unisolated)	8,500 isolated for measurement
Cold tier	Mailed equally	640 revived, 9,160 suppressed
Inbox placement (seed test)	82%	92%
Domain authentication	Partial	SPF, DKIM, DMARC + sending subdomain
Event tracking	2 of 5 events	All 5 verified incl. quiz funnel
Audience classification	None	Home / pro / unknown split built

The list got smaller on paper and larger in practice: from week three onward, every send reached people with a real chance of buying, and every flow could branch on whether it was talking to a home user or a salon.

Phase 2 — Core Flows and Dual-Audience Segmentation (Weeks 3-6)

With a clean, deliverable, classified list, we built the core flow architecture. The design principle: every customer moment that mattered got an automated response, and every response had a version for each audience.

Dual-track welcome series. Sign-up source determined the track. Home users entered a four-email series: an immediate brand introduction with a 10% first-order incentive, a hair-concern education email keyed to their quiz answers (color-treated, keratin-treated, damaged, oily scalp), a social-proof email built from customer before-and-afters, and a final urgency nudge before the incentive expired. Salon professionals entered a three-email track with trade pricing, bulk sizing, and the invoice process — no consumer discount language anywhere. In the first two weeks live, the home track converted 9% of new subscribers to a first purchase.

Post-purchase aftercare education (3 emails, days 2/9/20). Product-specific and unapologetically instructional: the keratin-kit buyer learned the 72-hour rule, the sulfate rule, and when to book a re-treatment; the color-care buyer learned wash frequency and heat protection. Day 20 checks in (“how is it going?”) and introduces the matching sulfate-free duo. This flow became the quiet workhorse — usage tickets started falling within three weeks, and it fed the cross-sell engine without a single discount.

Cart recovery (3 emails, hour 1 / day 1 / day 2). Rebuilt from the stock template to answer the questions Pakistani buyers actually have at checkout: COD availability, courier timelines to their city, and the returns process. The day-2 email added a free-shipping threshold nudge, since delivery-fee surprise was the single most cited abandonment reason in the exit survey we ran in week one.

Pro reorder flow. Every salon order entered a restock cycle: a day-28 reminder with one-click reorder of the same tub sizes and quantities, plus a stock-check prompt before Eid and wedding-season demand spikes. Pros do not need persuasion — they need the reorder to take eleven seconds.

Browse abandonment and VIP flows. Repeat product viewers without a purchase received a two-email nudge featuring the exact SKUs they had viewed. The top 10% of customers by lifetime spend entered a VIP flow with early access to launches and restock priority — wedding season makes stock-outs a loyalty problem in this category.

Flow	Live	Early signal (first 2 weeks)
Welcome (dual-track)	Week 3	9% home-track subscriber conversion
Post-purchase education	Week 4	Usage complaints visibly declining
Cart recovery	Week 4	12% abandoned-cart recovery
Pro reorder	Week 4	21% of pro orders reordering within 35 days
Browse abandonment	Week 5	First attributed orders
VIP early access	Week 6	Highest open rate of any flow (48%)

By the end of week six, email’s revenue share had moved from 6% to 13% — and the replenishment engine, the piece the whole programme was built around, had not even launched yet.

Phase 3 — Replenishment Engine and Campaign Rebuild (Weeks 7-11)

The consumption math. We built a per-SKU-class replenishment table from actual usage, not folklore:

Product class	Typical consumption	Trigger day
Keratin kit (home)	~6 applications over 45 days	Day 42
Keratin tub (salon)	~22 applications over 4 weeks	Day 28
Shampoo / conditioner	Daily use, ~5 weeks per bottle	Day 35
Bond-repair serum	Weekly use, ~10 weeks per bottle	Day 68

Flow mechanics. Each replenishment touch is a two-email sequence: a “running low?” email on the trigger day with a one-click reorder of the exact variant bought, and a day-later follow-up with a modest restock incentive for second-order hesitation. The logic is variant-aware (the 500ml buyer triggers differently from the 250ml buyer) and self-correcting — an early repurchase suppresses the sequence and re-anchors the clock from the new order date. No subscription commitment, which Pakistani beauty buyers consistently decline; just memory.

Campaigns rebuilt around segments. The fortnightly broadcast became weekly, segment-targeted sends: concern-led education and restock prompts for home users, stock-planning and trade updates for pros, early access for VIPs. Every send carried a subject-line A/B test — subject lines mixing English and Urdu outperformed English-only in 9 of 12 tests, a finding worth more than the entire creative refresh the brand had been quoted for.

Win-backs tuned by audience. Home buyers lapse at 75 days, pros at 45 — a salon that has not reordered in six weeks is already buying somewhere else. Both sequences escalate: a check-in, a concern-based nudge, a final incentive, then suppression to protect deliverability.

Metric	Week 6	Week 8	Week 11
Email revenue share	13%	17%	22%
Replenishment flow revenue	0	PKR 0.4M	PKR 0.7M
Open rate (active list)	27%	31%	34%
Pro 35-day reorder rate	21%	26%	31%

Phase 4 — Test, Attribute, and Compound (Weeks 12-17)

The final phase turned a working programme into a compounding one.

Flow-level testing. We tested the highest-leverage decisions rather than button colors: replenishment trigger timing (day 42 vs day 49 for keratin kits — day 42 won, lifting orders per thousand recipients by 18%, because later triggers surrendered restocks to whichever salon or retailer the customer walked past first), cart-recovery incentive shape (10% off vs a flat PKR 500 — the flat amount won on margin-adjusted revenue), and welcome incentive type (10% off vs free shipping — free shipping won on first-order AOV and downstream repeat rate).

Attribution layer. A shared dashboard broke revenue down by flow, segment, and campaign, with deliverability and list health alongside. This is what made the final numbers defensible instead of vibeful: by day 120 the brand’s owner could say “the replenishment engine alone does PKR 1.2M a month” and point at the specific flow.

The education dividend. Usage tickets settled at roughly 110 a month, down 42% from the 190 baseline — fewer complaints, fewer refund conversations, and support time redirected to pro-account service.

By day 120, email was generating roughly PKR 3.5M of a PKR 12.6M month — 28% — while paid spend stayed flat. The share gain was additive, not redistributive: total revenue grew because the brand stopped renting its existing customers from Facebook.

Final Results at 120 Days

Metric	Before	At 120 days	Change
Email revenue share	6%	28%	+22 pts
Email-attributed revenue (monthly)	PKR 0.6M	PKR 3.5M	+483%
Replenishment flow revenue	PKR 0	PKR 1.2M	New engine
60-day repeat purchase rate	21%	33%	+12 pts
Open rate (active list)	16%	36%	+125%
Click rate	0.8%	3.1%	+288%
Inbox placement (seed test)	82%	97%	+15 pts
Product-usage support tickets	~190/mo	~110/mo	-42%
Pro 35-day reorder rate	~8%	31%	+23 pts

Every number traces to a phase: deliverability and classification came from the week 1-2 cleanup, the core flow coverage from weeks 3-6, the replenishment engine and campaign discipline from weeks 7-11, and the durable 28% share from the testing and attribution work in the final stretch.

What Made This Work

1. Consumption math over generic timing. The replenishment engine worked because trigger days were calculated from applications per bottle and appointments per week, then validated by test. A generic “30 days after purchase” rule would have fired weeks late for salon tubs and weeks early for serums — annoying one audience and losing the other entirely.

2. Two audiences, two programmes. Separating 1,900 salon professionals from 17,600 home users changed everything downstream: pricing logic, education depth, reorder cadence, win-back timing. The pro reorder flow alone moved 35-day pro reorder rate from single digits to 31%, and it cost nothing but classification.

3. Discipline beats tooling. The brand already paid for the platform. The gains came from programme design — flow coverage, engagement-based sending, suppression, testing. Most “email problems” in Pakistani D2C are not platform problems; they are the absence of anyone owning the programme.

4. Deliverability first, always. Suppressing 9,160 dead subscribers felt like destroying an asset the brand had spent three years collecting. It was the opposite: it is why inbox placement reached 97% and every subsequent flow had a fair chance to perform.

5. Education as a revenue flow, not a support cost. The aftercare series cut usage tickets 42% while feeding cross-sells — the same content doing retention and service work simultaneously.

What Teams Can Apply

For Pakistani D2C brands in haircare, skincare, supplements, or any consumable category:

1. Calculate your product’s real consumption cycle before building flows. Trigger restock emails from how long the bottle actually lasts your average customer, then A/B the day. Timing is the entire mechanism.

2. Separate your business buyers from your consumers. If you serve salons, clinics, gyms, or retailers alongside home users, their reorder cadence, pricing, and objections are different. One list treated identically taxes both groups.

3. Sunset your inactive subscribers before you build anything else. A bloated list drags every send toward Gmail’s spam folder. Re-engage once, suppress the rest, and measure inbox placement — not just open rates — to know where you stand.

4. Cover the core moments before optimizing anything. Welcome, post-purchase education, cart recovery, replenishment, win-back. These five flows will out-earn any redesign of your newsletter template.

5. Test subject lines in the language your customers actually read. Mixed English-Urdu subject lines won 9 of 12 tests here. It costs nothing and compounds across every future send.

WeProms Digital has applied this lifecycle framework across Pakistani D2C brands in beauty, supplements, fashion, and food. The flow logic, timing, and creative change with each product cycle — but the sequence of hygiene first, classification second, flows third, and testing fourth stays the same.