Results: airmirate | Real Ecommerce Automation Case Studies, Abuja
FRAGILE: HANDLE WITH AUTOMATION
Results, real pipelines

Five stores. Five automation pipelines. Real numbers.

Not mockups, not projections. Here's exactly what we built, how we built it, and what changed, across WooCommerce, Shopify, social media DM commerce, and BigCommerce.

CASE 01 / 05
Mike S., Hingees
WooCommerce Clothing Brand Lagos, Nigeria

The situation

Mike runs a clothing store, a WooCommerce store selling apparels to customers across Abuja and Lagos. Orders had grown steadily for a year, but so had the WhatsApp and email pile of "where is my order" messages. She was manually checking WooCommerce order statuses against her courier's tracking page every morning, then copying updates into individual replies. Some customers waited two full days for a response. Returns were worse: each one meant a back and forth of five or six messages before a refund was even approved.

What we built

1
Audited the manual process

Shadowed Mike for two days to log every WISMO message, return request, and how long each took to resolve by hand.

2
Connected WooCommerce to n8n

Set up a webhook so every order status change in WooCommerce triggers an automation instantly, no manual checking required.

3
Built the WISMO auto-reply flow

Order status pulls from WooCommerce and the courier API, merges into a message template, and sends automatically by WhatsApp and email.

4
Automated return approvals

Return requests are checked against a rules engine (item condition, days since delivery); simple cases auto-approve, edge cases route to Mike on Slack.

5
Tested with real order data

Ran the flow against two weeks of historical orders before going live, to catch edge cases like split shipments and address changes.

6
Launched and monitored

Went live gradually, watching the first fifty automated replies closely before letting it run fully unattended.

The result

91%
WISMO tickets auto-resolved
7min
Avg. response, down from 2 days
6hrs
Saved weekly on support alone

"Running three platforms used to mean three separate messes. airmirate put everything on one line and it just works."

Mike S.
Founder, Hingees
CASE 02 / 05
Tariq A., Layer & Line
Shopify Fashion and apparel Maryland, USA

The situation

Layer & Line sells streetwear on Shopify, with a spike in traffic every time Tariq posts a new drop on Instagram. The problem: most of that traffic added items to cart and never checked out. Tariq could see the abandoned checkout count in Shopify but had no system to follow up, so every drop leaked sales he never got back. Manually messaging even the biggest carts took hours he didn't have between sourcing and packing orders.

What we built

1
Mapped the drop cycle

Studied three previous drops in Shopify analytics to see exactly when and why carts were abandoned during high traffic spikes.

2
Connected Shopify checkout events to n8n

Set an automation trigger on abandoned checkout, capturing customer, cart contents, and time since abandonment.

3
Built a three-touch recovery sequence

SMS at 1 hour, email at 6 hours with a soft nudge, and a final WhatsApp message at 24 hours with a small time-limited discount.

4
Added stock-aware logic

If the abandoned item sold out in the meantime, the flow automatically swaps in the closest in-stock alternative instead of a dead link.

5
Load-tested for drop-day traffic

Simulated a full drop-day spike to confirm the automation held up under sudden volume, not just steady daily traffic.

6
Launched on the next live drop

Went live during an actual product drop, watching recovery rates in real time and adjusting message timing on the fly.

The result

2.8x
Cart recovery rate
₦1.4M
Recovered in first drop cycle
0
Manual follow up messages sent

"I genuinely didn't believe the number until I saw it in Shopify myself. Carts that would have just disappeared are now actual orders, and I didn't touch a single one of them."

Tariq A.
Founder, Layer & Line
CASE 03 / 05
Funmi E., Voltage Gadgets
Shopify Electronics and gadgets Abuja, Nigeria

The situation

Voltage Gadgets sells phone accessories and small electronics on Shopify, a category with a specific problem: customers ask detailed pre-purchase questions (compatibility, warranty, charging specs) before they'll buy, and post-purchase questions (defects, warranty claims) after. Funmi's small team of two was fielding both, all day, across Shopify's inbox, Instagram DMs, and email, with no way to prioritize which question actually needed a human.

What we built

1
Catalogued common questions

Pulled six months of support history to identify the 20 questions that made up 80% of all inbound messages.

2
Built a product-aware AI agent

Connected an AI agent in n8n to the live Shopify product catalog, so answers about specs and compatibility are always accurate, not scripted.

3
Automated warranty claim intake

Defect reports now auto-collect order number, photo, and issue description in one structured flow instead of a scattered back and forth.

4
Set a confidence threshold

The agent answers confidently on catalog facts, but escalates ambiguous or emotional messages straight to Funmi's team on Slack.

5
Unified all three channels

Shopify inbox, Instagram DMs, and email now feed into the same automation instead of three separate manual queues.

6
Reviewed and refined weekly

First month included a weekly review of escalated conversations to keep improving the agent's answers.

The result

76%
Pre-purchase questions auto-answered
3x
More conversations handled per day
1
Unified queue, down from three

"We were basically two people trying to be a call center. Now the agent handles the repetitive spec questions, and my team only sees the conversations that actually need a human."

Funmi E.
Founder, Voltage Gadgets
CASE 04 / 05
Blessing O., Pressed & Poured
Social Media (Instagram and WhatsApp) Juices and cold-pressed drinks Abuja, Nigeria

The situation

Pressed & Poured has no website. Blessing sells entirely through Instagram posts and WhatsApp orders, which is common for food and drink brands starting out. The problem was volume without structure: orders came in as DMs, voice notes, and comments, with no consistent way to capture an address, confirm a payment, or know what was actually in stock that day. She was manually copying every order into a notebook before her evening delivery run, and mistakes were common.

What we built

1
Mapped the real order journey

Tracked exactly how orders arrived across Instagram comments, DMs, and WhatsApp to find where the process broke down.

2
Built a WhatsApp ordering flow

Set up an automated WhatsApp menu so customers select products and quantities in a guided flow instead of free-form messages.

3
Connected Instagram DMs into the same system

Comments and DMs on Instagram now route straight into the same WhatsApp ordering flow, so nothing gets missed on a busy post.

4
Automated daily stock updates

Blessing updates one simple stock sheet each morning, and the flow automatically stops offering items that have sold out that day.

5
Structured every order automatically

Each completed order auto-populates a delivery sheet with address, items, and payment status, ready for the evening run.

6
Piloted on a single week of orders

Ran the system in parallel with her notebook for one week to confirm accuracy before fully switching over.

The result

0
Missed or lost orders since launch
40%
More orders processed per evening run
1
System, replacing the paper notebook

"I didn't think I needed a website to need automation. Turns out the chaos was never the website, it was DMs and voice notes with no system behind them. That part is fixed now."

Blessing O.
Founder, Pressed & Poured
CASE 05 / 05
Emeka O., Bramwell Home
BigCommerce Home and furniture Abuja, Nigeria

The situation

Bramwell Home runs on BigCommerce, selling furniture and home decor with higher order values and longer delivery windows than a typical store. That combination meant every order generated multiple check-in messages from anxious customers over the two to three week delivery window, on top of the usual support and cart recovery load. Emeka's ops team of three couldn't keep pace with both proactive updates and reactive questions at the same time.

What we built

1
Audited the full order lifecycle

Mapped every stage from order placed to delivered, including the manufacturing and freight steps unique to furniture.

2
Connected BigCommerce to n8n

Built a webhook layer so every order and fulfillment status change fires an automation instantly, across the full multi-week timeline.

3
Built proactive milestone updates

Customers now get automatic updates at order confirmed, in production, shipped, and out for delivery, before they have to ask.

4
Layered in the after-sales system

Added the same WISMO and returns automation used in other case studies, tuned for BigCommerce's order data structure.

5
Added cart recovery for big-ticket items

High-value abandoned carts trigger a personal, lower-pressure follow up rather than a generic discount blast.

6
Rolled out in phases

Launched proactive updates first, then support automation two weeks later, then cart recovery, so the ops team could adjust gradually.

The result

63%
Fewer proactive check-in messages
2.1x
Cart recovery on big-ticket items
3
Systems running as one connected pipeline

"With a three week delivery window, silence is what kills customer trust. Now the customer hears from us before they even think to ask, and my team's time goes to the orders that actually need a person."

Emeka O.
Operations Lead, Bramwell Home
Your store could be case six

Ready to see your own pipeline built?

Every case here started with the same first step: a short call to map what's currently manual. Let's do the same for your store.

Book a call
Scroll to Top