Case study · a system for the business
Four channels, four hundred orders a month, 44 percent repeat customers. One system
When SEO brings orders, you can't afford to lose them. For Krasnodarflora I designed and built the system that carries an order from checkout to a bouquet photo in the customer's hands. About four hundred orders a month flow through it, and 44 percent are placed by customers who have bought before.
- ≈400
- orders a month through the system, April · June 2026
- 4
- sales channels merged into one feed
- 44%
- of orders come from returning customers
- 0
- invented reviews or ratings
Why the store needed its own system
Orders arrived from the site, two marketplaces and email, and lived in four different places. Florists assembled from memory, couriers called to re-check addresses, the owner reconciled revenue by hand.
Off-the-shelf CRMs didn't fit this pipeline. I built the system around the store's real processes: from "order accepted" to "customer rated the bouquet", working from a phone for the whole team.
What the system had to solve
- 1
Every channel in one feed
The site, marketplaces and email orders merge into a single feed with statuses. Email import parses the messages itself, no manual re-typing.
- 2
A florist pipeline
An order reaches the florist with its composition and delivery time. A photo of the finished bouquet goes to the customer for approval before the courier leaves.
- 3
A courier loop
Couriers get orders in Telegram, mark statuses, and the system computes their pay per trip on its own, crediting combined runs.
- 4
Honest feedback
After delivery the customer rates the order. Scores from confirmed orders flow to the storefront. Nothing is painted by hand.
- 5
Analytics you can trust
Revenue, channels, repeat purchases, monthly P&L. Calendar periods, bot-free data, cash reconciled against the bank.
- 6
Phone-first
The system ships as a PWA: the owner sees the business from a phone, florists and couriers get their tasks as push notifications.
What the system consists of
Orders and sales
- One feed for all channels with statuses
- Order import from email and marketplaces
- Amount and payment-link synchronization
- Unpaid orders and refunds under control
Production and delivery
- A florist pipeline with bouquet photos
- A courier Telegram bot with statuses
- Automatic courier pay calculation
- Every bouquet photographed for the customer
Reviews and storefront
- Order rating after delivery
- Real ratings on product cards
- A moderated review queue
- A short-link bouquet page for the customer
Money and analytics
- Monthly P&L by sales channel
- Cash-day reconciliation
- Analytics built on bot-free data
- Roles: owner, manager, florist, courier
What changed for the owner and for SEO
The owner sees the whole business from a phone: orders, money, florists and couriers. The customer gets a bouquet photo before delivery, and that promise is kept on every order.
It works for SEO too: real ratings and order counters on the storefront build trust and behavioral signals that no fake-review scheme can buy.
Two years of the system at work: seasonal March peaks and a steady summer flow. Numbers come from the order database, cancellations excluded.
Data: the store's order database · July 2024 · June 2026
Show data as a table
| Orders | |
|---|---|
| 07.24 | 328 |
| 08.24 | 320 |
| 09.24 | 346 |
| 10.24 | 338 |
| 11.24 | 401 |
| 12.24 | 280 |
| 01.25 | 253 |
| 02.25 | 350 |
| 03.25 | 519 |
| 04.25 | 249 |
| 05.25 | 210 |
| 06.25 | 200 |
| 07.25 | 204 |
| 08.25 | 231 |
| 09.25 | 247 |
| 10.25 | 232 |
| 11.25 | 305 |
| 12.25 | 252 |
| 01.26 | 222 |
| 02.26 | 281 |
| 03.26 | 485 |
| 04.26 | 203 |
| 05.26 | 216 |
| 06.26 | 202 |
Honest status
This is an internal system built for one store's processes, not a boxed product for sale. There are no interface screenshots with customer data in this case: those screens hold real names and phone numbers. The storefront in the screenshot is public. The repeat share is computed by customer phone number across confirmed orders, a 24-month average; month to month it ranges from 32 to 52 percent.
More case studies
SEO Website
From $800A fast, clean website built with SEO structure from the start.
Case study · my own store
Ten years in search, a four-year plateau, half of Google gone. My store, my numbers
Read the case
Case study · client, in progress
1,600 searches a month pass this site by. I'm taking a signage maker into search before the season
Read the case
Free, 2-3 days
Traffic burns without lead tracking
Tell me how your process works. I'll point out what to automate first and where inquiries are getting lost.
- A prioritized list of problems: what to fix first, what can wait
- Demand in your niche: how many people search for it and where you rank now
- What your team can handle in-house and what is worth handing to me
I reply in writing within a couple of hours. No calls needed and no sales script.
Numbers you can check
- ≈400
- orders a month in the flower shop I run
- 44%
- of those orders come from returning customers
- 10 years
- that site in search, full history on the table
The same case shows a 51% Google drop in May 2026. Most people hide that. I show it and write what I am doing about it.