The store is one of the cleanest migration candidates I've assessed — but a rebuild is not the cheapest fix for the fee problem, and the fee problem may not be where the money is going.
Technically: green light. 151 products, 159 variants, zero third-party apps, and inventory tracking switched off entirely. The three things that normally sink a Shopify migration — app dependencies, variant/stock complexity, and customer state — are all absent here.
Commercially: get one number first. The whole case rests on what share of revenue actually passes through the Shopify checkout. If most orders arrive by Instagram and WhatsApp as you describe, they're paying a subscription and a 2% penalty to support a checkout that handles the minority of their sales — and the right build isn't a checkout at all.
Finding — what's actually there
Store 7a8393-2.myshopify.com, running Shopify's free Horizon theme (v2.1.2), trading in ZAR. Everything below is pulled from the live storefront.
Cost — where the fees come from
That single fact is the source of the pain. Because they must use a local gateway, Shopify levies a third-party transaction fee on top of whatever the gateway charges. Two fees on every card sale.
| Layer | Basic | Grow | Advanced |
|---|---|---|---|
| Shopify third-party gateway fee | 2.00% | 1.00% | 0.60% |
| Gateway — Yoco standard local card | 2.95% | 2.95% | 2.95% |
| Effective rate | 4.95% | 3.95% | 3.55% |
Plus the subscription itself — roughly R550–R730/month on Basic depending on billing term and the USD/ZAR rate, since Shopify bills in dollars and the ZAR figure moves with the exchange rate.
Gateway alternatives sit in a similar band: PayFast at 3.5% + R2 (dropping to 2.0% on Instant EFT and Capitec Pay), Peach at 2.95% + R1.50 but with a R300/month account fee. Yoco's rate drops to about 1.35% above R50,000/month in volume, which makes it the strongest option if volume supports it.
The number that decides everything
Every rand of that 2% penalty is charged only on orders that pass through the Shopify checkout. If the bulk of orders arrive by DM and are settled by EFT or a payment link, the penalty is already being avoided on most revenue — and the saving from a rebuild is much smaller than the headline rate suggests. The subscription, however, is paid regardless of channel. On a low-checkout-volume store, the monthly fee is the real cost, not the percentage.
Cheap lever — worth doing this week either way
Moving Basic → Grow halves the third-party fee, 2% to 1%. The plan step-up costs on the order of R1,000/month more, so it pays for itself once roughly R100,000/month is going through the Shopify checkout. Below that, stay on Basic and switch to annual billing. This is a five-minute change that needs no code — check it against their real figures before considering anything larger.
Risk — what a rebuild has to replace
The storefront is the easy part. These are the things that currently come free and would become yours to build, buy, or consciously drop.
| Capability | Effort to replace | Recommendation |
|---|---|---|
| Checkout & PCI surface | High if rebuilt | Avoid — use hosted payment links |
| Gift cards (they sell them) | High — codes, balances, partial redemption | Handle manually or drop |
| Discount codes | Medium | Confirm if actually used |
| Order & shipping notification emails | Low | Transactional email service |
| Customer accounts & order history | Medium | Drop — low value at this size |
| VAT handling & compliant invoices | Medium | Must not be skipped |
| 301 redirects for ~210 URLs | Low but unskippable | Build the map before launch |
| Inventory tracking | None — already unused | No work required |
| App integrations | None — none installed | No work required |
The product tagging shows deliberate SEO investment, particularly on footwear — 19 sneaker products each carry the same dozen search-intent tags (handmade sneakers South Africa, sustainable sneakers, upcycled footwear). Whatever that ranking is currently worth, it is the one asset a careless migration can destroy outright. Around 210 URLs across products, collections and posts need a one-to-one redirect map, built and tested before cutover.
Recommendation
This is the substantive point. If orders genuinely arrive through Instagram and WhatsApp, then a checkout is not what the business runs on — it's an expensive appendix. What they actually need is:
That is a content site with a form and hosted payment links. It is a fundamentally smaller and cheaper thing than an ecommerce platform migration, it matches how they already sell, and it eliminates both the subscription and the 2% penalty.
Pull the real numbers, right-size the plan, and clean the catalogue. All reversible, all cheap.
Static catalogue and content pages, an order-request form routing to WhatsApp and email, and hosted payment links for the handful of card sales. Products import cleanly — Shopify's own product feed is already a well-formed export, and all 240 images are intact.
Rough sizing on the build: content migration is close to trivial, and there's no bespoke design to reproduce. The work is in the catalogue templates, the order flow, and the redirect discipline — call it two to three weeks, with the payment-link integration being the smallest part rather than the largest.
Before committing