Written by Kiwi Dynamics Team 1Letting support questions go unanswered overnight
A customer who messages at 9pm asking about their order and doesn't hear back until the next afternoon has had eighteen hours to get frustrated, or to complain publicly, or to chargeback. Support that only runs during business hours is a mistake that costs trust every single week, especially for a business selling to a timezone-spread customer base. It's one of the easiest mistakes to fix and one of the most commonly left unaddressed.
2Guessing at reorder quantities
Ordering stock based on a gut feeling or last year's numbers means either tying up cash in slow-moving inventory or running out of the products that are actually selling. Both sides of that mistake cost real money, one in storage and markdowns, the other in lost sales at the worst possible moment. It's a mistake that compounds every reorder cycle it isn't fixed.
3Publishing thin, rushed product listings
A product listing thrown up quickly, with a thin description and no attention to sizing or material detail, generates more return requests and more support questions than one written properly. The time saved writing it fast gets spent twice over answering questions the listing should have already answered. This mistake is easy to miss because the cost shows up somewhere else, in the support queue, not on the listing itself.
4Sending one generic cart recovery email
Sending the same 10 percent off code to every customer who abandons a cart, regardless of what they were looking at, leaves real revenue behind. A generic message recovers some carts, but a personalised one recovers meaningfully more, and the gap between those two numbers is pure margin left on the table. This mistake persists mostly because the generic version technically works, just not nearly as well as it could.
5Recommending the same products to everyone
Showing every visitor the same bestsellers list instead of recommendations based on what they've actually looked at is a missed opportunity on every single product page. Retailers who've never personalised recommendations are usually surprised by how much average order value moves once they do. Not testing this mistake is itself part of the mistake.
6Ignoring what reviews are actually saying
Hundreds of reviews sitting on a product page contain real, specific feedback, about sizing, durability, and quality, that almost nobody on the team ever actually reads in full. That means a pattern like a strap breaking after six months goes unnoticed until it shows up as a spike in returns instead. Ignoring reviews as a data source is a mistake that only gets more expensive as review volume grows.
7Reordering stock reactively, not proactively
Waiting until a shelf is empty to reorder means either a rush order at a worse price or a stockout on a product that was actively selling. Reactive reordering is one of the most direct hits to margin on this list, because both the rush freight cost and the lost sale are avoidable with a bit of lead time. It's a mistake driven entirely by not watching the right numbers closely enough.
8Handling every return with the same slow process
Treating every return the same way, with a reason code email, a confirmation, and a wait for approval, punishes the customer with the simplest, most obviously valid return just as much as the genuinely complicated one. That slow, uniform process is a mistake that shows up directly in customer satisfaction scores and repeat purchase rates. Most of those returns could be resolved in one message instead of four.
9Letting urgent tickets sit in a shared inbox
A support ticket that actually needs an urgent response, a damaged item, a payment issue, sitting in a shared inbox next to a dozen routine questions is a mistake that costs a business its most valuable customers first. Without triage, urgency gets decided by whoever happens to open the inbox next, not by what the ticket actually says. That's a mistake with real churn attached to it.
10Deciding on numbers that are a week old
Making stock, marketing, and staffing decisions based on a report that's a week old because compiling it properly takes half a day means every decision is made a beat behind reality. This mistake is subtle because the numbers aren't wrong, they're just late, and late numbers lead to slow reactions to trends that have already moved on. Fresh, reliable data changes the speed a business can actually operate at.
None of these mistakes are due to a lack of effort, they're what happens when manual process is stretched past what it can reasonably cover. Kiwi Dynamics builds the production AI that closes these gaps for NZ and AU retailers, shipped into the systems already in use and measured in hours and dollars, not slideware that looks good in a pitch and never actually runs.