The most common technical problems in online stores — and how to catch them early
Eight technical problems that quietly cost online stores orders — from updates and slow categories to broken integrations and backups — and how to check for them.
9 min read
Key takeaways
- The most expensive problems rarely show an error message — the customer simply leaves.
- Updates without a staging environment and a check of the critical path are a major source of outages.
- Integrations with couriers, ERP and invoicing need a log and alerts when they stop.
- A backup whose restore has never been tested is not a recovery plan.
- A short monthly checklist catches most problems before customers notice them.
Why small technical problems cost the most
When an online store goes down completely, someone notices within minutes. Silent faults are far more expensive: shipping calculation fails only for certain addresses, the confirmation email lands in spam, or the stock sync stops overnight. In most of these cases there is no error message — the customer just does not complete the order.
That is why technical maintenance is not only about reacting to outages, but about regularly checking the places where losses stay invisible. Below are the eight areas that most often cause such problems, with the first check for each.
| Problem | How it shows up | First check |
|---|---|---|
| Errors after updates | Broken pages, blank screens, missing features | Is there a staging environment and a list of recent changes? |
| Slow categories and products | High drop-off on mobile devices | Speed measured by page type |
| Cart problems | Orders started but not completed | A test order with different shipping and payment methods |
| Broken integrations | Wrong stock levels, missing shipping labels | Log of the last successful sync |
| Emails in spam | Customers ask whether their order went through | SPF, DKIM and DMARC records for the domain |
| Duplicate content | Indexed filter pages | Indexing report in Search Console |
| Untested backups | Invisible — until the day of the outage | When a backup was last restored |
| Weak access control | Shared accounts, former contractors with access | List of administrators and two-factor authentication |
1. Errors after platform, theme and extension updates
A WooCommerce store usually runs WordPress, WooCommerce, a theme and dozens of plugins, each with its own update cycle, plus the PHP version on the server. On Shopify the platform is updated centrally, but apps and the theme add their own code to the store. In both cases a seemingly harmless update can change how the cart or a template behaves.
- Update in a staging environment — a copy of the live store — first.
- Update one component at a time so you know what caused a problem.
- After every update, walk through the critical path: product, cart, payment, confirmation email.
- Keep a short change log and a plan for rolling back to the previous version.
2. Slow category and product pages
Speed is rarely the same across the whole store. The homepage may be fast while categories with many products and filters, or product pages with galleries and reviews, load slowly. Measuring only the homepage gives a false picture.
The most common causes are unoptimised images, scripts from apps that are no longer used, slow database queries when filtering, and hosting without caching.
- Measure the homepage, a category, a product and the cart separately.
- Check Core Web Vitals in Search Console — that is where the real-visitor data is.
- Review the installed apps and plugins and remove the ones you no longer use.
3. Cart and checkout problems
The cart is where technical errors cost money most directly. Typical examples are incorrect shipping calculation for certain locations, coupons that conflict with sale prices, and payments where the customer is charged but the order stays "pending" because the notification from the payment provider was not processed.
- Place a test order at least once a month and after every update — with different shipping and payment methods.
- Track the number of started but uncompleted payments and investigate sudden changes.
- Check that notifications from the payment provider (webhooks) arrive and are processed.
4. Broken integrations with couriers, warehouse and invoicing
Integrations usually run "in the background", so when they stop it is noticed late — when a customer orders an out-of-stock product or a shipping label is never created. Typical causes are a changed API key, an expired subscription, a change in the provider's API or a reached rate limit.
A good integration records every sync and alerts a responsible person when a step fails, instead of stopping silently.
- Know when the last successful sync happened for every integration.
- Set up an alert for a failed sync, not only for a complete outage.
- Keep access keys outside the code and document who manages them.
See also:API integrations and monitoring
5. Order emails that never reach customers
If customers often ask whether their order was received, check email deliverability. Sending directly from the web server without a properly configured domain often lands messages in spam.
- Check the domain's SPF, DKIM and DMARC records.
- Use a dedicated transactional email service instead of sending from the server.
- Send a test order to addresses at different email providers.
6. Duplicate and unnecessarily indexed content
Filters and sorting in categories can create thousands of URLs with almost identical content. If they are indexed, the search engine spends its resources on them instead of on important categories and products. Out-of-stock products without a clear policy — whether the page stays, redirects or is removed — create a similar problem.
7. Backups that have never been restored
Almost every host offers backups, but few stores have checked whether they can actually restore the store from them — with the database, uploaded files and settings. A backup stored on the same server does not help if the problem is the server itself.
- Keep at least one copy outside your main hosting.
- Test a restore in a staging environment at least once a quarter.
- Record how long the restore takes — that is your real readiness for an outage.
8. Weak access control and forgotten accounts
Over time, accounts of former employees, agencies and temporary contractors pile up in the store admin. Shared passwords and the lack of two-factor authentication turn every leaked credential into a risk for customer data.
- Every person should have their own account with the minimum permissions required.
- Enable two-factor authentication for all administrators.
- Review the access list every quarter and remove what is no longer needed.
Turning the list into a monthly routine
Most of the checks above take minutes when done regularly. If you have no in-house technical team, the most practical option is a monthly maintenance plan with clearly defined checks, or a one-off technical audit that shows where to start.
- A test order with different shipping and payment methods.
- A review of errors and warnings in Search Console.
- A check of the last successful sync for each integration.
- Planned updates in a staging environment.
- A review of the list of administrator accounts.
See also:Online store maintenanceTechnical audit of an online store