The network for creativity
Join 1.25M professional creatives like you
Connect with clients, get discovered, and run your business 100% commission-free
Creatives on Contra have earned over $150M and we are just getting started
Back to feedPost
Paystack tip that saves a support headache: never trust the frontend callback to confirm payment.
The redirect can fire before funds settle, or the user closes the tab.
Verify server side by calling the transaction verify endpoint with the reference, and mark the order paid only when status is success and the amount matches your expected value in kobo.
I log every webhook and reconcile daily.
On one checkout this took disputed orders to zero across 400+ transactions.
Building a checkout you can trust? Message me.
The network for creativity
Join 1.25M professional creatives like you
Connect with clients, get discovered, and run your business 100% commission-free
Creatives on Contra have earned over $150M and we are just getting started
Related posts
A question got answered correctly at 2am, no one awake to do it.
A customer needed a specific setup answer outside business hours. The bot pulled it straight from the docs and answered it correctly, no ticket, no wait until morning.
→ A good answer shouldn't depend on someone being awake to give it.
The bot pulling the exact doc snippet and replying instantly cuts out the ticket queue and keeps customers moving forward.
A reminder went out after the contract was already signed.
A sync delay meant the reminder fired an hour late. Harmless, but avoidable now signing status checks in real time before any reminder goes out.
→ Automated follow-up still needs a live check before it sends.
Nice catch on the sync lag, adding a live status check before the reminder will stop those hour old pings.
At Procys, I design the exception path alongside document extraction.
Invoices and identity documents arrive in different layouts. OCR and LLM reasoning produce candidate fields. Deterministic checks validate headers, line items, and tax values. When confidence or record comparisons leave a field uncertain, the workflow sends it to a reviewer with the source context.
A good document AI system should make uncertainty visible before data moves into business systems.
If your team processes invoices or identity documents, which exceptions still take the most manual time?
edge cases are where doc AI earns trust, love that you design for them 👌
