Manual QA for an online store's product dashboard and checkout

The client is a US retailer replatforming its online store ahead of a peak trading season. The product dashboard, catalogue, and checkout are where revenue is either captured or quietly lost, so the store had to be right the day it opened.

The internal team had strong developers but no dedicated QA function, and a hard launch date left no room to discover basic defects in front of real shoppers.

The catalogue, search, filters, and checkout had been built quickly against a fixed deadline, and nobody had tested the paths a real shopper actually walks end to end. Automated checks passed, yet the joined-up journey had never been exercised by hand.

Product data drove much of the storefront, so a single mismatched price, wrong image, or broken filter could undermine trust the moment the store opened. Layout also shifted across browsers and viewports in ways the team had not systematically checked.

With the launch date fixed, the retailer needed defects surfaced early enough to fix them calmly, not a scramble on launch weekend.

We wrote test charters covering the full purchase journey (browsing the catalogue, searching, filtering, adding to basket, and completing checkout as both a registered and a guest shopper) and ran them by hand against a maintained matrix of real browsers and devices in BrowserStack.

Alongside the scripted cases we ran session-based exploratory testing on the search, filter, and product-data flows, deliberately hunting the layout and data mismatches that pass automated checks but break the shopper's experience. Every defect went into Jira with exact steps, environment, and a screenshot, and cases were tracked in TestRail against the signed acceptance criteria.

We ran the cycle early enough that fixes could be verified and regressed before the freeze, closing with a clear coverage report and go/no-go recommendation rather than a last-minute status update.