Skip to content

Assess a shop relaunch by version and customer journey

Evaluate visible purchase journeys and internal processing together after a relaunch. Version links and comparable observations inform the next change.

Lena Fischer2 min read
Grey server with an open service panel, removable module and orange handles.

I would assess a shop relaunch by whether important customer transactions work reliably with the new version. The new design can be examined during acceptance. Whether orders continue to complete under later operating conditions needs additional observation.

Connect the visible journey with processing

Google SRE distinguishes visible behaviour from internal measurement signals. These perspectives complement each other in a shop: a customer may complete the purchase journey while a subsequent transfer is still pending. A fast web server response does not automatically answer whether the order was processed.

A fictional retailer introduces shop version S-neu. S-alt identifies the previous version. The following findings are invented evaluation examples, not measured project results.

Observation

What it establishes in the example

What still needs clarification

The reviewed product page in S-neu matches the design.

This presentation was checked.

Other views and the complete ordering journey

The purchase journey in S-neu visibly succeeds, but transfer remains pending.

Interface and subsequent processing give different states.

Associated order, cause and duration of the pending transfer

A report for S-neu includes different devices and visitor sources from the report for S-alt.

The compared groups differ.

Comparable transactions before judging the version's effect

The pending transfer is a specific software finding here. Another design detail would not resolve it. The team first needs a traceable link between the visible completion and the processed order.

Associate the change with the correct version

Record which version, configuration and integration processed an investigated transaction. Further changes may follow a relaunch. Without this link, a later error would be attributed indiscriminately to the initial release.

Keep the same definitions and time boundaries for before-and-after comparisons. If the offer, device mix or visitor sources change, those differences belong in the assessment. A better aggregate alone does not prove an effect of the new interface.

Match effort to actual operation

A small shop does not automatically need a comprehensive new monitoring platform. A few clearly chosen transactions and a maintained overview of pending transfers can be a starting point. If time-critical errors need rapid detection, that requirement calls for suitable automatic detection and response.

Sources

Written by

  • Lena FischerEditor for E-CommerceLena Fischer looks after the e-commerce articles at ASES. She writes about shop systems and how they connect to other applications, and explains the technical processes behind online retail.

Follow ASES

Read more

All articles

E-Commerce with ASES

Online shops, marketplaces and B2B portals