Ecommerce & Retail · Growth problem
Growth stalled after a replatform
A revenue dip after a replatform is common enough that it's tempting to assume the migration caused it, and specific enough in its usual causes that it's worth checking properly rather than assuming. SEO migration errors, tracking discontinuity, and genuine UX regressions are all real possibilities, but so is coincidental seasonality or a promotional calendar change that happened around the same time — the two need to be told apart carefully.
Symptoms
What this usually looks like
- Organic traffic or rankings dropped noticeably around the migration date
- Conversion rate changed after go-live, in either direction
- Analytics and revenue reporting show a discontinuity that makes before/after comparison difficult
- Checkout or payment issues emerged post-launch that weren't present before
- Product feed or marketplace listing issues appeared after the migration
Diagnostic questions
What we would test first
- Audit redirect coverage and accuracy from old URLs to new URLs, sampling both high- and low-traffic pages
- Compare organic rankings for a fixed keyword set, pre vs post migration
- Check analytics/tagging implementation for consistency across the pre- and post-migration periods
- Run a step-by-step checkout funnel comparison, pre vs post
- Review product feed error logs and marketplace listing counts around the migration date
- Identify any promotional, seasonal or media spend changes that coincided with the migration to isolate confounders
Root causes
Why it happens
- 01
SEO migration errors
Common technical issues include broken or missing redirects from old URLs, changes to URL structure without proper mapping, altered or lost metadata, and changes to internal linking structure — any of which can cause a genuine and measurable drop in organic visibility and traffic.
- 02
Tracking and analytics discontinuity
A platform migration frequently involves a change in analytics implementation, tagging, or even platform (e.g. moving to GA4 alongside the replatform), which can make before/after comparison unreliable unless both periods are checked against a consistent measurement methodology.
- 03
Genuine UX or conversion regression
New checkout flows, changed page layouts, altered payment options, or slower page speed on the new platform can each independently affect conversion rate — separate from any SEO impact — and need to be tested individually rather than assumed as a single combined effect.
- 04
Product feed and marketplace listing disruption
Product data feeds to Google Shopping, marketplaces or comparison sites can break or degrade during a migration if data mapping isn't handled correctly, which suppresses visibility and sales in those channels independently of the main website.
- 05
Confounding factors coinciding with the migration date
Seasonality, a change in promotional calendar, a paid media budget change, or a competitor action occurring around the same time as the migration can all contribute to a revenue change that gets wrongly attributed entirely to the platform move.
- 06
Governance gaps during the migration itself
Where SEO, analytics, and commercial stakeholders aren't formally involved in migration planning and testing, issues that would have been caught pre-launch (broken redirects, missing tracking) can go live and only get discovered once the commercial impact is already visible.
Evidence
The numbers we would look at
These are the metrics that make the constraint visible, and the cuts that stop them being reassuring by accident.
| Metric | What it tells you | How we read it |
|---|---|---|
| Organic traffic and rankings, pre vs post migration | The clearest signal of an SEO migration issue specifically.Compare matched date ranges and control for seasonality; check redirect coverage directly, not just traffic trend. | Compare matched date ranges and control for seasonality; check redirect coverage directly, not just traffic trend. |
| Conversion rate, pre vs post, matched by traffic source and device | Isolates a genuine UX/conversion regression from a traffic-mix change.Ensure tracking methodology is consistent across both periods before comparing. | Ensure tracking methodology is consistent across both periods before comparing. |
| Checkout completion rate, pre vs post | Identifies specific checkout or payment issues introduced by the new platform.Step-by-step funnel comparison, not just overall completion rate. | Step-by-step funnel comparison, not just overall completion rate. |
| Product feed health / marketplace visibility, pre vs post | Captures disruption to channels outside the core website that a site-only analysis would miss.Feed error rates and listing count in Google Shopping/marketplace accounts. | Feed error rates and listing count in Google Shopping/marketplace accounts. |
| Revenue, matched period comparison with confounders controlled | The commercial bottom line, but only meaningful once seasonality and promotional activity are accounted for.Use a like-for-like period from the prior year as well as immediate pre-migration, where possible. | Use a like-for-like period from the prior year as well as immediate pre-migration, where possible. |
Measurement traps
What can mislead you
- Looks fineRevenue dropped right after the migration, so the migration caused it
- Timing alone doesn't establish causation — checking whether seasonality, promotional calendar changes, or media spend changes occurred around the same date is essential before attributing the drop entirely to the platform move.
- Looks fineOur analytics shows a big change, so something is clearly broken
- A change in analytics implementation or tagging during the migration can itself produce a reporting discontinuity that looks like a real business change but is actually a measurement artefact.
- Looks fineOrganic traffic recovered within a few weeks, so the SEO migration went fine
- Traffic volume recovering doesn't confirm rankings or revenue recovered on the same pages or for the same keywords — a page-level or keyword-level check is needed to confirm a genuine like-for-like recovery.
Outcome
What better looks like
Not a promised number. A clearer basis for the next investment decision.
- The business can distinguish, with evidence, which part of any post-migration change is attributable to the platform move versus other factors
- SEO, analytics and checkout issues arising from the migration are identified and prioritised individually, not treated as one undifferentiated 'the migration broke things' problem
- Future migrations are planned with SEO, analytics and commercial stakeholders involved from the outset, with defined pre-launch checks
- Pre/post reporting uses a consistent measurement methodology so future comparisons are reliable
Where a Growth Diagnostic would start
A three to four week senior review across demand, discovery, acquisition, conversion, retention, measurement and capability — sequenced so this problem is either confirmed as the constraint or ruled out early. Read alongside the ecommerce & retail model page for how we frame the wider system.
Related problems
Ecommerce & Retail
Traffic isn't converting
Visitors are arriving but not buying. A structured look at where conversion is breaking down, and why.
Ecommerce & Retail
SEO traffic isn't driving revenue
Organic visibility looks strong but it isn't translating into sales. A look at intent, landing pages and product availability.
Ecommerce & Retail
Marketing attribution is unclear
Every platform claims the credit and the numbers don't add up. A pragmatic approach to measurement that decisions can actually be based on.
Questions about this problem
- How soon after a migration should we investigate a revenue drop?
- If a meaningful drop persists beyond the first 2–4 weeks after go-live — long enough to rule out short-term settling effects — it's worth a structured look, since redirect and SEO issues in particular tend not to self-resolve without intervention.
- We didn't set up any special tracking before the migration — is it too late to diagnose this?
- It's more difficult without a clean pre-migration baseline, but standard historical GA4, Search Console and platform data usually still allow a reasonably reliable before/after comparison, provided the migration date and any confounding events are clearly identified.
- Is this useful before a planned migration, not just after one?
- Very much so — reviewing the migration plan against known SEO and analytics risks beforehand, and agreeing pre-launch checks, is generally far cheaper than diagnosing and fixing issues after they've already affected revenue.