Migrating from one e-commerce platform to another: what can go wrong with 3,000 products?

The agency is right about passwords; hashing mechanisms don't translate between different systems. When we migrated, we sent customers a magic login link that prompted them to update their password. For order history, watch out for ID collisions—if your old order numbers clash with the new platform's ID sequence, your invoices will get completely mixed up.

We went through a similar migration two years ago with 4,000 products. A 15 percent dip in organic traffic during the first month is completely normal, don't panic. Even though we handled all our redirects properly, it took Google six weeks to crawl the new architecture. 14,500 Euro is actually quite reasonable in the German market for this scope and timeline.

If the agency says 8 weeks, plan for at least 12. Attribute mapping for variant products (colors, sizes, materials) almost always requires manual intervention. The agency will probably just run a basic CSV import, call it a day, and leave you stuck fixing missing product descriptions yourself.

Make sure to get these clauses added to the agency contract: 1) A 301 redirect mapping table covering all legacy URLs, 2) A complete, lossless transfer of meta titles and descriptions, 3) Retention of image alt tags, 4) If the site is multilingual, validation of hreflang and language tags against the new setup.

we pulled this crazy stunt last year too. product images were the biggest headache by far. old file names had spaces or special characters and the new platform threw a fit. honestly spin up a staging environment and run test orders with at least 100 products definitely do not flip the main domain before thoroughly testing checkout end-to-end.

Correction: I misremembered the figure, it was a bit lower.

What are you doing about your existing accounting or shipping integrations? Moving products is the easy part; are the middleware pipelines that tie the order flow to invoicing and shipping labels being rebuilt? Is setting up those integrations included in the agency's quote?

Don't push to production on a Friday—do it Tuesday night. If you go live mid-week during normal business hours for your team and the agency, you can jump straight on any checkout or cart issues that pop up in the first 24 hours. Launch over a weekend and there'll be nobody around to fix anything.

3000 products isn't huge but if you have tons of variants it's a nightmare. we had colors and sizes created as separate items and the entire category tree got wrecked. walk the agency through your variant setup right at the start and make them sign off on it.

Let me summarize what's been said so far. The answer varies greatly by industry; there is no one-size-fits-all rule.

This is my opinion, I'm not claiming it's absolute truth.

Noted, thanks.

There's a trap here, let me mention it. Start with a small trial; don't commit to everything at once.

Good luck with that.

you're right but weekly written progress reports are much more usseful than asking for dates.

i'm also curious if anyone does it differently.

You're right.

There's a trap here, let me mention it. The system lives on after delivery; maintenance is a separate line item.

Hasty decisions become decisions you have to fix six months later. Hope this helps.

How did you solve this? Access credentials should be opened in the company's name, not personal accounts.

Just leaving this note, it might be useful.

I'd appreciate it if you shared the outcome.

I completely agree. Just because everyone does it doesn't mean it's right.

Just leaving this note, it might be useful.

I don't think this advice fits everyone. If scope grows, either time or budget must grow. There is no third option.

In legacy systems, the most expensive thing is unknown dependencies. Just leaving this note, it might be useful.