Every year, thousands of businesses move their websites to new platforms, new domains, or new designs. Most of them lose traffic. Some never get it back.
A study of 892 website migrations found that the average site takes 523 days to recover its pre-migration traffic. That's a year and a half. And 17% of those sites? They never recovered at all — even after 1,000 days.
The frustrating part is that nearly every migration disaster follows the same pattern: a rushed timeline, skipped steps, and an SEO strategy that was "someone else's problem." The businesses that avoid this pattern — the ones that treat migration as a strategic project, not a technical task — typically recover within weeks and often end up with more traffic than they started with.
This guide is your complete migration checklist, from first decision to 90-day review.
What Counts as a Website Migration?
The term "migration" covers more ground than most people realize. Not all migrations carry the same risk, and understanding which type you're dealing with determines how much planning you need.
| Migration Type | What Changes | Risk Level | Typical Recovery Time |
|---|---|---|---|
| HTTP to HTTPS | Security protocol only | Low | 1–2 weeks |
| Server/hosting change | Where files live, nothing user-facing | Low | 2–5 days |
| Design refresh (same URLs) | Visual appearance only | Low–Medium | 1–2 weeks |
| URL restructuring | Page addresses change | Medium–High | 4–8 weeks |
| CMS/platform change | Underlying technology changes | High | 4–12 weeks |
| Domain change | The entire web address changes | Very High | 2–6 months |
| Full redesign + replatform | Everything changes at once | Critical | 3–12 months |
The golden rule: the more things you change simultaneously, the higher the risk. A design refresh on the same platform with the same URLs is low-risk. A domain change combined with a CMS migration and URL restructuring is the digital equivalent of juggling chainsaws.
WooCommerce learned this the hard way. In November 2023, they migrated to a shorter domain — Woo.com. The result: a **90%+ drop in organic visibility**. Five months later, they reversed the migration entirely and went back to WooCommerce.com, which immediately restored their traffic. Even a well-known tech platform with significant resources couldn't brute-force a domain migration that wasn't properly executed.
When Should You Migrate?
Migration is never something to do casually. It's an investment of time, money, and risk. But there are clear signals that it's time:
- Your site is dangerously slow. Page speed directly affects search rankings (Google calls these Core Web Vitals) and conversion rates. If your current platform can't deliver fast load times, migration may be the only fix.
- Your platform limits what you can build. You need features your CMS can't support — personalization, multi-language, custom integrations, a mobile app — and workarounds are piling up.
- Security vulnerabilities keep appearing. Outdated platforms become targets. If your team spends more time patching security holes than building features, that's a sign.
- You're rebranding. A new company name or brand identity sometimes requires a new domain.
- You're merging or acquiring. Consolidating multiple websites into one is a common post-acquisition task.
- Developer talent is drying up. If your platform is so old that finding developers who know it is getting harder every year, migration becomes a matter of survival.
If none of these apply, don't migrate. A working website that ranks well is an asset. Don't fix what isn't broken.
Phase 1: Pre-Migration (4–8 Weeks Before Launch)
This is where migrations are won or lost. Every hour spent planning saves days of emergency fixes after launch.
Step 1: Audit Your Current Site
Before touching anything, take a complete snapshot of what you have. This becomes your baseline — the standard against which you'll measure whether the migration succeeded or failed.
Traffic and performance baseline:
| What to Document | Where to Get It | Why It Matters |
|---|---|---|
| Organic traffic by page (12+ months) | Google Analytics | Identifies seasonal patterns and top pages |
| Keyword rankings | Google Search Console | Baseline for measuring recovery |
| Backlink profile | Ahrefs, SEMrush, or Moz | Pages with external links need priority protection |
| Core Web Vitals scores | Google PageSpeed Insights | New site must match or beat these |
| Conversion rates by landing page | Google Analytics | Traffic without conversions is vanity metrics |
| Crawl errors and indexing status | Google Search Console | Fix existing problems before adding new ones |
Content inventory:
Crawl your entire site and create a spreadsheet of every URL. Not just blog posts and product pages — include PDFs, images, and any file that might appear in search results. For each URL, note:
- Current traffic volume
- Number of external backlinks
- Whether the content should migrate, be redirected, or be retired
- The designated owner (who reviews it and confirms correct migration)
This audit typically takes 1–3 weeks depending on site size. For a 50-page site, it's straightforward. For a 5,000-page site, it's a serious project — but skipping it is how migrations go wrong.
Step 2: Map Every URL Change
If any URLs are changing, create a redirect map: a spreadsheet with two columns — old URL and new URL. Every single page that's changing addresses needs an entry.
Three rules for redirect mapping:
- Map to equivalent content. Each old URL should redirect to the page on the new site that's closest in topic and intent. Redirecting your blog post about pricing strategies to your homepage doesn't transfer authority — Google treats that as a "soft 404" and ignores it.
- No redirect chains. If URL A redirects to URL B, which redirects to URL C — that's a chain. Google won't follow chains longer than five hops, and each hop degrades performance. Always redirect directly to the final destination.
- Use 301 redirects, not 302. A 301 tells search engines "this page has permanently moved here — transfer all authority." A 302 says "this is temporary — keep the old page in your index." Using 302s during a permanent migration is one of the most common mistakes and one of the most damaging.
What about pages you're deleting? If old content won't exist on the new site and there's no relevant page to redirect to, let it return a proper 404 error. This is better than redirecting it to your homepage. Google's own documentation confirms: having pages return 404 does not hurt your other pages' rankings. Mass-redirecting everything to your homepage, on the other hand, actively confuses search engines.
Step 3: Assemble Your Team
Website migration is not an IT project. It's a cross-functional business initiative.
Who needs to be involved:
| Role | Responsibility |
|---|---|
| Project owner | Overall timeline, decisions, stakeholder communication |
| SEO specialist | Redirect mapping, content preservation, monitoring |
| Developer(s) | Technical implementation, staging environment, redirects |
| Content owner(s) | Review migrated content, verify accuracy |
| Designer | Visual consistency, user experience |
| QA tester | Pre-launch validation, broken link checking |
The most expensive migration mistake isn't technical — it's organizational. When SEO is brought in three weeks before launch instead of three months before, the team is scrambling to fix problems that should have been prevented.
Step 4: Set Up Your Staging Environment
Build the new site on a staging server — a private copy that isn't visible to search engines or the public. This is where you test everything before it goes live.
Critical staging checklist:
- [ ] Staging site is blocked from search engines (robots.txt disallow or password protection)
- [ ] All redirects are testable on staging
- [ ] Content has been migrated and reviewed for accuracy
- [ ] Meta titles and descriptions are preserved (or improved) for key pages
- [ ] Schema/structured data is implemented correctly
- [ ] Internal links point to new URLs (not old ones that will redirect)
- [ ] Forms, checkout flows, and other interactive elements work
- [ ] Analytics tracking code is installed
- [ ] Page speed meets or exceeds the old site's performance
- [ ] Mobile experience is tested on actual devices
A cautionary detail: make absolutely sure your staging environment's robots.txt or canonical tags don't leak into production. If your staging site uses `canonical` tags pointing to `staging.yoursite.com`, and those tags go live, search engines may consolidate your ranking signals to URLs that don't exist publicly. This is more common than you'd think.
Phase 2: Migration Day
By the time launch day arrives, there should be no surprises. Everything has been tested, reviewed, and approved on staging.
Launch Day Checklist
Before flipping the switch:
- [ ] Lower your DNS TTL (time-to-live) to 300 seconds, at least 24 hours before migration. This ensures the internet picks up your new server address faster.
- [ ] Confirm all 301 redirects are ready to activate
- [ ] Confirm the robots.txt on the new site allows search engines to crawl it
- [ ] Confirm all `noindex` tags from the staging phase have been removed
- [ ] Have a developer available for the next 24–48 hours
Immediately after going live:
- [ ] Test redirects on your highest-traffic pages — manually check that old URLs forward correctly
- [ ] Verify the homepage, key landing pages, and conversion flows work
- [ ] Submit your updated XML sitemap to Google Search Console
- [ ] If changing domains: use Google's Change of Address tool in Search Console (this tells Google to transfer signals from old domain to new domain for 180 days)
- [ ] Verify your new site in Google Search Console (all variants: www/non-www, HTTP/HTTPS)
- [ ] Confirm Google Analytics is receiving data from the new site
- [ ] Revert your DNS TTL back to normal values
If changing domains: continue paying for your old domain. If someone else buys it, they could redirect it to harmful content associated with your brand.
Phase 3: Post-Migration Monitoring
Migration doesn't end at launch. The next 90 days determine whether you've succeeded or whether problems are silently compounding.
Week 1–2: Daily Monitoring
This is the highest-risk window. Check these daily:
| What to Monitor | Where | What to Look For |
|---|---|---|
| Crawl errors | Google Search Console | Spikes in 404s, server errors, redirect errors |
| Indexing status | Google Search Console | Are new pages being indexed? Are old pages being deindexed? |
| Organic traffic | Google Analytics | Drops exceeding 20% from baseline need investigation |
| Core Web Vitals | Google Search Console | Any metric regressions compared to pre-migration |
| Redirect functionality | Spot-check old URLs | Confirm redirects still work, no chains or loops |
A 5–15% dip in the first two weeks is normal — search engines need time to recrawl and reprocess your site. Anything beyond 20% that persists past week two signals a problem.
Week 3–6: Twice-Weekly Monitoring
Shift from daily to twice-weekly checks. Focus on:
- Keyword rankings for your top 50–100 terms. Rankings may fluctuate for 4–8 weeks before stabilizing.
- Page-by-page traffic comparison. Which specific pages have recovered? Which haven't?
- Conversion rates. Traffic recovery means nothing if conversions dropped. A migration that maintains visits but breaks the checkout flow is still a failure.
- Crawl errors. These should be trending toward zero. If they're increasing, something is still broken.
Week 7–12: Weekly Monitoring
By now, a well-executed migration should show traffic approaching or exceeding pre-migration levels. Focus shifts to optimization:
- Identify underperforming pages that haven't recovered and investigate why
- Update any remaining internal links that still point to old URLs through redirects
- Monitor backlink health — are external sites linking to your new URLs or still hitting redirects?
- Run a full site crawl (Screaming Frog, Sitebulb, or similar) to catch any lingering issues
Recovery Timeline by Migration Type
| Migration Type | Expected Dip | Recovery (Well-Executed) | Recovery (Poorly Executed) |
|---|---|---|---|
| HTTP to HTTPS | 5–10% | 1–2 weeks | 2–4 weeks |
| Design refresh (same URLs) | 5–15% | 1–2 weeks | 4–8 weeks |
| CMS/platform change | 15–30% | 4–8 weeks | 4–12 months |
| Domain change | 30–50% | 2–4 months | 6–18 months |
| Full redesign + replatform | 20–60% | 2–4 months | 6–18+ months |
The 10 Most Common Migration Mistakes
These are the errors that cause the majority of migration traffic losses. Every one of them is preventable.
| # | Mistake | What Happens | How to Prevent It |
|---|---|---|---|
| 1 | No redirect map | Old URLs return 404 errors, pages drop from Google's index | Map every URL before migration begins |
| 2 | Redirecting everything to the homepage | Google treats mass homepage redirects as soft 404s — authority is lost, not transferred | Redirect each page to its closest equivalent |
| 3 | Using 302 instead of 301 redirects | Search engines keep the old URL in their index and don't transfer authority | Always use 301 for permanent migrations |
| 4 | Redirect chains (A → B → C) | Each hop degrades performance; Google stops following after 5 hops | Redirect directly to the final destination |
| 5 | Forgetting robots.txt | Search engines can't crawl the new site — your entire site becomes invisible | Remove staging-phase crawler blocks before launch |
| 6 | Deleting high-traffic pages | Organic traffic from those pages disappears overnight | Audit traffic before deciding what to cut |
| 7 | Not updating internal links | Internal links point to old URLs, creating unnecessary redirect hops and wasting crawl budget | Update all internal links to new URLs |
| 8 | Ignoring meta titles and descriptions | Pages lose their click-through rate in search results, reducing traffic even with stable rankings | Migrate or improve all metadata |
| 9 | No post-launch monitoring | Problems compound undetected for weeks | Monitor daily for the first two weeks |
| 10 | SEO team involved too late | Architectural decisions are already locked in; fixing them is 10x more expensive | Involve SEO from day one of planning |
What to Do If Traffic Has Already Dropped
If you're reading this after a migration that went sideways, here's a triage framework organized by urgency:
First 48 hours (emergency):
- Test all 301 redirects — are they working? Are any returning 404 errors or using 302 instead of 301?
- Check robots.txt — is your site accidentally blocking search engines?
- Check for `noindex` tags left over from the staging environment
- Submit your updated XML sitemap to Google Search Console
- If you changed domains, file a Change of Address request in Search Console
Week one (diagnosis):
- Run a full site crawl to find broken links, missing pages, and redirect errors
- Compare before-and-after traffic in Google Analytics, filtered by organic traffic only
- Identify which specific pages lost the most traffic — these are your priorities
- Check Google Search Console for indexing errors, crawl issues, and coverage gaps
Week two (fix):
- Repair or add missing 301 redirects for every changed URL
- Restore any high-traffic content that was accidentally deleted
- Fix internal links pointing to 404 pages
- Resubmit your XML sitemap after fixes are made
- Verify canonical tags point to the correct new URLs
Week three and beyond (monitor and optimize):
- Track keyword rankings and traffic recovery weekly
- Re-optimize page titles and meta descriptions for underperforming pages
- Rebuild lost authority through outreach and link reclamation
- Schedule regular technical audits going forward
Recovery timeline expectations:
| Issue | Time to Fix | Time to See Results |
|---|---|---|
| Missing redirects | Hours to days | 1–4 weeks |
| Robots.txt blocking crawlers | Minutes | 1–2 weeks |
| Deleted content restoration | Days to weeks | 4–8 weeks |
| Wrong redirect type (302 → 301) | Hours | 2–6 weeks |
| Major structural/architectural issues | Weeks to months | 3–12 months |
How Much Does a Website Migration Cost?
Costs vary enormously depending on what's changing and how large the site is.
| Migration Type | Small Site (< 100 pages) | Medium Site (100–1,000 pages) | Large Site (1,000+ pages) |
|---|---|---|---|
| HTTP to HTTPS | $0–$500 | $500–$2,000 | $2,000–$5,000 |
| Server/hosting change | $500–$1,000 | $1,000–$2,500 | $2,500–$5,000 |
| CMS platform change | $1,000–$5,000 | $5,000–$15,000 | $15,000–$50,000+ |
| Domain change + redirects | $500–$2,500 | $2,500–$7,500 | $7,500–$20,000 |
| Full redesign + replatform | $3,000–$15,000 | $15,000–$50,000 | $50,000–$150,000+ |
Budget reality check: 64% of migration projects overrun their budget. The average overrun is 18%. Build a contingency buffer of at least 20% into your budget from the start.
The cost of getting it wrong: A business generating $2 million annually from organic traffic that suffers a 40% traffic drop for six weeks loses approximately $70,000 in revenue — not counting the cost of emergency fixes. One financial services firm lost $180,000 per month in leads after a botched migration. The SEO planning that prevents this typically costs a fraction of the potential losses.
Questions to Ask Before Hiring a Migration Team
These questions separate teams that know what they're doing from teams that will learn on your budget:
- "Walk me through your migration process." Look for specifics: audit, URL mapping, staging, testing, monitoring. Vague answers like "we'll handle the SEO" are a red flag.
- "How will you handle redirects?" They should describe one-to-one URL mapping with 301 redirects, not "we'll redirect everything to the homepage."
- "Will you do a pre-migration SEO audit?" If not, they're guessing.
- "When does your SEO team get involved?" The answer should be "from the beginning," not "before launch."
- "What's your post-launch monitoring plan?" Expect at least 30 days of active monitoring with defined checkpoints.
- "Can you share a case study of a past migration?" Ask for traffic data before and after. If they can't show this, they either haven't tracked it or the results weren't good.
- "Who is responsible for fixing issues found after launch?" This should be defined in your contract before work begins, with clear timelines for response.
The Bottom Line
Website migration is one of the riskiest things you can do to your online presence. It's also sometimes necessary. The difference between a migration that costs you a year of traffic and one that becomes a growth catalyst comes down to three things:
Plan obsessively. Document everything before you touch anything. Map every URL. Audit every page. Set your baseline.
Execute methodically. Test on staging. Validate redirects. Launch with developer support on standby. Don't combine multiple changes if you can do them sequentially.
Monitor relentlessly. The first 90 days after launch determine the outcome. Daily checks in week one. Twice-weekly in weeks two through six. Weekly through month three.
The companies that follow this checklist don't just survive migrations — they come out stronger. HireRoad merged three domains into one and exceeded their traffic targets by 14.5%. Expat Living recovered from a 68% traffic loss and ended up 43% above their pre-migration numbers within three years. These aren't lucky outcomes. They're the result of treating migration as a strategic project, not a technical checkbox.
The checklist exists. The data is clear. The only variable is whether you follow it.




