Maintenance

Once your Cloudflare Pages site is live, maintenance is minimal compared to WordPress. This guide covers day-to-day operations, monitoring, and long-term management.

---

The Maintenance Difference

WordPress (Before)

Task	Time	Frequency
WordPress core updates	15-30 min	Monthly
Plugin updates (10-20 plugins)	30-60 min	Monthly
Theme updates	15-30 min	Quarterly
Security scan review	15-30 min	Weekly
Backup verification	15-30 min	Monthly
Database optimization	15-30 min	Monthly
PHP version check	15-30 min	Quarterly
Hosting plan review	15 min	Annually
SSL certificate renewal	15-30 min	Annually (if not auto)
Performance tuning	30-60 min	Monthly
Comment spam cleanup	15-30 min	Weekly
Total	3-6 hours/month

Cloudflare Pages (After)

Task	Time	Frequency
Review Cloudflare analytics	10 min	Weekly
Check form submission delivery	5 min	Weekly
Lighthouse audit	10 min	Monthly
Review SendGrid activity	5 min	Monthly
Total	30-45 min/month

Tip

Time savings: 85-90% reduction in maintenance overhead. Moving from 3-6 hours/month to 30-45 minutes/month per site.

---

Maintenance Schedule

The following calendar shows the monthly routine at a glance:

gantt
    title Monthly Maintenance Routine
    dateFormat  YYYY-MM-DD
    axisFormat %b %d
    section Week 1
    Review analytics (10 min)        :a1, 2024-01-01, 1d
    section Week 2
    Check form submissions (5 min)   :b1, 2024-01-08, 1d
    section Week 3
    Review analytics (10 min)        :c1, 2024-01-15, 1d
    section Week 4
    Monthly audit (30 min)           :crit, d1, 2024-01-22, 1d

Week	Task	Time	Details
Week 1	Review analytics	10 min	Cloudflare Dashboard: traffic, errors, threats
Week 2	Check form submissions	5 min	Verify emails arriving, compare volume to baseline
Week 3	Review analytics	10 min	Cloudflare Dashboard: traffic, errors, threats
Week 4	Monthly audit	30 min	Lighthouse, SendGrid, dependency check
Total		~55 min/month	vs WordPress: 3-6 hours/month

---

Daily Operations

First 2 Weeks Post-Launch

Caution

During the first 2 weeks after launch, monitor daily until patterns stabilize.

• Form submissions: Check email inbox for expected form submissions
• Site accessibility: Load site in browser, verify pages work
• Error check: Cloudflare Dashboard > Analytics > quick scan for 5xx errors

Steady State

For most sites: nothing. Cloudflare Pages is designed to run without daily attention. Only monitor daily if:

High-Traffic Site

Site receives significant traffic or is revenue-generating.

Active PPC Campaigns

Running PPC campaigns that depend on form submissions.

Recent Changes

In the first 2 weeks after a major change or deployment.

---

Weekly Operations

Monitoring Checklist (10-15 Minutes)

1. Cloudflare Analytics

   • Go to https://dash.cloudflare.com/
   • Select domain > Analytics
   • Check: Total requests (normal range?)
   • Check: Error rates (any 5xx spikes?)
   • Check: Threat activity (any unusual patterns?)

2. Form Submissions

   • Check recipient inbox
   • Compare volume to baseline (significant drop = investigate)
   • If using SendGrid: check https://app.sendgrid.com/ > Email Activity

3. Quick Site Check

   • Load the site in a browser
   • Submit a test form (optional — weekly)
   • Check on mobile (optional)

---

Monthly Operations

Performance Audit (10-15 Minutes)

# Run Lighthouse audit
npm run audit:baseline

# Or manually:
npx lighthouse https://yoursite.com/ \
  --only-categories=performance \
  --output=json \
  --output-path=./lighthouse-monthly.json \
  --chrome-flags="--headless"

What to check:

• Lighthouse score stable (no significant drops)
• LCP still < 2.5s
• No new CLS issues
• TBT within acceptable range

Note

If score drops: Check if Cloudflare features changed (Rocket Loader, etc.), if new third-party scripts were added, or if large images were added.

SendGrid Review (5 Minutes)

1. Go to https://app.sendgrid.com/
2. Check: Email Activity > Deliverability
3. Check: No bounces or blocks
4. Check: Usage within free tier (< 100 emails/day)

Cloudflare Usage Review (5 Minutes)

1. Dashboard > Pages > your-project
2. Check: Build count this month (limit: 500/month free)
3. Check: Function invocations (limit: 100k/day free)
4. If approaching limits: consider reducing deploy frequency or upgrading

---

Quarterly Operations

Comprehensive Review (30-60 Minutes)

• Performance trend: Compare Lighthouse scores over 3 months
• SEO health: Check Google Search Console for issues
• Security headers: Test at https://securityheaders.com/
• SSL certificate: Verify valid and auto-renewing
• SendGrid account: Review sender reputation
• Content freshness: Review if content needs updating
• Dependencies: Check if Wrangler needs updating: npm outdated
• Documentation: Update any outdated procedures

Dependency Updates

# Check for outdated packages
npm outdated

# Update non-breaking changes
npm update

# Check Wrangler version
npx wrangler --version

# Update Wrangler
npm install wrangler@latest --save-dev

Tip

Unlike WordPress, there are no urgent security patches. Dependency updates are for features and bug fixes, not security emergencies.

---

Content Update Procedures

Updating Text Content

Change Type	Recommended Tier	Time
Phone number	Tier 1 (Browser) or Tier 3 (AI)	2-5 min
Business hours	Tier 1 or Tier 3	2-5 min
Service description	Tier 1 or Tier 2 (CMS)	5-10 min
New blog post	Tier 2 (CMS)	10-30 min
New page	Tier 2 (CMS) or Tier 4 (Dev)	15-60 min
Design changes	Tier 4 (Dev) or Tier 3 (AI)	30-120 min

Updating Forms

Change Recipients

1. Edit functions/api/submit-form.js
2. Update the RECIPIENT_EMAILS array
3. Deploy: npm run deploy
4. Test by submitting a form

Add/Remove Fields

1. Edit the HTML form (public/*/index.html)
2. Update the Function to extract the new field (functions/api/submit-form.js)
3. Deploy and test

Updating Images

1. Add new image to public/images/ (or appropriate directory)
2. Update the <img src=""> in HTML
3. Deploy

Tip

Optimize before uploading: Convert to WebP, use appropriate dimensions (don’t upload 4000px for a 400px display), and add width/height attributes to prevent CLS.

---

Monitoring Tools

Built-in (Free)

Cloudflare Analytics

Traffic, errors, threats, bandwidth. Access via Dashboard > Analytics.

Cloudflare Pages Logs

Deployment status, Function errors. Access via Dashboard > Pages > Deployments.

SendGrid Activity

Email delivery, bounces, opens. Access at https://app.sendgrid.com/.

Google Search Console

SEO health, crawl errors, rankings. Access at https://search.google.com/search-console/.

Cloudflare Web Analytics

Privacy-friendly visitor analytics. Access via Dashboard > Analytics > Web Analytics.

External (Optional)

Tool	Purpose	Cost
UptimeRobot	Uptime monitoring, alerts	Free (50 monitors)
Pingdom	Performance monitoring	$10+/month
Google Analytics (GA4)	Detailed visitor analytics	Free
PageSpeed Insights	Performance scoring	Free
GTmetrix	Performance analysis	Free (limited)

Setting Up UptimeRobot (Recommended)

Free monitoring with email/SMS alerts:

1. Sign up at https://uptimerobot.com/
2. Add monitor:
   • Monitor Type: HTTPS
   • Friendly Name: “YourSite.com”
   • URL: https://yoursite.com/
   • Monitoring Interval: 5 minutes
3. Configure alert contacts (email, SMS, Slack)
4. Add monitors for each site

Note

What it does: Pings your site every 5 minutes. If it’s down, you get an alert. Keeps a log of uptime percentage.

---

Rollback Procedures

When to Rollback

• Site showing errors after a deployment
• Forms broken after a code change
• Visual layout broken
• Performance significantly degraded

How to Rollback

Dashboard (Fastest)

Time: ~30 seconds

1. Go to https://dash.cloudflare.com/
2. Pages > Your Project > Deployments
3. Find the last working deployment
4. Click the three dots menu > “Rollback to this deployment”
5. Confirm
6. Site reverts immediately

Git Revert

Time: ~2-5 minutes

# View recent commits
git log --oneline -10

# Revert the last commit
git revert HEAD

# Deploy the reverted version
npm run deploy

Emergency Checkout

Time: ~2-5 minutes

# Checkout a specific known-good commit
git checkout <commit-hash> -- public/ functions/

# Deploy immediately
npm run deploy

# Then fix the issue on a branch

Post-Rollback

1. Verify site works correctly
2. Identify what caused the issue
3. Fix the issue in code
4. Test fix locally
5. Deploy fix to staging
6. Test staging
7. Deploy to production

---

Scaling Considerations

Single Site

No special considerations. Free tier handles any single site easily.

10-50 Sites

Consideration	Recommendation
Cloudflare account	One account for all sites (Acme Medical)
Build limits	500 builds/month shared. Budget ~10 deploys/site/month
SendGrid	One account, multiple sender addresses
Monitoring	UptimeRobot with monitors for each site
Repository	One repo per site, or monorepo with subdirectories

50-300 Sites

Consideration	Recommendation
Cloudflare plan	Consider Pro ($20/month) for advanced analytics
Build limits	May need paid plan for more builds
SendGrid	May need paid plan for more emails
Monitoring	Centralized dashboard (UptimeRobot, Grafana)
Content editing	AI-assisted editing becomes essential
Batch operations	Scripted batch deploys, automated testing
Repository	Monorepo with shared components

Automation at Scale

For 100+ sites, automate common operations:

Batch Lighthouse Audits

audit-all-sites.sh

#!/bin/bash
SITES=(
  "https://site1.com"
  "https://site2.com"
  "https://site3.com"
)

for site in "${SITES[@]}"; do
  echo "Auditing $site..."
  npx lighthouse "$site" \
    --only-categories=performance \
    --output=json \
    --output-path="./audits/$(echo $site | sed 's/https:\/\///;s/\//-/g').json" \
    --chrome-flags="--headless"
done

echo "All audits complete. Results in ./audits/"

Batch Deploy (Monorepo)

deploy-all.sh

#!/bin/bash
PROJECTS=(
  "cloudflare-builder-drsmith"
  "cloudflare-builder-clinicsite"
  "cloudflare-lp-riverside"
)

for project in "${PROJECTS[@]}"; do
  echo "Deploying $project..."
  cd "$project"
  npx wrangler pages deploy public
  cd ..
done

---

Emergency Procedures

Danger

For production incidents, act quickly but methodically. Every previous deployment is saved and can be restored.

Site Down

Site Down

1. Check Cloudflare Status: https://www.cloudflarestatus.com/
2. If Cloudflare is up: Check your deployment in Dashboard > Pages
3. If recent deployment: Rollback immediately (see Rollback section)
4. If no recent changes: Check DNS with dig yoursite.com
5. Contact Cloudflare support if needed

Forms Broken

Forms Broken

1. Check browser console for JavaScript errors (F12)
2. Check SendGrid dashboard for delivery issues
3. Verify SENDGRID_API_KEY in environment variables
4. Test locally: npm run serve
5. If local works but production doesn’t: redeploy with npm run deploy
6. If still broken: rollback to last working deployment

Bad Deployment

Accidental Bad Deployment

1. Don’t panic — every previous deployment is saved
2. Go to Dashboard > Pages > Deployments
3. Find the last known-good deployment
4. Click Rollback
5. Fix the issue, test locally, redeploy

SSL Issues

SSL Certificate Issues

1. SSL certificates are automatic on Cloudflare
2. If expired: Check Dashboard > SSL/TLS > Edge Certificates
3. Usually resolves automatically within hours
4. If persistent: Contact Cloudflare support

SendGrid Suspended

SendGrid Account Suspended

1. Check SendGrid dashboard for suspension notice
2. Review email activity for spam complaints
3. Contact SendGrid support
4. Backup: Switch to alternative email service (Resend, Mailgun)
5. Update SENDGRID_API_KEY in environment variables
6. Update functions/api/submit-form.js if API format differs

---

Long-Term Planning

Annual Review Checklist

• Cost review: Verify still on free tier, check for unexpected charges
• Performance review: Run full Lighthouse audit on all sites
• SEO review: Check Google Search Console for all sites
• Security review: Check security headers, review access permissions
• Content review: Flag outdated content for client review
• Technology review: Check for Wrangler/Astro/Cloudflare updates
• Process review: Identify bottlenecks, improve workflows
• Training review: New team members trained, documentation current

When to Consider Upgrading

Signal	Upgrade Path	Cost
>500 builds/month	Cloudflare Pro	$20/month
>100 emails/day	SendGrid Essentials	$15/month
Need advanced analytics	Cloudflare Pro	$20/month
Need WAF rules	Cloudflare Pro	$20/month
>100k Function requests/day	Cloudflare Workers Paid	$5/month

Future Enhancements

Astro Migration

When: Building new sites. Effort: Medium.

Decap CMS

When: Non-devs need to edit. Effort: Medium.

AI-Assisted Editing

When: Managing 10+ sites. Effort: Low.

Automated Testing

When: Deploying frequently. Effort: Medium.

GitHub Actions CI/CD

When: Wanting auto-deploy. Effort: Low.

Cloudflare Turnstile

When: Getting form spam. Effort: Low.

Cloudflare Web Analytics

When: Wanting privacy-friendly analytics. Effort: Low.

---

Documentation Maintenance

Keep These Updated

Document	Update When
DEVELOPER-HANDOFF.md	Team changes, process changes
QUICK-REFERENCE.md	New commands, URL changes
This maintenance guide	New procedures or tools
Contact information	Team member changes

Document Changes

For every significant change to a site:

1. Note what changed
2. Note when
3. Note who made the change
4. Update relevant documentation

---

Summary: Monthly Maintenance Routine

Week 1

Review analytics — 10 min. Check Cloudflare Dashboard for traffic and error patterns.

Week 2

Check form submissions — 5 min. Verify emails arriving and volume is normal.

Week 3

Review analytics — 10 min. Check Cloudflare Dashboard for traffic and error patterns.

Week 4

Monthly audit — 30 min. Lighthouse, SendGrid review, dependency check.

Total: ~55 minutes/month per site

vs WordPress: 3-6 hours/month per site