Skip to main content

Automatic Issue Detection

TraceLog watches your site 24/7 and tells you when something is wrong — before your customers do. You don't need to configure anything: issues appear automatically in your dashboard as soon as TraceLog has enough data.

What an Issue Looks Like​

Every issue TraceLog detects includes:

  • What happened — a plain-language summary (e.g., "Conversion rate dropped 28% on mobile this week")
  • Severity — how urgent it is: info, warning, or critical
  • Affected sessions — how many visitors were impacted
  • Revenue impact — estimated weekly and monthly cost, when revenue tracking is configured
  • Likely cause — what TraceLog thinks is behind the problem (e.g., "JS error on checkout page affecting Safari 17+")
  • Suggested action — a specific recommendation for what to fix (AI-generated based on your data)

Issues appear as ranked cards on the Overview — your project's home — ordered by revenue impact alongside recommendations. Old /issues bookmarks redirect there automatically.

What to Do When You See an Issue​

  1. Read the summary — understand what changed and how many sessions are affected
  2. Check the revenue impact — prioritize issues that cost the most
  3. Review the suggested action — TraceLog explains the likely cause and what to fix
  4. Ask — open Ask and pose follow-up questions ("Which pages are affected?", "When did this start?")
  5. Fix and monitor — after fixing, resolve the card on the Overview; TraceLog measures whether the fix worked and what it recovered
tip

Issues with revenue impact are the highest-priority items. A "critical" issue without revenue impact (like a tracking gap) is less urgent than a "warning" issue that costs €2,000/week.

What TraceLog Detects​

30+ built-in detectors run automatically across the areas below. You don't configure or enable them — they all work out of the box.

Revenue & Conversions​

TraceLog watches your conversion funnel and alerts you when something drops:

  • Conversion rate drops (volume or percentage)
  • Revenue per session changes
  • Revenue shifts across traffic sources
  • Engagement rate degradation or bounce rate spikes
  • Session depth or duration collapse
  • Underperforming campaigns

Errors & Performance​

Catches technical problems that affect user experience:

  • JavaScript errors — classified by severity and whether users showed frustration signals (rage clicks, immediate exits)
  • New error types appearing for the first time
  • Core Web Vitals degradation (LCP, INP, CLS, TTFB)
  • Performance regressions compared to your baseline
  • SEO-impacting issues on organic landing pages

Traffic & Acquisition​

Monitors where your visitors come from and whether something changed:

  • Traffic spikes or drops (daily and weekly baselines)
  • Shifts in geographic, device, or source distribution
  • Low-quality traffic sources (high bounce, low engagement)
  • Bot traffic detection (zero-duration sessions, suspicious patterns)

Device & Browser​

Finds problems that only affect specific devices:

  • Error or engagement regressions on specific devices or browsers
  • Mobile-specific issues (separate from desktop baseline)

Pages & Funnels​

Watches individual pages and defined funnels:

  • Page-level anomalies in views or exit rates
  • Drop-offs in conversion funnels you've defined

Custom Events​

If you're sending custom events, TraceLog monitors them too:

  • Unusual changes in event firing patterns (sudden spikes or drops)

Opportunities​

Not all issues are problems. TraceLog also surfaces things going well:

  • Traffic growth and improving metrics
  • High-converting visitor segments you could double down on
  • Landing pages with untapped potential
  • Best-performing traffic channels
  • Time-of-day patterns in user behavior
  • High-retention pages worth investing in

Revenue Impact​

When revenue tracking is configured, issues include an estimated financial cost:

What you seeWhat it means
€2,800/weekEstimated revenue loss from this issue
Confidence: HighBased on 100+ conversions — reliable estimate
Confidence: MediumBased on 20–99 conversions — directional estimate
Confidence: LowUnder 20 conversions — rough estimate
No revenue shownNot enough data for a meaningful estimate (hidden, not guessed)

Issues are still detected without revenue tracking — they just won't include financial projections. Enable revenue tracking in Project Settings > Revenue to see the cost of every issue.

When Issues Appear​

  • Issue detection starts working from your first session
  • Statistical detectors (conversion drops, traffic anomalies) need 5,000+ sessions/month to reliably distinguish real problems from normal variance
  • Below that threshold, TraceLog still detects errors, performance issues, and tracking health problems — these don't require statistical baselines

Proactive Email Alerts​

TraceLog doesn't just show issues in the dashboard — it emails you when something urgent happens:

  • Proactive alerts run every 6 hours, checking for conversion drops and traffic anomalies
  • Issue alerts are sent when a high-impact issue is detected. The email includes the human-readable title, the suggested action to take, and the impact line (revenue, sessions, or affected page) so you can decide whether to act without opening the dashboard
  • Follow-up emails go out after several days if a critical issue is still active and unresolved — same format, with a small "Open Nd" badge and the original detection date so you know it isn't a fresh alert
  • First issue notification lets you know when TraceLog finds its first issue in your project

Configure which alerts you receive and who gets them in Project Settings > Notifications. You can also set up metric-based alert rules for custom thresholds.