Traffic Change Investigation

Fictional example report for Sample & Sons.

Executive summary

The observed change is a booking increase, not a traffic decline: bookings rose by 160 while sessions rose by 2,000. The central question is why bookings grew much faster than visits. The saved data shows higher booking rates within channels and simultaneous changes to the homepage experience. It cannot establish the cause without event checks, release and campaign dates, and lead-quality records. Investigate measurement first, then audience and experience, before making a budget or redesign decision.

Business context

A fictional B2B consultancy helping growing service businesses improve operations and customer experience. The primary goal is qualified consultation bookings. The target audience is operations leaders at growing service businesses.

Evidence and findings

Finding 1

Separate the size of the change from its explanation. Baseline sessions and bookings were 10,000 and 200; current values are 12,000 and 360. Holding the baseline 2% rate constant gives 240 bookings at the current traffic volume. That leaves 120 bookings above the constant-rate scenario, within a total increase of 160. This calculation identifies a conversion-efficiency question; it does not prove the site produced 120 incremental customers or isolate a particular intervention.

Finding 2

The saved channel mix is stable, so a broad mix shift is not the explanation here. Organic, Direct, Paid, and Referral retain session shares of 50%, 20%, 20%, and 10% in both windows. Organic bookings rise from 120 to 210 and its rate from 2.4% to 3.5%. Paid bookings rise from 20 to 48 and its rate from 1% to 2%. Direct rises from 40 to 72, and Referral from 20 to 30. Improvements occur within every listed channel. Finer changes in campaign, query, device, visitor intent, or tracking can still matter and are not ruled out by stable top-level shares.

Finding 3

The homepage changes are plausible contributors, with no causal test. Snapshots show a clearer booking CTA, while mobile LCP improves from 4.2s to 2.6s and interaction and layout metrics improve too. These observations support investigating the website experience. They do not specify when each change went live or show a control group. If bookings increased before a release, that release cannot explain the earlier movement. A dated change log is necessary to test the proposed sequence.

Finding 4

Measurement and business quality remain competing explanations. A duplicate booking event, changed consent behavior, altered key-event configuration, or a different attribution setup could change reported counts without equivalent business growth. None of those defects is demonstrated by the saved audit. The data also lacks campaign spend, CRM qualification, experiment assignments, and a complete release timeline. Treat these as questions to verify, not as diagnosed problems. Reconciliation with actual bookings is the fastest way to determine whether the business increase is real.

Recommended next steps

1. Priority 1 · Analytics specialist + booking owner · First day: Review event configuration and release history, test successful and unsuccessful booking paths, and compare daily analytics booking totals with booking-system records. Investigate duplicate or missing events if the two series diverge. Resolve measurement issues before interpreting conversion uplift or adjusting campaign budgets.
2. Priority 2 · Marketing + developer · First week: Build a timeline of campaign starts, audience changes, page releases, consent changes, and service availability. Compare daily sessions and bookings around those dates, then segment within Organic and Paid. Check whether the increase is sustained, concentrated in a few days, or associated with a particular landing page or device.
3. Priority 3 · Sales owner + analyst · Following weeks: Review fit, attendance, and downstream outcomes for the additional bookings. Label the final explanation as confirmed, supported, or unresolved according to the evidence. Keep budgets stable until cost and quality data justify a change; preserve useful website improvements even if they cannot be credited with the entire increase.

Sources and reporting window

• Baseline audit: September 1, 2026, with GA/GSC data from August 4–31.
• Current audit: September 29, 2026, with GA/GSC data from September 1–28.
• Saved services performance and snapshot.
• Saved GEO evidence.

These equal-length, 28-day windows compare two completed audits across the same four pages.

Applying this to your business

Use this investigation method for an unexpected jump or drop in your own enquiries, orders, or appointments. First verify what changed in the measurement, then separate traffic volume, audience mix, and within-segment conversion. Compare the timing with real business events such as a campaign, promotion, outage, or seasonal demand. An investigation report should help rule explanations in or out, not turn the most visible recent edit into an automatic cause.

Confidence and limitations

The measurements support the direction of change, not a causal claim. Booking events do not establish revenue or lead quality. Seasonality, channel mix, and measurement changes can affect the comparison. All numbers, business details, and findings are authored examples rather than a live assessment.

How to validate progress

The investigation is resolved when reported bookings reconcile with operational records, the relevant timeline is documented, and remaining explanations are explicitly qualified. Continue monitoring equal-length windows and daily patterns. Stable channel shares and higher within-channel rates support an efficiency change here; they do not eliminate seasonality or finer audience differences. If cause remains uncertain, propose a controlled next test rather than rewriting history as a proven success story.

Use Chat to ask follow-up questions against the saved report and audit evidence.