Web analytics business intelligence: connect results

Web analytics business intelligence connects acquisition-channel traffic with purchasing behavior and reconciled order revenue. Compare the same periods and channel definitions, then separate changes in traffic, conversion, and order value.

A channel can generate more revenue while converting a smaller share of its sessions. The useful management question is which change explains the result, and what evidence is still missing before changing spending. The business intelligence overview supplies the broader decision framework.

Define the measures before combining exports

Use monthly website and completed-order exports. Document the reporting time zone, date range, currency, channel grouping, and revenue treatment. In this example, revenue means completed-order value after discounts and refunds, excluding sales tax and shipping charges.

If the website report comes from Google Analytics, its session documentation defines a session as a period of interaction with a website or app. Sessions are not unique people. Keep that denominator consistent between months.

Use session acquisition channels for a session-based comparison. Google's Traffic acquisition report documentation distinguishes traffic acquisition, covering new and returning users, from user acquisition, covering new users. An order assigned to an original customer-acquisition channel cannot automatically be compared with traffic assigned to the latest session's channel.

Give every order one channel under the documented policy. Keep unassigned orders visible. If existing exports cannot support a consistent assignment, report that gap rather than presenting unmatched totals as a channel conversion analysis.

Build a small monthly control table

The fictional US retailer below has one completed order per purchasing session. That makes purchase-session conversion and orders per session equal in this example; your business may differ.

Place month-start date in Raw!A, channel in B, sessions in C, purchasing sessions in D, completed orders in E, and net revenue in F. Rows 2 through 7 contain:

August totals are 20,000 sessions, 540 orders, and $108,000 net revenue. September totals are 24,000 sessions, 636 orders, and $127,200. Match the order counts and amounts to the completed-order export for the same scope. Then bridge that export to the accounting control total, explaining any timing or recognition differences.

Do not join a channel session total to every order and then sum sessions: that repeats the denominator. Aggregate both sources to the same month-and-channel level first.

Calculate conversion and revenue per session

Put the selected channel in Summary!A2 and September 1, 2026, in L2. In Microsoft Excel, calculate September sessions in B2 with:

=SUMIFS(Raw!$C$2:$C$7,Raw!$B$2:$B$7,$A2,Raw!$A$2:$A$7,$L$2)

Calculate purchasing sessions in C2, orders in D2, and revenue in E2, respectively:

=SUMIFS(Raw!$D$2:$D$7,Raw!$B$2:$B$7,$A2,Raw!$A$2:$A$7,$L$2)
=SUMIFS(Raw!$E$2:$E$7,Raw!$B$2:$B$7,$A2,Raw!$A$2:$A$7,$L$2)
=SUMIFS(Raw!$F$2:$F$7,Raw!$B$2:$B$7,$A2,Raw!$A$2:$A$7,$L$2)

Microsoft's SUMIFS documentation describes the multiple-criteria aggregation used here. Covirage is not affiliated with Google or Microsoft.

In F2, calculate purchase-session conversion:

=IF(B2=0,"",C2/B2)

In G2, calculate revenue per session:

=IF(B2=0,"",E2/B2)

In H2, calculate average order value:

=IF(D2=0,"",E2/D2)

Repeat the summary using August's month-start date. The comparison becomes:

Round rates to two decimal places and dollar ratios to cents. Total conversion divides total purchasing sessions by total sessions; do not average channel percentages. Each channel's average order value is $200 in both periods, calculated from its revenue and order count.

Explain the change without claiming a cause

Paid search revenue grew by $3,200 even though conversion declined from 2.00% to 1.80%. With average order value unchanged, calculate the traffic effect at the old rate, followed by the conversion effect at current traffic:

These effects reconcile to $43,200 minus $40,000. They are an arithmetic explanation, not evidence that a campaign caused either movement. Review audience, offer, device, and product mix before recommending a spending change. Business intelligence for ecommerce covers the wider commercial context.

Where comparisons go wrong

• Purchase events counted as purchasing sessions. Duplicate events or multiple orders distort a session conversion numerator.
• Revenue definitions differ. Gross website purchase value and revenue after refunds need a documented bridge.
• Channel assignment changes. Reclassifying traffic can move reported performance without changing total sales.
• Zero traffic becomes zero conversion. A missing denominator is unknown, not measured failure.

Use business intelligence charts to present the reconciled comparison. For recurring periods, automated business intelligence explains which checks to standardize.

Compare your own web and order files in Covirage

Bring your channel summaries and completed-order export with the period and revenue definitions. Covirage maps the columns, and its tools compute and check the comparisons; the external AI model explains the results with every figure traced to its rows. Create a free Covirage account, which takes about a minute, then set up your business and bring a sample of your own data.

See revenue driver analysis for the supporting business questions.