When numbers don't match

Transactions and conversions differ

• Transactions is the number of orders (orders placed with the store).
• Conversions is the number of sessions with a purchase (purchases the tracking tag saw).

Ad blockers, and orders entered by hand in your store admin, don’t show up in tracking, so transactions comes out higher. CVR is calculated from the latter — so that the numerator and denominator come from the same layer (tracking). Shopify’s reports show the same difference.

Revenue in the Acquisition and Pages tables doesn’t match Overview revenue

Overview revenue is built from orders; revenue in the Acquisition, Pages, and device tables is built from tracked purchases. Orders don’t record “where the visitor came from”, so attributing revenue to sources and pages requires tracking. That’s why the note under those tables says “Revenue in this table is built from tracked purchases”, and why the totals don’t match Overview (a difference of a few percent in practice).

Revenue changes when you filter

While a filter is applied, Overview revenue, transactions, and AOV are also built from tracking (for the same reason as above). Remove the filter and they return to the order numbers.

A later funnel stage is larger than an earlier one

This happens when there’s a path to checkout that bypasses the cart, such as “Buy now”. If that doesn’t apply to you, check whether the tracking tag is installed twice.

Also, each funnel stage is “the number of sessions that reached that stage”. The final purchase stage won’t match transactions (for the same reason as “Transactions and conversions” above).

The cart and order-complete pages appear as landing pages

They aren’t real entry points, so their CVR and revenue come out high. Don’t use them for entry-point comparisons (the table marks them as “functional pages”).

Improvement cards don’t change when you change the period or filter

Improvement cards are chosen across the whole site, over the last 28 days. If we re-selected them within the filtered range only, the population would shrink and the cards would be swapped out wholesale, making them useless as a priority order. Within each card, the target pages and sources are named explicitly.

Speed scores move from day to day

Google PageSpeed Insights is a single measurement at that moment. It goes up and down with network and server conditions. For pages where the measurement was too heavy to complete, we show real-visitor measurements (last 28 days) instead.

Numbers differ from GA4

• Sessions: the 30-minute break is the same, but details differ — for example, GA4 continues a session across midnight.
• Users: NextRise’s identifier changes daily, so “users” over multiple days comes out higher than GA4 (the same person is counted as a different person on different days).
• Revenue: NextRise uses the order line-item total (excluding tax and shipping). Stores that send a tax- and shipping-inclusive value in GA4’s purchase event will see a higher figure in GA4.
• Channels: the classification rules follow GA4’s default channel grouping, but if you use a value outside the rules in utm_medium (for example sns), both tools put it under Unassigned.

Data quality differs between the GA4-imported period and after it

Before the boundary (the day you installed the tracking tag), the data is GA4’s daily totals; after it, our own tracking. GA4 holds only the top N, so the tail of each breakdown is grouped as “(other)” and per-product views cover only the top items. That’s why breakdowns appear to change partway through a period that spans the boundary.

If it still doesn’t match

Remove the period, comparison, and filter settings entirely, then compare Overview revenue and transactions against the numbers in your store admin for the same period and the same time zone. If there’s still a large difference, tell us the site name and period through Contact.