Skip to content

Area: User guide

How analytics works

Analytics has two pages. Dashboard reports your store for a period: revenue, traffic, engagement, conversions, your audience and your messaging channels. Customer Insights looks at the people: lifecycle stages, starred segments, funnels and cohorts. Everything is computed from the events your shop sends, so the numbers are only as complete as your tracking. See How events get into Releva.

Analytics reads data only. Nothing here sends a message or changes a campaign.

Every dashboard tab shares one date range, chosen with the button above the tabs. It opens on the last 14 days, including today. The presets are 1 week, 2 weeks, 1 month and 3 months; the calendar lets you pick any start and end day in the past.

Each figure is compared with the previous period: a range of the same length that ends the day before your start date. That is what every vs previous period and Previous: value refers to.

The dashboard range is kept while you switch tabs but not when you leave the page: it is back to 14 days on your next visit. The attributed orders page and each Customer Insights tab keep a period of their own.

An order is attributed to Releva when the buyer opened or clicked a Releva message, or clicked an on-site block, before ordering: within 7 days for messages, 1 day for on-site surfaces. It counts orders Releva took part in, not orders it can prove it caused.

Under Settings › Shop & Integration › Events & Analytics, Campaign Attribution chooses whether an open counts: Opens & Clicks (the default) or Clicks only. With Clicks, the At a Glance caption reads “clicked” instead of “opened or clicked”.

One order can be credited to several surfaces at once: a buyer who clicked an email and then a product block counts for both. That is why the tiles in the Attributed revenue breakdown overlap and do not add up to the attributed total.

Refunds are subtracted from revenue on At a Glance and on the Revenue tab, but never from order counts. The attributed orders page flags refunds in its Status column without subtracting them, so its totals can read a little above the At a Glance card.

Tab What it shows
At a Glance Attributed revenue against Total store revenue with their shares; your starred funnel; the attributed revenue breakdown by mode (One-off, Continuous & triggered), by channel and by on-site surface; Top performing continuous & triggered campaigns, Recent one-off campaigns, Top performing product blocks and Top performing banner blocks; View attributed orders.
Traffic Page views and assisted page views with their previous values, Releva Contributed vs Other Page Views, and the top 3 page views by UTM source and medium.
Engagement How often people acted on what Releva showed them, and Where people engaged: clicks by surface, on the site and in messages.
Conversion The events your shop counts as a conversion, and Does engaging with Releva correlate with buying?: the conversion rate of people who engaged with a Releva surface against those who did not. This is correlation, not a controlled test; for causal lift use a control group.
Revenue Releva Contributed vs Other Revenue, New vs Repeat Customers (orders, revenue and shares by first purchase), revenue over time, and store revenue and average order value by UTM source and medium.
User Base Registered Users, Subscribers and Subscribes and Unsubscribes over the period.
Marketing Per channel an Activity chart and a Sends chart; emails also get opens by weekday and by hour and bounces. Email, push notification and webhook charts are always shown; SMS, Viber and WhatsApp charts appear only when the channel is set up.
Data Quality The health of your Meta pixel and Conversions API: event match quality and how pixel purchases compare with the purchases Releva sees. Without a pixel it reads No Facebook pixel configured.

The line Conversion metric: on At a Glance names the event actions your shop counts as a purchase. On a shop with several purchase actions, every one of them is counted.

The dashboard has no export button. Exports of campaign and channel results live under Analytics › Exports.

For every card and tile on each tab, with what it counts and what it is compared with, see What each Analytics metric means. For a walkthrough, see Check how your store is doing.

View attributed orders on At a Glance opens Orders attributed to Releva for the same period. Each row is one order with its time and purchase action, order id, refund status, profile, the attributed amount out of the order value, and the surfaces it was Credited to: a campaign, a channel message, a product block, a banner, a story, an ad, the sales agent or search as you type. Expanding a row with Show products lists each product line and what it was attributed to; a line with no credited surface is not in the attributed total.

The period, page and page size are in the URL, so a link to this page can be shared.

RFM scores every profile on recency, frequency and monetary value and places it in a lifecycle stage: Champions, Loyal, Recent, At risk, Needs attention or Inactive. The tab compares the distribution on two days, Start date (30 days ago by default) and End date (today), with the change per stage and the average monetary score. It is a comparison of two snapshots, not a window. A shop whose history collection has only just started sees History collection has only just started, so there is no earlier day to compare against yet.

Reports the segments you have starred on Data Platform › Segments, up to 25 per shop: members, size change, conversions and conversion value, with one column per channel. See Track a segment in Audience Performance.

A funnel is an ordered list of 2 to 6 steps. Each step is a Standard event (Purchase, Refund, Cart Add or Cart Create, resolved against your shop’s own action names when the funnel is measured) or a set of Event actions from the actions your shop has sent; the step is reached by any of them. An optional Audience segment restricts the funnel to a Behavior segment.

A funnel counts the profiles that completed the steps in order within the selected period; the conversion window is the period itself. The card reports the profiles at each step and the % of previous step. A shop can have up to 5 funnels, and one of them is starred: that one is shown on At a Glance. The funnel form has no star control; the starred funnel is the shop’s default one, so edit it if At a Glance should follow different steps. Deleting a funnel does not keep its history.

See Create a funnel.

A cohort is the group of customers who made their first purchase in the same month or week. The table follows each cohort over the periods after that (0 mo, 1 mo, …) with a row for All cohorts. Choose the Interval (Monthly or Weekly), the Measure (Profiles, Orders or Revenue) and whether to show counts (#) or percentages (%). The default range is the last 365 days. Revenue nets refunds; order counts do not change.

Releva marks the links it serves so you can find them in Google Analytics 4 (GA4) as well.

  • A product block link carries rlv_rid (the block’s token) and rlv_rpid (the product). A banner link carries rlv_bid. In a GA4 Exploration, add a segment with the condition Page path + query string contains rlv_rid for product blocks or rlv_bid for banners to see the sessions and purchasers that came through them.
  • A link in a Releva message goes through Releva’s click tracking, which adds utm_source=releva, utm_medium=<channel> and utm_campaign=<message name> (the name of the email, SMS or other message that was clicked) to the destination. In GA4 these appear as the Session source, Session medium and Session campaign dimensions, so an Exploration with Session campaign as a row dimension lists each message by name; filter on contains with part of a name to compare the variants of an A/B test. The UTM tagging is on by default for every shop; subscription-management links are never tagged.

The GA4 Exploration steps themselves are Google’s; see Google’s own help for the current layout.

  • AnalyticsCheck how your store is doing

    Pick a period and read your attributed and total revenue, the funnel and the attribution breakdown.

  • AnalyticsCreate a funnel

    Define a funnel from your events and see how many profiles reach each step.

  • Campaigns & WorkflowsHow campaigns work

    Learn how campaign goals and statistics feed the revenue Releva attributes.

  • Store dataHow orders and refunds work

    Learn how orders and refunds get into Releva, the source of every revenue number.