Analytics metrics glossary

Metric

Definition

Formula

Notes

Active subscribers

Unique customers who hold at least one active subscription as of the end of the selected period. A customer with multiple active subscriptions is counted once.

Count of distinct customers with subscriber status = Active as of end date.

Also shown as "Active" in the Subscriber activity breakdown panel.

Active subscribed products quantity

Total quantity of products currently subscribed to across all active subscriptions. E.g. one subscriber with 2 active subscriptions of 3 products each contributes 6.

Sum of line-item quantity across all active subscriptions, as of end date.

Counts product units, not customers or contracts - a single customer can contribute many units.

Active MRR

Monthly recurring revenue from all active subscriptions, including those currently in payment recovery.

Sum of line-item MRR (normalized to a monthly cadence) across active subscriptions. E.g. a $60 order billed every 2 months = $30 MRR.

Subscriptions in dunning still count as Active until retries are exhausted, so their MRR stays in this figure. One-time add-ons are excluded. Shipping and taxes are excluded.

Subscribers trend

How the active subscriber base has moved over time, showing additions (new, reactivated, resumed) and reductions (paused, churned, expired) per bucket.

Running balance: previous bucket's active count + additions − reductions = current bucket's active count.

Since each bucket builds on the last, an error in one period's count carries forward into every later bucket.

Subscriber additions

Subscribers who started a new subscription, resumed a paused one, or reactivated a cancelled/expired one during the period.

New + Reactivated + Resumed subscribers (deduplicated per customer).

"Reactivated" includes a customer with zero active subscriptions starting a brand-new one - it isn't limited to reactivating the same subscription.

Subscriber reductions

Subscribers who paused, cancelled, or expired all of their subscriptions during the period.

Paused + Churned + Expired subscribers (deduplicated per customer).

"Churned" here is the same concept as Subscribers lost in Cancellation analytics.

Net active subscribers change

Net gain or loss in active subscribers over the period.

Active subscribers (period end) − Active subscribers (day before period start).

Should algebraically match Additions − Reductions; if the two disagree, treat it as a data flag worth checking.

Active subscribers by selling plan & delivery interval

Distribution of active subscribers and MRR across each selling plan and delivery frequency combination.

Per plan/frequency pair: subscriber count, MRR, and % share of total active subscribers.

Also viewable by selling plan alone or delivery interval alone. A subscriber holding plans across multiple groups is counted once per group, so totals can exceed the overall active count.

Active subscriptions

Total number of subscription contracts currently active, at the contract (not customer) level.

Count of subscription contracts with status = Active as of end date.

One customer with 3 active subscriptions contributes 3 here but only 1 to Active subscribers.

Subscriptions trend

How the active subscription base has moved over time, showing additions and reductions per bucket, at the contract level.

Running balance: previous bucket's active count + additions − reductions = current bucket's count.

Contract-level counterpart to Subscribers trend — use this when a customer's multiple subscriptions should count separately.

Subscription additions

Subscription contracts that started new, resumed, or reactivated during the period.

New + Reactivated + Resumed contracts.

Contract-level version of Subscriber additions.

Subscription reductions

Subscription contracts that were paused, cancelled, or expired during the period.

Paused + Cancelled + Expired contracts.

This is the contract-level equivalent of Subscriptions cancelled plus pauses/expirations.

Net active subscriptions change

Net gain or loss in active subscription contracts over the period.

Active subscriptions (period end) − Active subscriptions (day before period start).

Contract-level counterpart to Net active subscribers change.

Subscribed product additions

Subscribed product quantity added during the period, split by how it was added (new, resumed, reactivated).

Sum of line-item quantity added, split by addition mode.

Unit-level counterpart to Subscriber/Subscription additions — measures product quantity rather than customers or contracts.

Subscribed product reductions

Subscribed product quantity removed during the period, split by how it was removed (paused, cancelled, expired).

Sum of line-item quantity removed, split by reduction mode.

Unit-level counterpart to Subscriber/Subscription reductions.

Net subscribed products change

Net gain or loss in actively subscribed product quantity over the period.

Active subscribed product quantity (period end) − Active subscribed product quantity (day before period start).

Useful for spotting quiet revenue erosion from quantity decreases even when subscriber counts hold steady.

Metric

Definition

Formula

Notes

Non-subscribed customers

Customers who placed at least one order during the period and have never purchased a subscription before. Represents your pool of potential new subscribers.

Count of customers with ≥1 order in the period, excluding anyone with any prior subscriber history before the period started.

"Never subscribed" is checked strictly before the period's start date — a customer who subscribes mid-period can still appear here for an earlier order.

Subscriber acquisition rate

Percentage of non-subscribed customers who became subscribers during the period.

Acquired subscribers ÷ Non-subscribed customers × 100.

Shows "–" rather than 0% when there's no non-subscribed base to convert.

Acquired subscribers

Customers who became subscribers for the first time during the period (true first-time subscribers only).

Count of customers whose first-ever subscription started within the period.

Excludes reactivations and resumes — those are only counted in the broader Subscriber additions metric on the Overview tab.

0-day subscriber churn

Newly acquired subscribers who cancelled on the same day they subscribed, where the acquisition itself also happened in the selected period.

Count of subscribers where both acquisition date and churn date fall in the period, with a same-day gap.

Not the same calculation as 0-day churn contribution in Cancellation analytics — that one only requires the churn date to be in-period, not the acquisition date. The two aren't directly comparable.

Acquisition rate trend

How subscriber acquisition rate has changed over time.

Per time bucket: Acquired subscribers ÷ Non-subscribed customers × 100.

Trend view of Subscriber acquisition rate, shown daily/weekly/monthly.

Subscriber acquisition by product

Top 10 products driving new subscriber sign-ups, viewable by quantity or by revenue.

Per product: quantity and revenue from each newly acquired subscriber's first order.

Attribution is based on a subscriber's very first subscription order — later cross-sells to the same subscriber aren't included.

Acquisition by selling plan and delivery interval

New subscriber acquisition broken down by selling plan and delivery frequency combination.

Per plan/frequency pair: count of newly acquired subscribers attributable to that combination.

Useful for spotting which plan/frequency combos convert best at first purchase.

Metric

Definition

Formula

Notes

Subscriber vs Non-subscriber revenue

Total revenue split between customers who were subscribers (active/paused) vs. one-time buyers, based on the customer's status on the order date.

Subscriber revenue = revenue from orders placed while the customer was Active/Paused; Non-subscriber revenue = Total − Subscriber revenue.

This is a customer-level split (was the buyer a subscriber that day?), different from the order-level split below. Includes shipping and taxes, net of refunds.

Overall revenue

Total revenue from all orders in the period, inclusive of discounts, shipping, and taxes, net of refunds.

Sum of current order total across all orders.

Shown in both the Subscriber vs Non-subscriber panel and the Subscription vs Non-subscription panel — same figure.

Subscription vs Non-subscription revenue

Revenue, order count, and AOV compared between orders containing at least one subscription item vs. orders with only one-time items.

Subscription revenue = revenue from orders with ≥1 subscription line item; AOV = revenue ÷ order count for each side.

Order-level split (does the order contain a subscription item?), not customer-level. A subscriber's one-time-only order counts on the non-subscription side. Full cart value counts toward Subscription revenue even if most items were one-time.

Subscription revenue %

Share of overall revenue that comes from subscription orders.

Subscription revenue ÷ Overall revenue × 100.

Quick health check for how reliant the business is on recurring vs. one-time purchases.

Recurring vs Checkout subscription revenue

Revenue, order count, and AOV compared between recurring orders (billed by Loop) and checkout orders (the Shopify checkout order that first creates a subscription).

Checkout revenue = revenue from orders that created a subscription; Recurring revenue = Subscription revenue − Checkout revenue.

Helps separate "new subscription starts" value from "ongoing subscription" value within total subscription revenue.

Metric

Definition

Formula

Notes

Subscriber vs Non-subscriber LTV

Cumulative revenue per customer, tracked from acquisition date to today, compared between subscribers and non-subscribers.

Cumulative revenue ÷ customer count for each cohort, tracked forward from acquisition.

Segments subscribers further by which order number triggered their first subscription (Order #1, #2, #3, #3+).

Customer distribution

How many customers fall into each LTV segment (non-subscribed, or subscribed on order #1, #2, #3, or #3+).

Count of customers per segment.

Pairs with LTV breakdown to show whether early or late converters make up more of your base.

LTV breakdown

Detailed metrics per LTV segment: customer count, total orders, total revenue, orders per customer, and overall LTV.

Per segment: LTV = Total revenue ÷ Customer count; Orders per customer = Total orders ÷ Customer count.

The segment an order counts toward is based on how many orders a customer placed before their first subscription started.

LTV trend

How LTV has evolved over time for each segment, including non-subscribers.

Per time bucket, per segment: cumulative revenue to date ÷ fixed cohort customer count.

The chart's date range always extends to today regardless of the selected end date, so it keeps growing even for a fixed historical acquisition window.

Metric

Definition

Formula

Notes

Total attempted

Total value of all subscription payment attempts in the last 90 days — your baseline for billing volume.

Sum of the discounted order value for the latest billing attempt per order.

Counts one (the latest) attempt per order, not every individual retry.

Success (overall)

Percentage of payment attempts that were ultimately successful in the last 90 days, across all attempts including recovered dunning payments.

Realized revenue ÷ Total attempted × 100.

Broader than Success rate below — this includes recoveries. Don't use the two interchangeably despite the similar name.

Recovered

Revenue recaptured by Loop's retry engine after an initial payment failure.

Sum of revenue from successful dunning attempts.

Revenue that would otherwise have been lost without automated recovery.

Under recovery

Value of payments currently mid-retry, with attempts remaining and not yet resolved.

Sum of failed attempts with retries left, excluding any subscription that was cancelled/paused/expired/skipped before retries finished.

Still "in play" — may convert to Recovered or Lost depending on outcome.

Lost (overall)

Revenue that could not be recovered after all retries were exhausted, including skipped orders and subscriptions paused/cancelled/expired after a failed payment.

Sum of failed attempts with zero retries remaining.

Calculated across all cohorts — first attempt, first recovery cycle, and subsequent recovery cycles combined. Reflects permanent revenue leakage.

Success rate (first attempt)

Percentage of orders billed successfully on the very first attempt, excluding anything already in a dunning cycle.

First-attempt successful revenue ÷ First-attempt total revenue × 100.

The cleanest signal of card/checkout health, since no prior failure influences the outcome.

Backup attempt rate

Percentage of first-attempt failures where a backup payment method was also tried.

Backup-method attempts ÷ Total first-attempt failures × 100.

A higher rate means more customers have a fallback card on file, improving same-day recovery odds.

Backup recovered

Orders/revenue collected using a customer's backup payment method on the first billing attempt.

Sum of successful attempts made via a backup payment method.

Requires the merchant's backup-payment-on-dunning setting to be enabled to generate volume.

Recovery rate — First recovery cycle

Percentage of eligible payments recovered during the first recovery cycle (retries on the same order after the original failure).

Recovered ÷ (Attempted − Under recovery) × 100.

Excludes still-pending retries so the rate reflects only closed outcomes.

Lost — First recovery cycle

Orders/revenue that exited the first recovery cycle without being collected.

Sum of first-cycle attempts that ran out of retries or were abandoned via pause/cancel/expire/skip.

Shows revenue leakage immediately after first-cycle retries are exhausted.

Recovery rate — Subsequent recovery cycle

Percentage of eligible payments recovered across all retry cycles after the first (i.e. on later subscription orders).

Recovered ÷ (Attempted − Under recovery) × 100, scoped to 2nd+ cycles.

Captures incremental recovery beyond the first cycle — a strong first-cycle rate paired with a weak subsequent rate suggests prolonged dunning isn't adding much value.

Lost — Subsequent recovery cycle

Orders/revenue that couldn't be recovered even after extended retry attempts across later cycles.

Same "lost" classification as the first cycle, scoped to 2nd+ cycles.

Long-tail counterpart to Lost — First recovery cycle.

Metric

Definition

Formula

Notes

Recovery contribution split

Share of recovered revenue driven by automatic retries vs. customer-initiated card updates.

Recovered-by-card-update ÷ Total recovered × 100 (and the retry equivalent).

Also available as a trend over time. Card-update detection depends on the payment method being flagged as updated after the last successful charge — updates made through channels that don't set this flag get misattributed to "retry."

Failure reason wise recovery contribution

Recovery performance broken down by the reason a payment failed, split by retry vs. card update.

Per failure reason: Attempts, Recovered, Recovery rate = Recovered ÷ Attempts × 100.

Failure reasons are the raw processor error text — near-duplicate wording from different gateways can appear as separate rows rather than one merged category.

Metric

Definition

Formula

Notes

Subscriber churn rate

Percentage of subscribers who cancelled all of their subscriptions during the period.

Subscribers churned ÷ [(Active + Paused subscribers at period end) + Subscribers churned] × 100.

The active/paused base is a snapshot taken exactly at period end, not an average — a volatile period (lots of adds/pauses right before the cutoff) can shift the rate independent of actual churn behavior.

Subscribers lost

Absolute number of subscribers who cancelled all of their subscriptions during the period.

Count of distinct customers transitioning to fully churned status in the period.

If the same subscriber churns, reactivates, and churns again on the same calendar day, they're only counted once for that day.

0-day churn contribution

Percentage of churned subscribers who cancelled the same day they subscribed, relative to total churn in the period.

0-day churned subscribers ÷ Total churned subscribers × 100.

Compares calendar dates rather than a strict 24-hour window, so the actual elapsed time can range from minutes up to just under 48 hours. High values usually signal checkout-expectation mismatches or low-quality signups.

Upcoming order contribution

Percentage of churned subscribers who cancelled right after receiving an upcoming-order reminder, relative to total churn in the period.

Churned subscribers correlated with an upcoming-order reminder ÷ Total churned subscribers × 100.

Excludes subscriptions that churned due to a failed payment (dunning) — this isolates "surprise charge" driven cancellations. It's a timing correlation, not proof of causation.

Subscription cancellation rate

Percentage of individual subscription contracts cancelled during the period.

Subscriptions cancelled ÷ [(Active + Paused subscriptions at period end) + Subscriptions cancelled] × 100.

Contract-level, not customer-level — a customer with 3 subscriptions who cancels 1 contributes to this rate but doesn't count toward Subscriber churn rate. Excludes naturally expired (prepaid/fixed-term) contracts from the base.

Subscriptions cancelled

Absolute number of subscription contracts cancelled during the period.

Count of distinct contracts transitioning to Cancelled status in the period.

A single customer cancelling 2 subscriptions in one day counts as 2 here but only 1 toward Subscribers lost (if those were their last subscriptions).

Orders before cancellation

Average number of completed orders a subscription had before it was cancelled.

Sum of completed orders across all cancelled subscriptions ÷ Subscriptions cancelled.

Measures lifespan in order count, not calendar time, so it doesn't translate directly across different delivery frequencies. As an average, a few long-tenured cancellations can skew it upward.

MRR lost via cancellations

Monthly recurring revenue lost due to cancellations during the period.

Sum of recurring line-item MRR (at time of cancellation) across all cancelled subscriptions.

Captures MRR at its last known value before cancellation — doesn't net out a cancel-then-reactivate cycle that happened within the same window. Excludes shipping, taxes, and one-time charges.

Top 10 products by churned MRR

The 10 products contributing the most to lost MRR, alongside their churn rate.

Per product: Cancelled MRR ÷ (Active + Paused + Cancelled MRR) × 100, ranked by total MRR.

Only products currently on an active or paused subscription are eligible to rank — a product with cancellations but zero active/paused presence today won't appear.

Order wise cancellations

Breakdown of cancellations by the number of orders placed beforehand, to spot early-stage churn patterns.

Cancellations bucketed by completed order count.

⚠️ Not currently available as a standalone breakdown — only the average version (Orders before cancellation) is live. Treat this row as directional intent, not a confirmed chart.

Reason wise cancellations

Distribution of the reasons subscribers gave for cancelling. Also available as a trend over time.

Count of cancellations grouped by cancellation reason, most common first.

"Cancelled by customer"/"Cancelled by API" reasons are relabeled "No cancellation reason captured" since no specific reason was collected. Cancellations that bypass reason capture (e.g. some admin/API cancels) inflate this bucket without reflecting a real behavioral driver.

Channel wise cancellations

Where cancellations originated — e.g. self-serve customer portal vs. merchant admin vs. API.

Count and lost MRR grouped by cancellation source.

Admin-initiated cancellations may reflect merchant/support action rather than customer sentiment — don't read this the same way as self-serve cancellations.

Frequency wise cancellations

Cancellations broken down by delivery frequency (e.g. every 30/60/90 days), alongside current active and paused counts for context.

Active, Paused (snapshot), and Cancelled counts per delivery frequency; top 10 by volume.

Frequencies stored in different raw formats (e.g. "30 DAY" vs. "1 MONTH") are treated as distinct buckets rather than merged.

Selling plan wise cancellations

Cancellations broken down by selling plan, alongside current active and paused quantities for context.

Active, Paused (snapshot), and Cancelled quantities per selling plan; top 10 by volume.

Line items without a selling plan assigned are excluded from this view.

Metric

Definition

Formula

Notes

Save rate

Percentage of cancellation attempts that were successfully saved.

Saves ÷ Attempts × 100.

Dormant/abandoned cancellation sessions with no final outcome aren't counted as attempts either way. Can be filtered by ARL-covered vs. non-ARL regions.

Saved MRR

Monthly recurring revenue retained through the cancellation flow during the period.

Sum of discounted recurring MRR (normalized monthly) for every saved subscription.

Reflects MRR at the moment of the save — excludes one-time add-ons, shipping, and taxes.

Additional revenue generated

Revenue earned from subscriptions after they were saved via the cancellation flow.

Sum of line-item revenue from successful orders placed after the save event.

Only counts successful orders; excludes shipping, taxes, and orders placed before the save.

Additional orders generated

Number (and average per subscription) of orders placed on subscriptions after being saved.

Count of successful post-save orders; average = that count ÷ subscriptions saved in the period.

Only successful orders that occurred after the save event are counted.

Save rate distribution

Save rate and outcomes broken down by each step of the cancellation funnel — Benefits, Reason-based treatments, Offers, and Winback offers. Also available as a trend over time.

Per funnel step: Attempts, Saves, Save rate, Additional revenue, Additional orders.

"Winback offers" saves come primarily from post-cancellation win-back flows, so they can show activity even with low attempt volume in this view.

Benefit page wise saves

Save effectiveness of each individual benefit page shown during cancellation. Also available as a trend over time.

Per benefit page: Attempts, Saves, Save rate, Additional revenue, Additional orders.

Only sessions attributable to a specific benefit page are counted; generic/unattributed sessions are excluded.

Reason wise saves

Save effectiveness for each cancellation reason a subscriber selects.

Per reason: Attempts, Saves, Save rate, Additional revenue, Additional orders.

Feeds directly into Reason specific treatment saves and Reason specific offer saves below.

Reason specific treatment saves

Which retention treatments (offers, delay, pause, swap, etc.) worked best for a given cancellation reason.

Count of saved sessions grouped by treatment type, optionally filtered to one reason.

Helps decide which treatment to prioritize per reason rather than using a one-size-fits-all flow.

Reason specific offer saves

Distribution of which specific offers subscribers accepted for a given cancellation reason.

Count of saved sessions grouped by offer, restricted to sessions that reached and accepted an offer.

Only counts sessions that actually reached the Offer step — sessions saved via a different treatment (e.g. pause) aren't included here.

Offer wise saves

Save effectiveness of each individual offer. Also available as a trend over time.

Per offer: Attempts, Saves, Save rate, Additional revenue, Additional orders.

Trend view also includes win-back offer saves sourced from the post-cancellation Instant Reactivation flow.