New · Instant Pix in Brazil
sparkio
Start now

Platform · Analytics

See why payments pass or fail.

Approval rate, decline reasons, funnel and latency for every market and method — read straight from the ledger that authorizes and settles your money. No tag to install, no export to reconcile.

Start nowTalk to sales
Analytics · OverviewSep 22 – Oct 21
Date range
Gross volume
$121,840.20
1,436 payments
Approval rate
92.4%
1,436 of 1,554 attempts
Average ticket
$84.85
per approved payment
Recovered
$3,212.40
37 payments retried

Approval rate7-day rolling

Use the arrow keys to move between points; Home and End jump to the first and last.

Simulated · Northlight demo data

Approval rate

See where approvals slip.

Approval split by market and by payment method, on the same window and the same definition, against your account average. A weak market or a slow-paying voucher stands out at a glance.

  • By market, method, card brand and issuer
  • One definition: a payment retried on a second route counts once
  • Every bar compared with your account average
  • Open any bar to see the payments behind it
How a payment is authorized
Approval rateLast 30 days
Break down by
Sort by
Account · 92.4% · 1,436 of 1,554 attempts
Approval rate by market, last 30 days (simulated Northlight data)
MarketAttemptsApprovedApproval
United States69265194.1%
United Kingdom15214293.4%
Eurozone20218792.6%
Brazil24422491.8%
Mexico17715788.7%
India877586.2%

Decline reasons

Every decline, with a next step.

Issuer responses are grouped into reasons you can act on. Next to each one, SparkIO suggests the fix it can run for you — a retry schedule, a second route, a 3‑D Secure change or a card update — and you switch it on.

  • The raw issuer code stays on every payment
  • Soft and hard declines told apart
  • One suggested fix per reason, applied with one switch
  • Every change logged with who made it
Read about smart routing
Declines118 declined · last 30 days

Suggested fix

34.7% of declines

Retry schedule

Soft decline

Saved cards and renewals retry on day 3, 5 and 7 after a failed charge, at the hour the card last succeeded. One-off checkouts never retry.

on 51 insufficient_funds

retry day 3 · 5 · 7

scope saved_cards, renewals

Apply to new payments

≈ $3,479 declined in 30 days

Simulated · nothing changes on a real account

Decline reasons, last 30 days (simulated Northlight data)
ReasonCodeDeclinesSuggested fix
Insufficient funds5141Retry schedule
Do not honor0529Route B, every market
Suspected fraud59173‑D Secure before risky payments
Expired card5412Card updater
3-D Secure abandoned3DS11Fewer 3‑D Secure challenges
Other—8Review in events

Checkout funnel

From checkout view to money in the bank.

One funnel from the moment a shopper opens checkout to the moment the money lands in your balance. The gateway and the acquirer are the same account, so no step falls between two tools.

  • Views, attempts, approvals and settlement in one place
  • Step conversion and drop-off at every stage
  • Async payments shown as pending, never lost
  • Split by market, method or device
How settlement works
Checkout funnelLast 30 days
Show
2,180 → 1,409

  1. 71.3% started a payment · 626 left

  2. 92.4% approved · 118 declined

  3. 98.1% settled · 27 pending

64.6% of checkout views ended as money in the balance
Checkout funnel, last 30 days (simulated Northlight data)
StageCountShare of viewsStep conversion
Checkout views2,180100.0%—
Payment attempts1,55471.3%71.3%
Approved1,43665.9%92.4%
Settled to balance1,40964.6%98.1%

Latency & routing

Every route, timed in milliseconds.

Median and slowest authorization times for each route, next to the payments its fallback recovered. When a route slows down, it shows here first — and the routing rule is one click away.

  • p50 and p95 per route and per market
  • Payments recovered by fallback, in count and value
  • Routes ranked by speed or by the tail
  • Routing rules linked from every row
Define routing
RoutesAuthorization time · last 30 days
p50 authorization
212 ms
median
p95 authorization
480 ms
95th percentile
Recovered
37
$3,212.40
Rank by
37 recovered: 31 on Route B fallback, 6 by scheduled retry
Authorization time by route, last 30 days (simulated)
RouteScopep50p95Recovered
Pix (instant rail)Pix · Brazil168 ms352 ms0
Route A (acquirer_1)Cards & wallets · US, GB196 ms452 ms0
Route C (acquirer_3)Cards & wallets · Eurozone224 ms498 ms0
Route D (acquirer_4)Cards · BR, MX, IN238 ms541 ms0
Route B (acquirer_2)Fallback · US, GB241 ms518 ms31
UPI (instant rail)UPI · India291 ms664 ms0

Reports & exports

Take the numbers where you work.

Download any view as CSV, email it on a schedule, sync it to your data warehouse or pull it from the API. Every export uses the same definitions as the dashboard, so the numbers match wherever they land.

  • CSV of any view, any range
  • Scheduled emails to anyone on your team
  • Warehouse sync every hour
  • A Reports API with the dashboard's filters
Read the API reference
ReportsApproval by market · last 30 days
Export as

northlight-approval-by-market-2026-10-21.csv

6 rows · 0.2 KB

CSV preview: approval by market, last 30 days
marketattemptsapprovedapproval_rate
US69265194.1
BR24422491.8
EU20218792.6
GB15214293.4
MX17715788.7
IN877586.2
Simulated · Northlight demo data

Ask Spark

Ask Spark why approvals moved.

Spark reads the same analytics you do. Ask in plain words and get the answer with the numbers behind it, plus the change it would make. Nothing changes until you approve it.

  • Answers show the chart and the payments behind them
  • Suggestions stay drafts until someone approves them
  • Read-only by default; roles decide what it can change
  • In the dashboard, or in your own AI over MCP
Meet Spark
SparkNorthlight · Analytics

Why is Mexico's approval rate below the rest?

Spark

Approval rate · last 30 days

  • Mexico · cards & wallets94.5%
  • Account92.4%
  • Mexico · OXXO74.0%

Suggestion

Email a reminder 24 hours before each OXXO voucher expires.

Sources · approval by market, by method · Sep 22 – Oct 21

Simulated · answers are computed from the demo data

Mexico approved 88.7% of 177 payment attempts in the last 30 days, 3.7 points under your account's 92.4%. The gap is OXXO. Vouchers are 28% of Mexico's attempts and 74.0% of them get paid. Cards and wallets in Mexico approve at 94.5%, above your average. Suggestion: Email a reminder 24 hours before each OXXO voucher expires.

Also in analytics

For the questions that come next.

Smaller tools that turn a chart into an answer.

  • Real-time events

    Every authorization, decline, refund and payout lands as an event seconds after it happens, filterable by any field.

  • Custom segments

    Save a slice — first-time buyers in Brazil paying with Pix, say — and reuse it in every chart, alert and export.

  • Cohorts

    Group customers by the month of their first payment and follow repeat purchases and renewals over time.

  • Alerts

    An email or a webhook when approval drops, a decline code spikes or a route slows past a threshold you set.

  • Multi-currency normalization

    Totals in your reporting currency at each payment's capture-day rate, with the original amount kept on every row.

  • Roles

    Finance, support and marketing each see the views they need; amounts and customer details follow each role.

Questions

Questions about analytics.

Where does the data come from?
From the same ledger that authorizes, captures and settles your payments. SparkIO is both the gateway and the acquirer, so there is no tag to install and no second system to reconcile: every chart is built from the payments themselves.
How is approval rate calculated?
Approved payments divided by payment attempts in the selected range. A payment declined on one route and approved on a second counts once, as approved, and also appears in Recovered. Charts never sample: every payment in the range is counted.
How fresh are the numbers?
Authorizations and declines appear within seconds. Settlement and payout figures update as the money moves, so a payment shows as approved first and settled when it lands in your balance.
How are different currencies compared?
Totals are converted to your reporting currency at the rate of the day each payment was captured. The original amount and currency stay on every row, in every export and in the API.
Can I get the raw data out?
Yes. Any view downloads as CSV, reports can be emailed on a schedule, the warehouse sync copies payments, declines and settlements every hour, and the Reports API takes the same filters as the dashboard.
Who on my team can see it?
Owners decide. Roles control which views, markets and amounts each teammate sees and whether they can export, and every export is written to the audit log.

Your first payment is your first data point.

Analytics is on from the first transaction, sandbox included. Nothing to install, nothing to connect.

Start nowTalk to sales