Act with urgency on the costliest payment errors
Not every payment decline costs the same. A gateway error code tells you a payment failed. It doesn't tell you whether that failure cost you $75 or $750, or whether it was the first error in the sequence or the one that finally lost you a customer. This recipe shows you how Chargebee Reveal cuts through the error noise so you can prioritize the glitches actually draining revenue.
Why error frequency is never the full story.
And what to look for instead.
The most common error is rarely the most expensive one. An error that fires 500 times on $20 transactions matters less than one that fires 80 times on $400 transactions. This recipe re-sorts your errors by what they truly cost, so the fix list matches the revenue at stake.
Not just "this error happened 2,039 times," but $282,997 sat behind it. Two numbers next to each other change which error you call urgent.
A customer can hit three different errors before giving up. Only the last one mattered. Track that one separately and act with precision.
Every error comes with a plain-English reason and a next step. No decoding gateway jargon before you can act on it.
From “we have a lot of declines” to a fix list ranked by revenue impact
If you already have Chargebee Reveal live, jump straight to Step 3. If not, follow the first two steps to get set up. In this recipe, you'll start with the shape of your errors, then move to what they cost, then to which one actually ends the sale, then to who it's happening to.
Connect your gateway API keys
Open Reveal and go to Sources in the left navigation. Add each processor you run today with a read-only key. One-time setup, no engineering work.
- Add each gateway from Sources › Add source using read-only keys
- Step-by-step connect guides: Stripe, Adyen, Braintree, PayPal, Cybersource, Shopify Payments
- Connect every gateway you run, an error that looks small blended can be huge on one account alone
Let your payment data populate
Once sources are connected, Reveal starts importing your transaction history. Full ingestion typically finishes within a day. Nothing to do on your end but wait for it to finish.
- Transaction and error data populates per gateway as it comes in
- Insights and Performance views unlock as data is ready
- Come back when the Error Distribution tab shows numbers instead of a loading state
Decode the overall error distribution
Go to Performance › Error Distribution. The donut chart breaks every decline down by gateway error code, as a share of total error volume. Hover any slice and you get the exact percentage and error count. Below it, currency-specific tables show the same breakdown scoped to USD, EUR, GBP, and every other currency you process in.
- Hover a slice for exact percentage and error count
- Scroll the currency tables to see if the same error dominates everywhere, or if it's currency-specific
- Click Currency Wise Error Distribution to flip the whole view to a per-currency donut layout
general_decline or insufficient_funds, which tells you little on its own. The specific, named errors further down the legend, fraud flags, network declines, do_not_honor, are usually where the fixable patterns live. Rank your errors by true dollar impact
Click into any error, or scroll down, and you land on Error Impact on Revenue. This is the same kind of donut, but sized by dollar volume instead of raw count. Set the dropdown to by: Average and you get the average transaction value sitting behind each error type. The table below lists every gateway error code with a plain-English reason, the total volume, and the count, and you can toggle between viewing by Count or by Volume in the filter panel.
- Set the dropdown to by: Average for average transaction value per error
- Toggle Count / Volume in the filter panel to change what the donut sizes itself by
- Read the Gateway Error Reasons column, it explains the code in plain language
Catch the error that lost you the sale
Open the same dropdown and pick by: Last Occurrence instead. This re-sorts the donut and table around a different question: not "which errors happen," but "which error was the final one before the customer gave up." A $500 order can hit a 3DS failure, retry into an insufficient funds error, and finally abandon on a network decline. All three happened. Only the network decline cost you the sale.
- by: Average tells you the typical dollar value at stake per error type
- by: Last Occurrence tells you which error is the true breaking point
- Compare both views, an error that's rarely "last" but frequently "first" points to a retry-timing fix, not a checkout fix
do_not_honor or a similar hard decline shows up disproportionately often as the last error but rarely as the first, that's your customer communication gap. Those customers hit a wall with no clear next step and quietly left. A retry nudge won't fix it. A payment method switch prompt might. Trace revenue leaks to specific regions, issuers, or card types
Open the Filter Panel in the top right and apply a breakdown: Customer Country, Card Type, Card Issuer, Issuing Country, Funding Type, Payment Method, Gateway Account, or 3DS flow. Every step above still applies, you're just scoping it. This is how "3DS failures average $300 per transaction" becomes "3DS failures average $300 per transaction, and they're concentrated in customers on Issuer X in Region Y."
- Filter to one segment at a time first, card type, then geography, then issuer
- Watch for an error that's unremarkable overall but severe in one segment
- Segment-specific problems usually need a segment-specific fix, not a blanket policy change
Two ways to slice the error data,
what to use when
You need visibility into both errors by average value and the final error that costs you a customer. Both live in the same Chargebee Reveal dropdown, and both matter. They just answer different questions.
The average transaction value behind each error type. Answers: "if I fix this error, what's a typical recovered sale worth?" High-value errors deserve engineering time even if they're less frequent.
Insufficient funds: avg $75/txn
The final error a customer hit before abandoning, across every retry attempt. Answers: "which failure is the true breaking point, not just one step along the way?"
How to turn error insights into a fix list
Combining Average Revenue Impact with Last Occurrence, plus a segment breakdown, gets you to a specific action plan around your payment errors. Here's what teams typically do with it.
Rank your fix backlog by dollar impact, not error count. A rare, expensive error can outrank a common, cheap one.
If a specific error carries a high average value, give it a more generous retry policy than low-value soft declines.
If failures concentrate in one geography or card type, the fix belongs in that flow specifically, not a global change.
If Card Issuer breakdown shows one issuer driving disproportionate do_not_honor volume, that's a routing conversation, not a retry fix.
Errors that show up heavily as "last occurrence" are silent losses. A proactive nudge at that exact failure point recovers sales a generic retry never would.
Find out which errors are actually costing you money
Connect your gateways and Reveal returns the full error picture in a day: what's failing, what it's worth, and which one is ending the sale. If you'd rather have the Reveal team run it for you, we do that too, no engineering setup on your side.
Already on Chargebee Billing? Reveal is provisioned within one business day. Reach out to your account manager. Not on Chargebee yet? Reveal works standalone across 40+ gateways, book a free full-stack payments audit.