Here's a scenario I've watched play out more than once. A founder signs up for a dunning tool, sets up a sequence, and three months later the CFO asks a simple question: "Is this thing working?"
The founder opens the dashboard. It shows a few charts. Emails sent, opens, clicks. Maybe a recovery number. Nothing that clearly answers whether the tool is paying for itself, or whether switching off tomorrow would cost the business real money.
That's a failure of the dashboard, not the founder. Payment recovery software is one of the few SaaS purchases where you can, and should, prove ROI down to the dollar. If your tool can't show you that, it's either not tracking the right things or it's hiding behind vanity metrics.
This post covers the metrics that actually matter, how to evaluate a tool's reporting, and what a useful ROI dashboard looks like in practice.
Why ROI dashboards are non-negotiable for dunning tools
Dunning sits in a rare category. Unlike brand marketing or developer tooling, the ROI is directly measurable. Every recovered payment is a specific dollar amount on a specific invoice. You can trace it back to a specific message in a specific sequence.
If a tool can't show you that chain of causation, it's either not instrumented properly or it's deliberately vague. Neither is acceptable when the whole point of the product is to recover revenue you'd otherwise lose.
A good ROI dashboard answers three questions at a glance:
- How much revenue did this tool recover this month?
- Would I have recovered that anyway without the tool?
- What did it cost me to recover it?
Everything else is supporting detail.
The five metrics that matter
1. Recovered revenue
The headline number. Total dollars recovered from failed payments within a defined window, usually the last 30 days and month to date.
This should be net of refunds and chargebacks. If a customer's card goes through after a dunning email but they refund a week later, that's not recovered revenue. A serious dashboard accounts for this. A weak one double-counts.
Realistic benchmark: a tool recovering 60 to 75% of recoverable failed payments is doing well. Anything above 80% is excellent. Below 40% means either the tool is underperforming or your failure mix is unusually hard (lots of hard declines, fraud blocks, or closed accounts).
2. Recovery rate
Recovered payments divided by total recoverable failed payments, shown as a percentage.
The subtle part is the denominator. "Recoverable" should exclude payments where the customer cancelled, the card was reported stolen, or the account closed. Including those in the denominator makes the tool look worse than it is and buries the real signal.
A good dashboard lets you see recovery rate broken down by failure reason. Expired cards should recover at 70 to 85%. Insufficient funds at 50 to 65%. Hard declines much lower. If those segments aren't visible, you can't diagnose what's working.
3. MRR saved
Recovered revenue is a point-in-time number. MRR saved is the recurring impact. If you recover a $99 monthly subscription, that's not $99 recovered, it's $99 per month of MRR preserved, assuming the customer stays.
This matters because the lifetime value of a recovered subscription is usually 10 to 20 times the single invoice. A tool showing recovered revenue without MRR saved is undercounting its own value.
A quick example. Say you recover $5,000 in failed invoices this month across 50 customers. Average subscription is $100. With typical SaaS churn, those 50 customers will generate roughly $5,000 per month for the next 10 to 15 months on average. MRR saved is $5,000. Annual revenue saved is $50,000 to $75,000.
4. Cost per recovery
Total cost of the tool (subscription plus any usage fees) divided by number of recovered payments.
This is the number that makes ROI conversations with finance easy. If you're paying $49 per month and recovered 80 payments, your cost per recovery is $0.61. If the average recovered payment is $85, your ROI is 139x on that line item.
Tools that charge a percentage of recovered revenue get expensive here, because cost per recovery scales with average payment size. Flat-fee tools stay predictable.
5. Time to first recovery
Median time between a payment failing and the recovery closing. Measured in hours.
Faster is better, but only up to a point. Sending the first retry email three seconds after failure looks aggressive. Retry timing matters because Stripe's own smart retries have a natural rhythm, and customers need time to notice the first SMS before you hit them with a second one.
A healthy dashboard shows time to first recovery as a distribution, not just an average. You want to see whether most recoveries happen in the first 24 hours (likely card issues) or the first 72 (likely funds issues) or the full 14 days (the stubborn ones).
Metrics that sound good but aren't useful
Some dashboards lean heavily on engagement metrics. Open rates, click rates, delivery rates. These are fine as diagnostics when something breaks, but they're not ROI metrics.
A 95% email open rate means nothing if nobody updates their card. A 15% open rate with a high conversion can be better than a 40% open rate with a low one. Don't let engagement theater distract from recovered revenue.
Same with "emails sent" or "sequences triggered." Those are activity metrics. They measure whether the tool is running, not whether it's working.
What a useful ROI dashboard looks like
Here's roughly what we think a dunning dashboard should show, in rough priority order:
- Recovered revenue this month, with comparison to last month
- Recovery rate (recovered / recoverable), with segment breakdown
- MRR saved this month, with projected annual impact
- Cost per recovery
- Time to first recovery distribution
- Top failure reasons in the current period
- Per-sequence performance (which sequences recover the most)
- Per-channel performance (email vs SMS conversion)
That's it. Everything else is nice to have. If the first five are missing or buried, the dashboard isn't doing its job.
How to evaluate a tool's reporting before you buy
When you're in a trial, run a few checks:
- Can you see a specific recovered payment and trace it back to the message that drove the update? If yes, the tool has proper attribution. If no, the recovery numbers are soft.
- Does the dashboard update in real time, or on a daily batch? Real time matters when you're testing sequence changes and want to see impact today, not tomorrow.
- Can you export raw data? You'll want to reconcile the tool's numbers against Stripe and your own MRR tracking.
- Does recovery rate exclude uncollectable payments from the denominator? If not, every number looks worse than reality.
- Is MRR saved calculated, or do you have to compute it yourself? If you have to compute it, the tool is understating its own value.
Where Rekko fits
We built Rekko's dashboard around recovered revenue as the anchor metric. Not emails sent, not opens, not clicks. The front page shows recovered revenue, recovery rate, MRR saved, and cost per recovery for the current month. Everything else is one click away.
Because Rekko is flat fee, cost per recovery stays predictable as you scale. You're not paying a percentage of recovered revenue, so the ROI math gets better the more the tool works.
Attribution is traced per message. You can click any recovered payment and see exactly which sequence step closed it, email or SMS. That makes it easy to tune your sequences and prove value to finance without manual reconciliation.
If you want to compare how different tools handle ROI reporting, we've put together pages on Churnkey, ProfitWell Retain, and Stunning.
The honest caveat
No dashboard makes a bad sequence good. If your messages are generic, your timing is off, or you're not using SMS at all, a beautiful ROI dashboard just shows you clearly how much money you're leaving on the table. The dashboard is the feedback loop, not the fix. The fix is better sequences, multi-channel coverage, and honest testing.
But you can't improve what you can't see. So start with the dashboard.
Try Rekko free
See what recovered revenue, recovery rate, and cost per recovery look like on your own failed payments.
- Real-time ROI dashboard
- Per-message attribution
- Email plus SMS sequences
- Flat monthly pricing
Start your 14-day free trial. No credit card required. Full details on pricing.