If the outcome does not move, you should not be paying for the platform.
That sentence is easy to write and hard to sign, which is why most of this category does not sign it. Making it real needs three unglamorous things: a baseline drawn from your own reporting and agreed before anything runs, a clear line between what we control and what you do, and a remedy written in one sentence rather than a paragraph of conditions.
A guarantee is only as good as the definitions underneath it.
Your historical performance on your book, from your own reporting, agreed in writing before a single call is placed. Not our estimate, not an industry benchmark, and not a figure derived after the fact. This is the part everyone skips and it is the only part that makes the rest enforceable. If your reporting cannot produce a defensible baseline, that is worth discovering in week one rather than in the argument at the end.
One number, chosen by you, that your own board already reads. Recovery rate on a defined bucket, renewal conversion inside a stated window, save rate on a churn cohort. Not automation rate, not containment, not a satisfaction score. We do not publish a fixed list of guaranteed metrics, because the metric that matters is yours. Anything we could improve without improving your business is disqualified by definition.
What we control and what you control, written down before we start. We carry execution: reach, consistency, coverage, the quality of the conversation, the handoff into your floor. You carry the inputs: data quality, the access we need, your own policy, and any constraint your compliance team introduces. A guarantee that quietly covers your inputs is a guarantee we would break, and then neither of us has anything.
What happens if we miss, in one sentence, decided before you sign. If we miss, you do not pay for the miss. Our commercials are a fixed fee plus an overage tied to the outcomes we agree; when those outcomes are not met, the overage is waived. Not a service credit, not a discount after the fact — the downside is priced in before we start. The test of any commercial promise is whether the remedy can be read in a single breath by someone who is annoyed.
What we would stand behind, and what we would refuse to.
- Reach and coverage. That the accounts you gave us were attempted, at the hours the rules allow, the number of times agreed.
- Consistency. That the rulebook, tone and disclosure were the same on every call, and that you can verify it on all of them rather than a sample.
- The handoff. That the calls needing a person reached one with the context attached, inside an agreed time.
- Movement on the outcome metric against the agreed baseline, on a book where the volume makes the result meaningful.
- An absolute recovery or renewal percentage. That depends on your book, your pricing and your customers' ability to pay. Anyone guaranteeing it is either not reading the contract or not planning to honour it.
- A result on a list we cannot reach. If a third of the numbers are wrong, no execution fixes it, and the honest first project is the data.
- An outcome where your policy changes mid-programme. A new settlement rule or a changed collections strategy resets the baseline. Fair to both sides.
- Anything on volume too small to measure. A few hundred calls cannot separate our effect from ordinary variance, and a guarantee on noise is theatre.
Four steps, and three of them happen before we start.
Agree the metric and pull the history
Your operations team produces the baseline from your own reporting. If the two of us cannot agree on what normal looks like, there is nothing to guarantee and we should stop here.
Write the boundary down
Which inputs are yours, which execution is ours, and what happens if a compliance constraint changes the volume. This is a short document and it prevents the long argument.
Fix the measurement window before anything runs
We baseline for a minimum of three months on high-volume flows before the guarantee takes effect. Long enough that variance does not decide the result, short enough that you are not committed for a year to find out. Small-batch calling never produces a stable baseline, so we do not guarantee against it.
Run it, and report against the baseline weekly
No surprises at the end. If the number is not moving by the midpoint, we will say so and tell you what we think is wrong, which is usually the list or the handoff rather than the model.
Is this real, or is it marketing?
It is real, and it is in contracts today. This page describes the shape; the exact numbers and the exposure are defined per engagement and sit in the contract, not on a web page. If we miss, you do not pay for the miss: the overage tied to the outcomes we agree is waived. If you want to test whether we mean it, ask for the baseline clause in writing before a pilot starts.
What stops you from setting an easy baseline?
You do. The baseline is your historical performance on your book, drawn from your own reporting, agreed by your operations team before anything runs. We do not get to choose the denominator, and if we tried, the people who would catch it are the ones sitting in the room.
What if the miss is our fault?
Then it is not covered, and the contract has to say so plainly or the guarantee is meaningless. A stale list, an integration your team could not prioritise, a compliance constraint that arrived mid-pilot, a change in your own collections policy. We carry execution risk, not the risk of the inputs changing underneath us.
Do you do this on every programme?
No. It needs a process with a measurable outcome, enough volume for the result to be statistically meaningful, and a baseline that actually exists in your reporting. Collections, renewals and retention usually qualify. A low-volume process with no history does not, and we would rather run a straightforward pilot there.
What is the remedy if you miss?
You do not pay for the miss. Our commercials are a fixed fee plus an overage tied to the outcomes we agree; when those outcomes are not met, the overage is waived. Not a service credit and not a discount after the fact — the downside is priced in before we start. One sentence in the contract rather than a paragraph with conditions. The specific exposure sits in the contract, defined per engagement.
The mechanism is live. So is the promise around it.
If we miss, you do not pay for the miss: our commercials are a fixed fee plus an overage tied to the outcomes we agree, and when those outcomes are not met the overage is waived. We do not publish a fixed list of guaranteed metrics, because the metric that matters is yours — a recovery rate on a bucket, a renewal rate inside a window, a save rate on a cohort, agreed with you before go-live. We baseline for a minimum of three months on high-volume flows before the guarantee takes effect; small-batch calling never produces a stable baseline, so we do not guarantee against it. The specific exposure sits in the contract, defined per engagement — not on a web page.