Hypothesis
Recovering the balance matters less than not losing the customer while doing it.
Desired result: Lower cost per resolved contact; shorter wait/handling time; higher consistent resolution and accessibility

Credit, collections and care teams
A customer is behind on payment. A blunt reminder in the wrong language damages a relationship that is still worth keeping.
A pilot's requirements are specified: the data, the authority and the consent design.
Seen by the customer · Voice, Kiswahili
From the record, not a run: what would reach them, on Voice, IVR, WhatsApp, SMS, in Kiswahili, English, Local language.
What happened
A payment becomes overdue or a promise-to-pay is missed.
A customer is behind on payment. A blunt reminder in the wrong language damages a relationship that is still worth keeping.
What it reads
Checks before anything is sent
Identity, vulnerability, contact policy, authority and affordability pass.
Every option considered, including doing nothing
No action is a valid outcome, and it is recorded with its reason. Nothing is computed on this page: the options are the record's own, and no run has decided anything.
The shape of the record, with no values. A reference appears only after a run is saved.
Hypothesis
Recovering the balance matters less than not losing the customer while doing it.
Desired result: Lower cost per resolved contact; shorter wait/handling time; higher consistent resolution and accessibility
Modelled not observed
No figure is modelled for this journey. The mechanism that could produce the result is stated instead.
Receive → disclose AI → authenticate → detect intent → retrieve/act → confirm → log → hand off when limits trigger → sample for QA
Observed
Nothing yet. Measurement begins in a pilot’s validate stage, on your systems, against a comparison agreed first.
Intent-level control or phased rollout; compare resolved contacts, repeat contacts and cost per resolved intent; human QA sample
Measured on: Containment; end-to-end resolution; transfer rate; first-contact resolution; average handling time; repeat contact; CSAT; abandonment; unsafe-response rate
The result wanted.
Lower cost per resolved contact; shorter wait/handling time; higher consistent resolution and accessibility
The mechanism that could produce it.
Receive → disclose AI → authenticate → detect intent → retrieve/act → confirm → log → hand off when limits trigger → sample for QA
The records kept to show it.
Audio/transcript; intent/confidence; retrieved source; tool-call response; handoff reason; resolution code; QA review; repeat-contact result
The cost, and the ways it can go wrong.
Costs: Implementation fee + platform/usage subscription per interaction/minute; optional managed conversation operations
If it fails: Apologise without inventing; do not transact on uncertainty; transfer with captured context; retain trace; alert operations on repeated failures
The measure that shows it helped.
Intent-level control or phased rollout; compare resolved contacts, repeat contacts and cost per resolved intent; human QA sample
Measured on: Containment; end-to-end resolution; transfer rate; first-contact resolution; average handling time; repeat contact; CSAT; abandonment; unsafe-response rate
Nothing has been observed for this product. Every line above is the record's own design intent; measurement begins in a pilot's validate stage, on your systems, with the comparison agreed first.
The business side
| Step | What the customer experiences | What the operator does |
|---|---|---|
| The momentThe moment that started it. | A payment becomes overdue or a promise-to-pay is missed. | Reads balance and payment history, previous arrangements and how they went, language and channel preference, hardship and vulnerability markers. |
| The decisionThe decision to be made. | Nothing reaches the customer yet. | Explains the balance and proposes a fair payment or hardship path. |
| The safeguardsThe conditions that stop it. | Still nothing. No action is sent until every check has passed. | Identity, vulnerability, contact policy, authority and affordability pass. |
| The actionThe action that reaches the customer. | Secure a promise, payment or named human review in local language. Reaches them on Voice, IVR, WhatsApp, SMS, in Kiswahili, English, Local language. | Recommends. The system that holds the right confirms, charges or provisions. |
| When it goes wrongRefusal, failure and recovery. | The customer is flagged as vulnerable partway through the conversation. | Automated collection stops, the case moves to a person, and the stop is recorded with the reason. |
| The resultThe change it made. | What changed for them is what is counted; nothing else is claimed. | Recovered account or resolved evidenced case. |
| The proofThe proof anyone can check. | Can be answered for, later, from the record. | Balance source, conversation, plan, consent, payment and resolution trail. |
Operator systems remain authoritative. HeuriTel reads these to propose an action; it does not decide on their behalf.
What goes wrong
The customer is flagged as vulnerable partway through the conversation.
What happens then
Automated collection stops, the case moves to a person, and the stop is recorded with the reason.
A case, a complaint or an overdue balance is handled with the customer's context, a clear disclosure and a human within reach.
where the balance, the plan and the payment sit
how the reminder is worded and which channel carries it
A pilot's requirements are specified: the data, the authority and the consent design.