Hypothesis
A customer who is drifting away is cheaper to keep than to replace, and an offer sent over an unresolved fault costs more than it earns.
Decisioning and Next Best Action
Customer Growth & CVM
Help customers return to the right pack, not just the next promotion.
When a regular buying pattern changes, compare a relevant offer, a service check and no action before contacting the customer.
For CVM and commercial teams accountable for prepaid and postpaid value.
Available to try now
The whole journey, end to endRuns end to end
Runs end to end, configured for your organisation and market.
Open simulation
Verified work email required.
Against your own systems
Against your own systemsWorking session
Request a session A pilot prepares this decision on your data, systems and policy, with the comparison agreed first.
Regular data buyer: no pack for 6 weeks, 41% of the last one used, contacted 25 days ago.
The demonstration below runs on a generic sample. Tell us your market and it will use figures and a name that fit.
Configure a demonstrationThe decision
Should the operator offer a different pack, check whether the service is working, wait, or do nothing at all?
A customer who bought a data pack every month for a year has bought nothing for six weeks. The service works and the customer is silent.
One decision, from the signal to the record it leaves
- The signal
A regular buying pattern stops, and the gap passes the point where it stops looking like a pause.
- What it may read
- Recharge and pack purchase history
- Recent data usage against the pack they were buying
- Service quality at the locations they use
- Contact permission and when they were last contacted
- Open care or service cases
- What can refuse it
- Marketing consentHas this customer agreed to be contacted for offers, on this channel?Consent and preference system
- Contact frequencyWould this exceed how often the customer may be contacted?Campaign management system
- Fair valueDoes the proposed pack suit how much they actually use?Product catalogue and pricing
- Margin floorDoes the incentive leave the action worth taking?Commercial finance
- Open service caseIs there an unresolved fault or complaint on this account?Care and service management
- What is kept
- The inputs used, and how fresh each one was
- The eligibility rule and its version
- Every option considered, with the reason it was allowed or refused
- The option chosen, including when that was no action
- The policy and consent checks, and their results
- The named approver, where approval was required
- The execution request and the receipt that came back
- The comparison group the result will be measured against
The ranking, step by step
- 01Eligibility rules
- 02propensity/uplift or value score
- 03treatment constraints
- 04ranked action including no-action
- 05decision log
What it does
Three things it helps you decide or do, each visible in the run below.
Understand why the buying pattern changed
Recent usage, recharge and contact history are read before anything is proposed, so a customer who has gone quiet and a customer with a service fault are told apart.
Compare appropriate options and stop those that fail checks
A right-sized pack, a service check and no action are ranked together, and consent, contact frequency, fair value, margin and an open case can each refuse.
Keep the decision and explain how a pilot would measure its effect
Every run keeps its inputs, checks and outcome, and a pilot compares treated customers against a holdout before any value is claimed.
Run it
Recover a customer who has quietly stopped buying data packs
Move any input and the checks, the ranking and the decision are computed again.
The customer, as the records have them
Customer · ILL-NEU-0001 · 36 months · Regular data buyer
Unchanged, as the sample records have it.
Offer a pack that matches how they actually use data
Their usage fits a smaller pack than the one they were buying. A fair-value match is more likely to be taken than a discount on the pack they already stopped buying.
Small weekly pack · 190 · 1.5 GB for 7 days · 72 margin after the incentive
Every check permits it, so the ranking chose the option worth the most to both sides.
The checks, the options, the customer's outcome and the failure, row by row
The checks, and who owns each answer
| Check | Result | Why |
|---|---|---|
| Marketing consentConsent and preference system | Permits | Consented on SMS. |
| Contact frequencyCampaign management system | Permits | Last contacted 25 days ago; the floor is 14. |
| Fair valueProduct catalogue and pricing | Permits | Using 41% of the pack they were buying, so a smaller pack is the fair match. |
| Margin floorCommercial finance | Permits | 72 remains on Small weekly pack after the incentive. |
| Open service caseCare and service management | Permits | Every case on this account is resolved. |
The open-case check reads this market’s own records rather than a field on the customer, so it is not one of the controls.
The customer’s outcome
A smaller weekly pack matching your recent usage is available. Reply 1 to accept.
Reply 0 to stop receiving these offers.
English · SMSThe failure this market demonstrates
Usage or care data has not refreshed inside the window the decision requires, so eligibility cannot be established.
Until a market is configured, usage and care data have no agreed freshness window, and eligibility waits.
Remedy. Stale context returns no-action, with the source and its age named.
Handed to Data operations, with the source and freshness recorded.
Sample records, computed in your browser. Record set FX-NEUTRAL v3, scenario SCN-PACK-DROPOUT.
The improvement, and how it is measured
Modelled
A run on the sample records shows the decision, the checks it passed and the receipt the responsible system would return; incremental value is measured in the pilot.
Observed
Measured in a pilot’s validate stage, on your systems, against a comparison agreed first.
The result, the mechanism, the measurement, and the record’s five answers
- Desired result
- Incremental revenue or retention; lower contact waste; improved conversion with margin guardrails
- Mechanism
- Ingest → qualify → score → apply limits/consent → assign control → execute approved action → capture delivery and business outcome → monitor/retrain
- Measurement
Randomised holdout where feasible; intent-to-treat primary view; pre-declared denominator; guardrails for complaints, margin and opt-out
Measured on: Eligible population; treatment rate; conversion; incremental revenue; ARPU; churn; contact rate; margin; opt-out; decision latency
- 01Desired result
The result wanted.
Incremental revenue or retention; lower contact waste; improved conversion with margin guardrails
- 02Mechanism
The mechanism that could produce it.
Ingest → qualify → score → apply limits/consent → assign control → execute approved action → capture delivery and business outcome → monitor/retrain
- 03Evidence required
The records kept to show it.
Decision log; eligibility snapshot; price/order response; delivery receipt; control assignment; revenue/outcome ledger
- 04Costs and risks
The cost, and the ways it can go wrong.
Costs: Setup fee + annual product subscription; optional managed execution fee; optional verified-outcome fee with agreed baseline
If it fails: Fail closed on eligibility/consent/price; queue retryable events; do not duplicate orders; route exception to campaign operations
- 05Measurement approach
The measure that shows it helped.
Randomised holdout where feasible; intent-to-treat primary view; pre-declared denominator; guardrails for complaints, margin and opt-out
Measured on: Eligible population; treatment rate; conversion; incremental revenue; ARPU; churn; contact rate; margin; opt-out; decision latency
A saved run keeps the decision and its receipt: what was decided, which checks refused, and what the responsible system returned. It does not show incremental value. That needs the comparison above, over an agreed period, against a baseline finance has accepted.
The business side
- The change
- Accepted, provisioned save and incremental contribution.
- Running cost
- Setup fee + annual product subscription; optional managed execution fee; optional verified-outcome fee with agreed baseline
- What evidence supports it
- Cohort definition, model version, policy decision, assignment and outcome.
Technical detail
Inputs, decision logic, systems touched, records kept.
The four questions, field by field
What would we need from your systems, and how much history?
Attributes required
- customer_id
- account_type
- tenure_days
- active_products
- recharge_30d
- usage_30d
- revenue_30d
- last_contact
- consent_status
- outcome_label
Also useful, not required
- Network experience
- digital clickstream
- complaints
- location cohort
- household/account links
- partner purchases
8-12 weeks of customer, usage, revenue, product, contact and outcome history; stable customer key; one executable channel
How is the action chosen, and what stops it?
Actions it may choose between
- Offer
- service message
- reward
- channel change
- human follow-up
- suppress
- do nothing
Conditions that stop execution
- No consent
- ineligible product
- price/catalogue mismatch
- contact cap
- credit/fraud block
- channel unavailable
- control assignment
Fail closed on eligibility/consent/price; queue retryable events; do not duplicate orders; route exception to campaign operations
What does it connect to, and what stays authoritative?
Your systems involved
- CRM/CDP
- campaign manager
- product catalogue
- OCS
- order management
- billing
- consent registry
Interfaces
- Customer/profile API
- catalogue/eligibility API
- campaign API
- charging/order API
- consent API
- outcome event API
Channels
- SMS
- USSD
- app
- web
- outbound call
- agent desktop
Charging, billing, catalogue, consent and order systems remain authoritative; HeuriTel decides only within approved policy.
What is kept, and how would a claim be tested?
Evidence required
- Decision log
- eligibility snapshot
- price/order response
- delivery receipt
- control assignment
- revenue/outcome ledger
The measures
- Eligible population
- treatment rate
- conversion
- incremental revenue
- ARPU
- churn
- contact rate
- margin
- opt-out
- decision latency
Privacy and regulation
- Purpose limitation
- consent and suppression
- data minimisation
- retention
- explainability for material customer effects
- local marketing rules
Randomised holdout where feasible; intent-to-treat primary view; pre-declared denominator; guardrails for complaints, margin and opt-out
The 22 definitions underneath
22 in this product
- HeuriTel CVMDecision definitionHT-0001
Identify eligible records for cvm, choose from approved actions including no action, execute through the system of record, and measure the agreed business outcome.
/d/HT-0001 - HeuriTel Customer 360Decision definitionHT-0002
Identify eligible records for customer 360, choose from approved actions including no action, execute through the system of record, and measure the agreed business outcome.
/d/HT-0002 - HeuriTel Customer SegmentationDecision definitionHT-0003
Identify eligible records for customer segmentation, choose from approved actions including no action, execute through the system of record, and measure the agreed business outcome.
/d/HT-0003 - HeuriTel Micro-SegmentationDecision definitionHT-0004
Identify eligible records for micro-segmentation, choose from approved actions including no action, execute through the system of record, and measure the agreed business outcome.
/d/HT-0004 - HeuriTel Customer LifecycleDecision definitionHT-0005
Identify eligible records for customer lifecycle, choose from approved actions including no action, execute through the system of record, and measure the agreed business outcome.
/d/HT-0005 - HeuriTel Next Best ActionDecision definitionHT-0006
Rank eligible commercial or service actions for each customer, including do nothing, under consent, contact, margin and product rules.
/d/HT-0006 - HeuriTel Next Best OfferDecision definitionHT-0007
Rank only offers the customer can buy and the operator can fulfil, then capture order, margin and incremental conversion.
/d/HT-0007 - HeuriTel Churn PreventionDecision definitionHT-0008
Identify customers with a defined near-term churn risk, select only economically justified retention actions, and measure incremental saves against a holdout.
/d/HT-0008 - HeuriTel RetentionDecision definitionHT-0009
Identify eligible records for retention, choose from approved actions including no action, execute through the system of record, and measure the agreed business outcome.
/d/HT-0009 - HeuriTel Win-BackDecision definitionHT-0010
Identify eligible records for win-back, choose from approved actions including no action, execute through the system of record, and measure the agreed business outcome.
/d/HT-0010 - HeuriTel Dormant Customer ReactivationDecision definitionHT-0011
Identify eligible records for dormant customer reactivation, choose from approved actions including no action, execute through the system of record, and measure the agreed business outcome.
/d/HT-0011 - HeuriTel Customer Lifetime ValueDecision definitionHT-0012
Identify eligible records for customer lifetime value, choose from approved actions including no action, execute through the system of record, and measure the agreed business outcome.
/d/HT-0012 - HeuriTel ARPU GrowthDecision definitionHT-0013
Identify eligible records for arpu growth, choose from approved actions including no action, execute through the system of record, and measure the agreed business outcome.
/d/HT-0013 - HeuriTel UpsellDecision definitionHT-0014
Identify eligible records for upsell, choose from approved actions including no action, execute through the system of record, and measure the agreed business outcome.
/d/HT-0014 - HeuriTel Cross-SellDecision definitionHT-0015
Identify eligible records for cross-sell, choose from approved actions including no action, execute through the system of record, and measure the agreed business outcome.
/d/HT-0015 - HeuriTel Prepaid-to-Postpaid MigrationDecision definitionHT-0016
Identify eligible records for prepaid-to-postpaid migration, choose from approved actions including no action, execute through the system of record, and measure the agreed business outcome.
/d/HT-0016 - HeuriTel High-Value Customer ManagementDecision definitionHT-0017
Identify eligible records for high-value customer management, choose from approved actions including no action, execute through the system of record, and measure the agreed business outcome.
/d/HT-0017 - HeuriTel New Customer OnboardingDecision definitionHT-0018
Identify eligible records for new customer onboarding, choose from approved actions including no action, execute through the system of record, and measure the agreed business outcome.
/d/HT-0018 - HeuriTel Usage GrowthDecision definitionHT-0019
Identify eligible records for usage growth, choose from approved actions including no action, execute through the system of record, and measure the agreed business outcome.
/d/HT-0019 - HeuriTel Share-of-Wallet GrowthDecision definitionHT-0020
Identify eligible records for share-of-wallet growth, choose from approved actions including no action, execute through the system of record, and measure the agreed business outcome.
/d/HT-0020 - HeuriTel Recharge-to-Usage ConversionDecision definitionHT-0021
Use customer affordability, usage, renewal and margin data to improve recharge-to-usage conversion while controlling cannibalisation and measuring incremental value.
/d/HT-0021 - HeuriTel Customer Value MigrationDecision definitionHT-0022
Identify eligible records for customer value migration, choose from approved actions including no action, execute through the system of record, and measure the agreed business outcome.
/d/HT-0022
Implementing it
Four stages, and what has to be true before the next one starts.
12-16 weeks to pilot; 2-4 additional weeks per market after reusable interfaces exist.
01
Scope
One use case, one value unit and a denominator finance has agreed to. Without these there is nothing a later result can be compared against.
- Confirm value unit
- select use case
- baseline denominator
02
Connect
The attributes mapped from your systems, and the decision and outcome interfaces working in both directions.
- map attributes
- integrate decision and outcome APIs
03
Validate
Eligibility agreed, a dry run with nothing sent, then a controlled pilot with a holdout that is actually respected.
- build eligibility
- dry run
- controlled pilot
04
Operate
The live loop: decide, check, execute through your systems, capture what came back, and measure against the control.
- Ingest
- qualify
- score
- apply limits/consent
- assign control
- execute approved action
- capture delivery and business outcome
- monitor/retrain
The team, and the four states each assessed against your systems
Who does it. 1 telecom product lead · 1 CVM specialist · 1 data engineer · 1 ML engineer · 1 integration developer · 0.5 QA · 0.5 DevSecOps · 1 deployment coordinator
Readiness is assessed on your data, your integration and your authority.
Needs client data and integration review
The client’s data and integration position.Target product definition
What exists as running software.Not audited
Whether this product shows the entry anywhere.Not assessed
Where a named client has reached.