ISP-vertical operating system  /  Building since 2014

The operating system for ISPs who build their own future.

One database, one auth layer, one deploy. Billing, network operations, inventory, and support sit on the same data spine. Self-host on your hardware or run it as SaaS: your data, your jurisdiction, your choice.

Multi-tenant Multi-currency Multi-region Self-host or SaaS AI-native In production since 2014
The Problem

ISPs are fed off-the-shelf ERPs that don't know what an OLT is.

The category has been served by adapted-from-elsewhere software for two decades. Generic CRMs bolted to billing engines bolted to ticket systems, glued by sync layers, sold by sales teams who learned about ONUs in their pre-call brief.

The integrators promise unified, then deliver duct tape. The dashboards promise insight, then deliver the same nine widgets every other vertical gets. The roadmap promises industry focus, then ships features for retail.

ISPCQ was built by people who run an ISP, for people who run an ISP. Every module maps to operational reality: cabling crews, OLT firmware, BGP sessions, splitter trees, customer-IP reverse lookups, prepaid-credit reconciliation across multiple currencies and tax regimes.

This is what the surfaces look like when the engineers are also the operators.

Built in production, not in a boardroom

Two tenants. Two countries. Same code.

ISPCQ wasn't designed in a strategy deck. It runs in production today across two geographies, two currency regimes, and two regulatory environments, on real subscribers, real money, real regulators, all from a single codebase, deployed together. Every architectural claim on this page is operational reality, not a roadmap promise.

Two tenants, one codebase
South Africa Bulgaria

Two ISPs running in production today. Live customers, live billing, live regulators. Not a sandbox.

Countries
South Africa · Bulgaria
Currencies
ZAR · EUR (BG migrated from BGN on 2026-01-01)
Tenancy model
Single mono-repo. No per-tenant forks. No drift.
Deployment
Self-host both, or run either as SaaS. Same architecture.
Cadence
Same code, both tenants, same day.
Sovereignty  /  The philosophy

Your data. Your jurisdiction. Your deployment model.

Sovereignty is not about owning the source code. It is about your data being yours, exportable, never trapped, and your deployment being your call. Two ways to run ISPCQ; same architecture either side of the boundary.

01
Deploy anywhere.
Self-host on your hardware, your VPC, your region. Or run it as SaaS while you build the team to bring it in-house. Same architecture either side of the boundary, same data model, same migration path.
02
Multi-tenant by design.
One codebase; multiple ISP brands across jurisdictions, currencies, and regulators; no forking. Two tenants live across two countries today; more in flight. Add a tenant brand, not another instance.
03
Built by ISPs for ISPs.
Every feature exists because an operator asked for it. The roadmap is the production backlog. There is no committee writing requirements that haven't been pulled at 2am during a fibre cut.
04
Your data, exportable.
Customer records, billing history, network telemetry, support transcripts. Yours, queryable, leaving with you if you switch deployment models or operators. Sovereignty is portability, not lock-in.
Core modules

One system. Ten domains. No bolted-together CRM.

Stop stitching together spreadsheets and vendor EMS tools. Run your entire ISP from one operating layer.

01
Multi-currency / multi-rail

Dual billing engines with rule-based pricing, automated bank reconciliation, mobile money collection, and cashbox management.

  • Smart Price Matrix & bundles
  • Regional rails: mobile money, SEPA, card, banking
  • Multi-currency, proven in production
02
Vendor-agnostic / OLT-native

Vendor-agnostic OLT management and TR-069 automation. See signal levels inside the customer ticket.

  • Multi-vendor OLT support
  • Live optical signal graphs
  • Environmental monitoring
03
OTDR / cable / splitter trees

Complete fiber network documentation. Cables, cores, splitters, ducts, manholes. OTDR integration and cable-fault impact analysis.

  • OTDR measurement import & analysis
  • Service path tracing (OLT to ONT)
  • Cable-fault impact lists
04
Configurable workflows / kanban

Configurable status workflows with For-Review sign-off, chain-of-command close, kanban board, dynamic dashboard views, KPI metrics, and automatic escalation.

  • Kanban with drag & drop
  • Outage quick-log + wiredown cluster-detect
  • Auto-escalation + chain-of-command transitions
05
Serial-tracked / multi-warehouse

Track every serial number from GRN to customer premise. Multi-warehouse and vehicle-stock tracking.

  • AI invoice scanning
  • Barcode generation
  • RMA & warranty tracking
06
Biometric / leave / payslips

Attendance tracking with fingerprint import, leave management with auto-login blocking, shift scheduling, and payslips.

  • Biometric device import
  • Leave auto-blocks login
  • Working-days engine
07
voyage-4-nano / sqlite-vec

Built around AI from the ground up, not bolted on. The assistant reads the same database your team reads. Customer summaries, note categorization, document verification, after-hours triage.

  • Customer intelligence briefs
  • Auto-categorization
  • After-hours automated triage
08
2FA / self-service / reseller

Self-service portal with 2FA, invoice viewing, online payments, ONU reboot, Portal Badger self-repair, and reseller management.

  • Two-factor authentication
  • Portal Badger: one-click connection repair
  • Reseller portfolio view
09
PBX / 6 gateways / messaging

PBX phone-system integration, email with bounce handling, payment gateways, and messaging plugins.

  • PBX API (any capable PBX)
  • Six payment gateways
  • Email & messaging queue
10
ACL / audit / multi-tenant

Deployment sovereignty, role-based access control, 34 automated cron jobs, multi-tenant architecture.

  • Granular ACL with inheritance
  • Full audit trails
  • Multi-region configuration

Comparing ISPCQ to Splynx, Sonar, Powercode, UISP, or your current platform?  Read the comparison brief →

Aelita  /  Built-in, not bolted-on

The AI assistant that already runs the platform.

Aelita reads the same data spine that operators do: customer history, network telemetry, billing ledgers, support notes, OLT firmware logs. She is part of the operating layer, not a chatbot grafted on top.

Built on voyage-4-nano embeddings against a sqlite-vec store. Live in production since December 2025 (v3) inside a platform that has been shipping since 2014.

Embeddings
voyage-4-nano
Vector store
sqlite-vec
Surface
In-app, every module
Live since
v3 · Dec 2025
Now at
v9, Phase E vault
Get a demo

See it on your own data.

Bring an export of your billing, network, or support data. We'll show you the same operational view your team would see on day 14, with your customers, your tariffs, your OLT topology.

Demos run roughly 45 minutes. Self-hosted or SaaS, your call. We don't pitch; we show.

Direct

Comparison brief
Splynx, Sonar, UISP, others
Why ISPCQ
The full thesis