# AI-GTM Maturity (4 Levels)

> Four maturity levels of a GTM function's AI adoption: foundation, multi-channel, orchestrated, AI-native. The levels cannot be skipped; the next sensible step is always exactly one level up.

- Canonical URL: https://www.convios.com/en/methods/ai-gtm-maturity-levels
- Language version: https://www.convios.com/de/methodik/ai-gtm-reifegrad
- Status: Convios practice framework
- Method library: https://www.convios.com/en/methods — Markdown: https://www.convios.com/en/methods.md

## Problem

Leadership teams want "AI in sales" and buy tools or agents before the foundations exist. On poor CRM data, automation only scales the error; an agent on disconnected point solutions works with partial data. Before any tooling decision, you need the classification: which level is the GTM function on, and what is its current constraint?

## Approach

1. Locate the GTM function on the four levels: foundation (selling by hand, first CRM), multi-channel (channels interlocking), orchestrated (signal-driven), AI-native (agents do the legwork, humans decide).
2. Name the typical constraint of the current level: data quality at level 1, tool sprawl at level 2, missing clean signals at level 3, creativity and judgment at level 4.
3. Define the next step as exactly one level up, never a jump to level 4.
4. Recommend or buy tools and agents only after this diagnosis.
5. Repeat the classification regularly; the level is a state, not a completion.

## Example

A typical case: a software company with a good product wants to introduce "AI agents in sales" because competitors advertise it. The classification shows level 2: channels are running, but some twenty poorly connected tools produce no consistent data flow. The agent rollout is postponed; instead the team consolidates the stack and builds signal-driven routing, the transition to level 3. The agent question then comes up again on a viable foundation.

## Limits

The model is a synthesis of established methods and, as a stage model, not independently validated empirically; it is a diagnostic heuristic, not a calibrated scale. It classifies the GTM function, not the AI maturity of the whole company. Level boundaries blur in practice; mixed states are the norm.

## Metric

The current level with its named constraint, plus progress on the level-plus-one measures, such as data quality or signal coverage.

## Sources

- Capability Maturity Model for Software, Version 1.1 (CMU/SEI-93-TR-024) — Software Engineering Institute, Carnegie Mellon University (Paulk, Curtis, Chrissis, Weber), 1993 · academic and scholarly literature · supports the underlying mechanism. Carries the principle of maturity levels as ordered plateaus of development and argues in a dedicated section why skipping a level is counterproductive; the four sales levels themselves do not appear there. (https://www.sei.cmu.edu/documents/1092/1993_005_001_16211.pdf)
- Primary source: Capability Maturity Model for Software, Version 1.1 (CMU/SEI-93-TR-024) (https://www.sei.cmu.edu/documents/1092/1993_005_001_16211.pdf)

## Related

- Method: [Maturity Gate](https://www.convios.com/en/methods/maturity-gate)
- Method: [AI-GTM Transformation Loop (DSAE)](https://www.convios.com/en/methods/ai-gtm-transformation-loop)
- Method: [GTM Stack Signal Routing](https://www.convios.com/en/methods/gtm-stack-signal-routing)