Platform Intelligence

Knowing how a platform works is not the same as understanding your position in it.

Platform Intelligence is the structured ability to assess where a company stands in the AI-era platform ecosystem - which forces are acting on it, which relationships are durable, and what the dynamics already in motion mean for its position over the next 12-24 months.

Talk to AGG about the Framework

Platform Gravity

The force that pulls every company toward the platforms around it.

Platform gravity is the operating reality of every company that builds on, sells through or competes alongside a major platform, including AWS, Microsoft, Google Cloud, Salesforce, and AI foundation model providers including Anthropic and OpenAI. It operates whether or not the company has named it and it determines outcomes well before a startups financial data reflects it.

Gravity is continuous and mostly invisible

Today's key players are making platform roadmap decisions months before they become public.

CI/CD and continuous, progressive delivery is the order of the day. When critical platforms make changes, by the time it arrives, the window for an appropriate business or technical response has often already closed for startup B2B technology products in the path.

AGG knows the signals and how to find them.

It is happening from multiple layers simultaneously

Most companies have a primary platform relationship that takes their attention.

A company watching one relationship can miss a competitive convergence forming from an entirely different layer. Dependencies across these layers is critical to understand and continuously monitor for both technical roadmaps as well as GTM opportunities.

AGG helps organizations read the layer map simultaneously.

Vertical roadmaps move on their own cadence

A startup watching only the horizontal signal may be watching the wrong roadmap.

Major platforms maintain dedicated product roadmaps for specific industries (healthcare, financial services, manufacturing), that move independently. Platform insertion into vertical markets is one of the fastest growing areas of investment for historically horizontal players.

AGG monitors vertical roadmaps independent of horizontal offerings, including key ecosystem relationships.

Revenue is not the only barometer

Quarterly financials report the past, it's critical to understand the future.

By the time a platform shift appears in the financials, the window for a proactive response may already be closed. The structural signal lives in the platform's product organization — not in the in quarter revenue metrics your company or portfolio is reporting.

AGG starts with the product, not the revenue — because the platform's roadmap may already know what your financials don't yet reflect.

Platform Relationships

Four relationship modes. Each tells a different story.

Every company relates to the platforms around it through multiple modes simultaneously - selling to them, distributing through them, building on their infrastructure, or filling gaps in their product roadmaps. The most significant finding is when one mode looks healthy while another carries critical exposure.

Most Transactional

Sell-to: Platform as customer

The platform organization is itself a buyer of the portfolio company's products or services. Strategic value rises significantly if it is paired with a position deeper inside of the platform's product roadmap (such as a solution) - but on it's own is transactional and is most exposed to budget and consolidation risk.

Most Misunderstood

Sell-through: Platform as channel

The platform's marketplace, co-sell motion, or partner ecosystem is the company's route to market. A marketplace listing and a functioning distribution relationship are not the same thing — and the difference between them is not visible without looking for it. Field & PRM alignment is crucial - yet missing for most.

Biggest risk If only Platform Relationship

Build-on: Platform as foundation

The company's product is architecturally built on the platform's infrastructure, models, or capabilities. The depth of the dependency — whether it creates genuine switching costs or a surface layer the platform can replicate — is the primary diagnostic question this mode surfaces and is the most traditional adoption work platform companies provide.

Most Durable

Build-into: Platform as destination

This is the most invisible relationship, yet the most strategic: a company's product capabilities fill gaps in the platform's roadmap that the platform has chosen the ecosystem should own. Platform product teams run a continuous internal gap analysis. That assessment is not public. It determines whether a company is protected or exposed by the platform's next product cycle.

The AI-Era Platform Stack

A structured ecosystem, not a single relationship.

The actual platform ecosystem is layered and interdependent. Each layer depends on the layers beneath it. Changes at the foundation means a company at L5, L6, or L7 is shaped by forces it may not be watching. The stack is ordered by dependency: L1 is the physical foundation everything builds on. Each layer following it depends on the prior layer below. Platform gravity flows downward - lower layers pull on higher ones as they restructure.

L1 · Hardware · Foundation

L2 · Hyperscalers

L3 · Models The Catalyst

L4 · Orchestration Active battleground

L5 · Enterprise Verticalising

L6 · Vertical Cloud Blindspot

L7 · Applications Highest absorption risk

Most portfolio companies operate at L5–L7 — the layers carrying the highest absorption risk from vertical roadmap convergence at L6 and foundation model expansion from L3.

The stack is a real dependency chain

L3 foundation models run on L2 hyperscaler infrastructure, which runs on L1 GPU hardware. This is not a metaphor, it is an operational dependency. When NVIDIA's supply constraints shape inference economics at L1, that propagates upward through every layer above it.

Building on a platform is not new. The risk profile is.

Companies have always built on platforms and sold through them - Microsoft's partner network, Red Hat's channel, Salesforce's AppExchange. What changed in the AI era is who you are building on. A foundation model provider can now decide to enter your market in a product cycle, not a strategic planning cycle.

Every layer has its own gravity

From hardware substrate to vertical AI applications, each layer has its own ecosystem dynamics, roadmap velocity and pull on the companies above it. Understanding where a company sits across all seven is the foundation of the AGG assessment.

The Catalyst Layer

Every layer restructured. Here is what changed.

The AI era did not upgrade the prior platform stack, it restructured every layer in it. L3 Models is the catalyst: the layer that did not exist as enterprise infrastructure before, and whose emergence changed the operating conditions for every layer above it. Read top to bottom. Foundation at the top, applications at the bottom, gravity flowing down.

How to read platform gravity

L1 and L2 sit at the top - the foundation everything depends on. Gravity pulls downward: as these layers restructure around AI, every layer beneath them is affected. L7 applications at the bottom carry the highest pressure because they sit at the end of that chain. The labels on the right show AGG's assessment of each layer's current force on your portfolio company.

Pre-AI Era · What each layer looked like

AI era · What each layer is now

L1 · Hardware · Foundation

General-purpose CPU infrastructure - Intel, AMD, commodity servers. Stable and commoditised. Not a source of competitive differentiation.

↓

L1 · Hardware · Foundation

GPU-first infrastructure. NVIDIA dominant. Inference cost is now a strategic variable that shapes what is economically viable at every layer above it.

L2 · Hyperscalers

AWS, Azure, GCP as general-purpose cloud - compute, storage, networking. Partner programmes and marketplace mechanics were stable and predictable.

↓

L2 · Hyperscalers

AWS, Azure, GCP now AI-optimised. Marketplace mechanics, co-sell motions, and MDF programmes have been restructured around AI workloads and model deployment.

L3 · Models

This layer did not exist as enterprise infrastructure. Statistical ML was back-room analyst tooling - not something companies built products on or competed through.

↓

L3 · Models The Catalyst

Anthropic and OpenAI ignited the restructuring of every layer above this one. The first layer that is simultaneously infrastructure you build on and a company that could enter your market in the next product cycle.

L4 · Orchestration

Integration plumbing - Mulesoft, Zapier, Apigee. Connected systems to each other. Unglamorous and stable. Nobody fought over it.

↓

L4 · Orchestration Active battleground

The connection layer between AI and enterprise systems - retrieval, routing, agent memory, tool use. Every major platform is fighting to own it. No one has won. A portfolio company here faces acquisition, absorption, or becoming the standard.

L5 · Enterprise

Horizontal platforms - Salesforce, SAP, Oracle, with stable partner ecosystems. The primary distribution channel for most B2B software companies.

↓

L5 · Enterprise Verticalising

Salesforce, ServiceNow, Oracle now building vertical AI products with dedicated industry teams. Ecosystem partners who built on these platforms face absorption from the same platform they depend on.

L6 · Vertical Cloud

Forming but limited - Epic in healthcare, Veeva in life sciences. Small ecosystems, slow cadence, following horizontal platform signals rather than setting their own.

↓

L6 · Vertical Cloud Blindspot

Epic, Veeva, nCino now run independent AI roadmaps that move faster than horizontal signals. Their announcements happen at vertical industry conferences, not the events most investment teams monitor.

L7 · Applications

Vertical SaaS with defined feature sets and stable roadmaps. Competed on features and distribution. The primary arena for portfolio company differentiation.

↓

L7 · Applications Highest absorption risk

Features absorbed into platform copilots and agents from above. Compressed by L5 horizontals verticalising and L3 models expanding upward. The layer where gravity lands hardest.

Reads all seven

The labels on the right - The Catalyst, Active battleground, Blindspot, Highest absorption risk — are AGG's diagnostic read of each layer's current pressure. Where a portfolio company sits relative to these forces determines its classification:

Accelerate Reposition Monitor Triage

A single-layer view misses the cascade. The AGG diagnostic reads all seven simultaneously.

The AI Era

Two critical changes in the AI era.

As a baseline, time to respond to platform movements has compressed from years to quarters. In the AI era, the signal and the announcement can arrive in the same cycle often slip streamed into the core platform at the same time. Two key shifts are happening within this reframing of time:

The infrastructure / application boundary dissolving

In every prior era, infrastructure companies built the substrate and application companies built the products. That separation held. The AI era removed it. Foundation model providers are now simultaneously the layer companies build on and a direct competitor entering their vertical markets with named products and dedicated go-to-market organization. In addition, the application era as we have known it is coming to an end.

Vertical roadmaps became primary, not secondary

In prior eras, the horizontal roadmap was the signal to watch. In the AI era, major platforms maintain dedicated product teams for specific industries whose roadmaps move independently and faster. These announcements appear at vertical industry conferences, not general technology events. A company watching only the horizontal signal is watching the wrong roadmap.

What Platform Intelligence Produces

Platform intelligence from AGG tells B2B technology companies where they are.

Platform Intelligence is not market research or a competitive landscape report. It is the structured ability to assess, with specificity, where a company stands in the AI-era platform ecosystem and what the dynamics already in motion mean for its position over the next 12–24 months.

Explore the Platform Intelligence Framework →

A clear position in the ecosystem

Where the company sits, which direction pressure is arriving from, and which platform forces are relevant to its specific position — not a generic view of the AI landscape, but a specific structural read across the seven-layer stack.

An honest read of each relationship mode

What each platform relationship actually means, which mode is generating real value, which is cosmetic, and where the exposure is hiding behind the metric that looks healthiest.

A specific classification and next action

One of four classifications (Accelerate, Reposition, Monitor, or Triage), with evidence-based rationale and a concrete next action. Not a directional view. A specific, defensible position to act from.

Start with the Framework

Platform position is not static. Neither is the diagnostic.

AGG provides Platform Intelligence Framework diagnostics for VC firms, CVC teams, and B2B technology companies navigating AI replatforming. The diagnostic produces a classification. The classification determines what happens next.

View the Framework

Seeing this from an investor perspective? Explore what AGG does for investment firms →

Building a B2B technology company? Explore what AGG does for tech companies →