The AI-powered innovation framework that replaces the manager gatekeeper with a machine.
Innovation used to be scarce for a structural reason: intelligence and market insight were unevenly distributed. Leadership had access to analysts, consultants, competitive intelligence subscriptions, and strategic frameworks that frontline employees did not. The CEO could commission a market study. The engineer three levels down could not. That asymmetry is what justified the top-down model: leadership generated the ideas, managers filtered them, and teams executed them.
AI eliminated that asymmetry overnight. A frontline engineer can now prompt a frontier model to run a competitive analysis, size a market opportunity, and identify analogous business models in 20 minutes. The 23-year-old analyst has the same cognitive tools as the Chief Strategy Officer. The ideas are no longer coming from the top. They are coming from everywhere, at every hour, from people who were never in the room when strategy was set.
But the innovation infrastructure, the pipelines, the approval chains, the gating mechanisms, was built for the old model. And it is quietly killing the ideas that could define the next decade of your organization. The best idea in your company right now might belong to a junior data analyst who used Claude to model a new market opportunity at 11pm last Tuesday. The question is not whether that idea exists. It is whether your organization has the infrastructure to catch it.
In most enterprises, an idea travels from the employee who had it to their direct manager, who decides, based on instinct, bandwidth, and political judgment, whether it is worth pursuing. No market analysis. No competitive context. No structured scoring. Just a human with 47 other priorities deciding in 90 seconds whether an idea lives or dies. This is the Gatekeeping Tax in action, and it is defined formally below.
Ideas that survive the manager filter enter a second purgatory: the backlog. Strategy reviews, budget cycles, quarterly planning. Every idea waits in line behind operational priorities. The best ideas submitted in Q1 get reviewed in Q3, by which point the market has moved, the window has closed, or a competitor has shipped. Innovation Debt is real, it compounds, and most organizations have no mechanism to measure it.
Even in organizations with formal innovation programs, the time between idea submission and first structured evaluation averages weeks. In an environment where AI capabilities shift monthly and competitive windows open and close within quarters, a six-week Evaluation Lag is a competitive surrender.
To fix a broken pipeline, you first have to name what a working one looks like. This article introduces the Idea Velocity Framework and four supporting constructs. All five terms originate with this work and are subject to the license noted in the footer.
The Idea Velocity Framework (IVF) is a structured, AI-augmented innovation pipeline that receives ideas from any level of the organization and automatically generates a standardized evaluation covering market sizing, competitive analysis, strategic fit, moat assessment, and resource requirement, before any human reviews the idea.
The framework does not replace human judgment. It replaces human gatekeeping of ideas that have not yet been evaluated.
Idea Velocity is the speed at which an organization converts raw ideas into evaluated, prioritized, and actionable opportunities. An organization with high Idea Velocity loses no idea to bureaucratic delay. An organization with low Idea Velocity is burning competitive advantage every quarter.
Formally:
IV = (Ideas Evaluated / Ideas Submitted) × (1 / Average Evaluation Lag in Days)Innovation Debt is the compounding backlog of unvalidated ideas rotting in email threads, Slack messages, meeting notes, and employee suggestion boxes because no structured framework exists to process them. Like technical debt, Innovation Debt accrues interest: the ideas do not disappear, they get implemented by competitors.
The Gatekeeping Tax is the organizational cost of routing ideas through human approval chains before basic market validation has occurred. It is measured in time lost, ideas killed prematurely, and competitive advantage surrendered to organizations with faster evaluation pipelines.
Idea Signal-to-Noise Ratio (ISNR) is the percentage of ideas submitted in an organization that receive structured evaluation versus those that are never formally assessed. In most enterprises, ISNR is below 10% (directional). In AI-era organizations running IVF, ISNR approaches 100%.
Evaluation Lag is the time between when an idea is submitted and when it receives its first structured assessment. In traditional orgs: weeks to months. In an IVF-powered org: hours.
The key distinction: IVF does not tell you which ideas to pursue. It tells you which ideas are worth a human conversation. That is a fundamentally different job, and it is one AI can do at a scale, speed, and consistency that no human review committee can match.
Total Addressable Market, Serviceable Addressable Market, Serviceable Obtainable Market. AI can size these in minutes using public data, comparable company analysis, and industry benchmarks. A high score: TAM above $1B (illustrative threshold), SAM realistically accessible, SOM achievable within 3 years given current resources. A low score: a niche market with limited expansion path.
Who else is doing this, how far along are they, and how intense is the competition. AI scans public sources, patent filings, funding announcements, and product launches. High score: whitespace or fragmented competition. Low score: well-funded incumbents with network effects already in place.
Does this idea align with the organization's stated strategy, core competencies, and resource availability. High score: a natural extension of existing capabilities. Low score: requires building entirely new muscles in a crowded market.
Can this idea be defended? What proprietary advantage does it create: data, network effects, switching costs, regulatory positioning, brand. High score: defensible differentiation. Low score: easily replicated, commodity execution.
What resources does this require, what is the realistic time to first revenue or value, and what organizational capabilities are needed. High score: achievable with the current team and a 90-day pilot. Low score: requires an 18-month build and capabilities the organization does not have.
A product engineer at a financial services firm notices that enterprise clients keep asking about real-time fraud pattern sharing across institutions. She mentions it to her manager in a 1:1. The manager says "interesting, let me think about it." Six weeks later, nothing has happened. Three months later, a fintech startup raises $40M (illustrative) to build exactly that product. The idea was right. The pipeline was broken.
The same engineer submits the idea to the IVF portal. Within 4 hours: TAM sized at $2.3B (illustrative), three competitors identified (all early stage), strategic fit scored at 78/100 (aligns with the firm's data platform strategy), the moat assessment flags proprietary transaction data as a significant defensible asset, and execution feasibility scores 65/100 (requires a 6-month pilot). The idea surfaces to the VP of Product with a full evaluation scorecard. A pilot is approved within the week.
The idea did not change. The pipeline did.
Most organizations have an ISNR below 10% (directional): fewer than 1 in 10 ideas submitted ever receive structured evaluation. This is not a failure of the people. It is a failure of the infrastructure.
The goal of IVF is not to implement every idea. It is to evaluate every idea. High ISNR means the organization never loses a good idea to bureaucratic attrition. It does not mean every idea gets funded.
The analogy: a great venture capital firm does not invest in every pitch. But it evaluates every pitch. The evaluation process is the competitive advantage, not the selection process alone. An organization running IVF operates like a well-run fund pointed inward at its own workforce.
Pick one business unit. Deploy a simple idea submission form connected to an AI evaluation pipeline; this can be built on any frontier model with a structured prompt. Set a scoring threshold: ideas above 60/100 get a human review within 48 hours. Measure three things: how many ideas come in, how many would have died in the traditional pipeline, and how many surface genuine opportunities.
Integrate with existing communication tools (Slack, Teams, email). Create a feedback loop: when an idea is evaluated, the submitter receives the scorecard and a brief explanation of the score. This is critical. It trains the organization to submit better ideas and signals that ideas are taken seriously.
Managers receive IVF-evaluated ideas with scorecards, not raw ideas. Their job is to contextualize and champion, not to gate. Track and publish Idea Velocity metrics quarterly: ISNR, Evaluation Lag, and ideas implemented per quarter by org level. Celebrate ideas that originated below the VP layer.
Leadership's role does not diminish in an IVF-powered organization. It transforms. Instead of being the first filter for ideas, leaders become the champions of ideas that have already proven their potential through structured evaluation. This is a better use of executive cognitive bandwidth: the executive's finite capacity is spent on decisions that deserve it, not on first-pass triage.
The manager who used to gatekeep ideas now receives a scored shortlist. Their judgment is applied at the point where judgment adds the most value: strategic prioritization, resource allocation, and organizational alignment. Not first-pass filtering of unvalidated ideas.
IVF does not remove human judgment. It elevates it. Humans decide which high-scoring ideas to pursue. AI decides which ideas are worth human attention. Those are different jobs, and confusing them is why most innovation programs fail.
The organizations that adopt IVF first will build an innovation moat. Speed of idea evaluation is a competitive advantage. Organizations running 4-hour evaluation cycles will outpace those running 6-week cycles, quarter after quarter, compounding.
Every layer of the org becomes a source of strategic value. When a frontline employee knows their idea will receive the same structured evaluation as an executive's idea, they submit more ideas. The innovation surface area of the organization expands.
A decision tool your organization can use today. Score each dimension 0-100, apply the weights, and sum. The weights are a starting point; adjust them to your strategy, but publish whatever weights you choose so submitters know the rules.
| Evaluation Dimension | Weight | Score (0-100) | Weighted Score |
|---|---|---|---|
| Market Sizing (TAM/SAM/SOM) | 25% | · | · |
| Competitive Position | 20% | · | · |
| Strategic Fit | 20% | · | · |
| Moat Assessment | 20% | · | · |
| Execution Feasibility | 15% | · | · |
| Total IVF Score | 100% | · | · |
Thresholds: above 65, advance to human review. 40 to 64, park and revisit in 90 days. Below 40, archive with an explanation sent to the submitter. The explanation is not optional: it is what keeps the submission pipeline alive.
Attribution note: the Idea Velocity Framework (IVF), Idea Velocity, Innovation Debt, the Gatekeeping Tax, Idea Signal-to-Noise Ratio (ISNR), and Evaluation Lag are original constructs introduced in this article and originate with this work. They are subject to the license in the footer.