Market simulation engagements

Run the launch
before you launch it.

Manic Edge builds a simulated version of your market, with your customers modeled as a population of AI agents, and runs your decision through it. Watch how the community reacts, how competitors counter, and what breaks. Before anything ships.

Start an engagement How it works

01 / The product

A rehearsal space for
decisions you can't take back.

The problem

Markets grade in public.

Pricing changes, paywalls, repositioning: you find out how customers really react only after it's live, when the damage is already done and the community is already angry.

What we do

We simulate your market.

We seed a population of agents from your real customer data, inject the decision you're weighing, and run the simulation multiple times. Then we report what recurs.

What you get

Signal, not vibes.

Reactions that recur across independent runs, ranked by recurrence. Every claim tagged as observed in your data or emergent from simulation, so you always know what you're trusting.

engagement readout / run 2 of 2 

      

      
Real motif analysis from an engine run. One-off reactions are weather. Recurrence is signal.

    


    
    

      

        
02 / The engagement

        
Five steps.
Two weeks. One decision.

      


      
    
        
  1. 
          01
          
    
            
    Scope
    
            
    One decision, sharply defined. We take the engagement only if it passes
            three gates: a discrete intervention, an identifiable population, and real
            data to seed from.
    
          
    
        
    2. 
        
  2. 
          02
          
    
            
    Seed
    
            
    We build the agent population from your data: support logs, research,
            community threads. Weighted like your actual customer base, with hard
            constraints, not adjectives.
    
          
    
        
    3. 
        
  3. 
          03
          
    
            
    Run
    
            
    Your decision goes into the simulation as the single injected variable.
            We run it at least twice, independently, so recurrence means something.
    
          
    
        
    4. 
        
  4. 
          04
          
    
            
    Read
    
            
    A three-pass readout: what your data already says, what emerged in
            simulation, and an audit trail on every claim. We defend it live.
    
          
    
        
    5. 
        
  5. 
          05
          
    
            
    Interview
    
            
    The simulation stays open. Sit down with the agents and ask the
            follow-ups yourself: the question you'd never get to ask a real
            customer the day after a launch goes wrong.
    
          
    
        
    6. 
      

    


    
    

      

        
03 / Pricing

        
Three tiers.
Fixed fee, agreed in writing.

      


      

        

          
Tier 01

          
Community Run

          
$3,500 USD / two weeks

          
Your customers, simulated. One decision run through a
          population seeded from your data.

          
    
            
  • Seeded agent population from your data
    • 
            
  • Two+ independent simulation runs
    • 
            
  • Recurrence readout, every claim tagged
    • 
            
  • Live agent interview session
    • 
          

          Start here
        


        

          
Tier 02

          
War Game

          
$8,500 USD / three weeks

          
Customers and competitors. We profile your
          rivals from the public record, especially how they responded last time,
          and simulate their strategy room reacting to your move.

          
    
            
  • Everything in Community Run
    • 
            
  • Competitor profiles from public record
    • 
            
  • Simulated competitor deliberation
    • 
            
  • Counter-move scenarios ranked by recurrence
    • 
          

          Start an engagement
        


        

          
Tier 03

          
Full Loop

          
Custom / scoped per project

          
Move, counter-move, market reaction, interleaved over
          multiple rounds. Custom market shapes, from B2B buying committees to
          multi-platform communities.

          
    
            
  • Everything in War Game
    • 
            
  • Multi-round move / counter-move loop
    • 
            
  • Custom platform & population shapes
    • 
            
  • Board-level strategic readout
    • 
          

          Talk to us
        

      


      
Scenario plans ranked by recurrence, never
      "Competitor X will do Y." Built from public information and your own data.
      That honesty is the product.

    


    
    

      

        
04 / Who's behind it

        
Built by a practitioner,
not a theorist.

      

      

        

          The simulation engine, the methodology, and the judgment come from
          real-world deployment, not theory. We sign our name to every readout.
        

        
    
          
  • [AWARD] Gold Stevie for AI leadership
    • 
          
  • [DEPLOYED] production AI agents at enterprise scale
    • 
          
  • [MEASURED] $1.5M in operational savings, 70% self-service deflection
    • 
          
  • [METHOD] every readout tagged, audited, and defended live
    • 
        

      

    


    
    

      

        

          
05 / Contact

          
What are you
about to decide?

          
Tell us the move you're weighing. We'll reply within one business day
          with whether it's simulatable, which tier fits, and a fixed fee.

        

        

          
          
          
          
          
          
          
          
        

      

    


    
    

      

        

          
Manic AI · newsletter & podcast

          
Field notes from the edge of applied AI.

          
What we're building, breaking, and learning. No spam, unsubscribe anytime.