Turn business ideas into testable assumptions and experiments.
EMT Copilot helps enterprise innovation teams identify what must be true, assess existing evidence, and select relevant experiments using Precoil's proven methodology and library of 50+ experiments, all within your organization's Microsoft 365 Copilot environment.
Powered by Precoil’s Extract–Map–Test method and 50+ experiment library. Created by David J Bland, lead author of Testing Business Ideas.Stop starting from scratch every time you test an idea.
Innovation teams juggle workshops, spreadsheets, templates, and generic AI prompts to figure out what to test next.
EMT Copilot brings a consistent approach to those decisions, without replacing the team's judgment.
Common Approach
Manually list assumptions in a workshop or document.
Debate priorities without a shared view of evidence.
Search templates or ask a generic chatbot for test ideas.
With EMT Copilot
Extract assumptions from an idea or business model.
Compare importance with evidence strength.
Explore 50+ experiments from a proven, curated library.
Four things your team can do with the Copilot Agent.
Start with a rough idea, an opportunity brief, or a business model. Use the agent to move toward a clearer next experiment.
Turn ideas into explicit assumptions.
1
Surface desirability, viability, and feasibility assumptions that might otherwise remain implicit.
Takeaway: A list of statements that can be challenged and tested.
Assess the importance of each assumption alongside the strength of evidence already available.
Takeaway: A clearer view of the riskiest assumptions.
Find what needs evidence first.
2
Get structured experiment suggestions and how to run instructions informed by Precoil's 50+ experiment library.
Takeaway: Options for what to test next and why.
Select experiments suited to the risk.
3
Use a purpose-built agent in a compatible Microsoft 365 Copilot environment rather than inventing prompts for every idea.
Takeaway: A repeatable starting point inside an existing enterprise tool.
Work in Microsoft 365 Copilot.
4
Organizations we’ve worked with…
Not just a generic AI prompt.
General purpose AI can suggest experiments. EMT Copilot applies a specific, structured experimentation method to help teams decide which suggestions deserve attention.
Precoil EMT System
Extract
Every growth decision carries assumptions about customers and economics are often implicit.
We surface those assumptions so teams can see exactly what must be true for impact.
Map
Not all assumptions carry equal weight, some will break the growth decision if wrong.
We prioritize the riskiest assumptions based on importance and evidence so teams focus on what actually matters.
Test
Instead of debating opinions, we design targeted experiments to generate real evidence quickly.
This allows teams to validate assumptions before committing significant time & money.
One idea, start to finish.
Plug and Play Solar is a removable balcony solar kit for apartment renters. Here is one session with EMT Copilot: the team's own assumptions and documents go in, and a traced assumption set, a map critique, and a library-grounded experiment come out.
Excerpted from EMT Copilot v1.7 output in Microsoft 365 Copilot. Demonstration case, not a client engagement.
Start from the team's thinking, not a blank page.
The team shares a one-sentence idea, the twelve assumptions they had already written down, and the documents they had on hand: their canvases, a draft PRD, and two industry reports.
Takeaway: No prompt writing. The agent asks for what it needs, one step at a time.
Every assumption traced to where it came from.
The agent blends the team's list with what the documents support, sorts everything into desirability, viability, and feasibility, and shows its work in the Source column.
Sources received: Plug and Play Solar - BMC · Plug and Play Solar - VPC · Plug and Play Solar - PRD · Plug-in_PV_Safety_Whitepaper_-_Final · [SolarPower Europe report]
| # | Assumption | Source |
|---|---|---|
| D1 | We believe apartment renters want to generate some of their own electricity despite not owning their homes. | team Plug and Play Solar - VPC × Reduce my monthly electricity bill |
| D3 | We believe apartment renters will seek landlord or property-manager approval when approval is required. | team Plug and Play Solar - VPC × Landlord may not approve |
| D5 | We believe apartment renters will trust a removable electricity-generating product mounted to their balcony to be safe for themselves, neighbors, and property. | inferred from Plug-in_PV_Safety_Whitepaper_-_Final |
| # | Assumption | Source |
|---|---|---|
| V1 | We believe enough apartment renters will purchase the kit at the planned retail price. | team Plug and Play Solar - BMC × Revenue Streams |
| V2 | We believe landlord and property-manager restrictions will leave a sufficiently large eligible market in the launch region. | team Plug and Play Solar - BMC × Customer Segments |
| V5 | We believe the retail price can hold once dedicated-circuit or certified-installation requirements are priced in. | inferred from Plug-in_PV_Safety_Whitepaper_-_Final |
| # | Assumption | Source |
|---|---|---|
| F4 | We believe the kit and installation process can comply with the applicable product-safety, electrical, utility, and building requirements in the launch market. | team Plug-in_PV_Safety_Whitepaper_-_Final × certification |
| F5 | We believe a plug-and-play design can meet emerging safety expectations without dedicated circuits, special receptacles, or other changes that undermine the renter value proposition. | inferred from Plug-in_PV_Safety_Whitepaper_-_Final; [SolarPower Europe report] |
The material is strong on customer value, installation, and safety. It says little about channel access and commercial execution: acquisition cost, property-manager partnerships, fulfillment economics, warranty exposure, and support burden. A financial model and a launch-region regulatory analysis would close those gaps.
F5 pulls against V1 and V2. The value proposition depends on a cheap, simple, renter-installed product, but both reports describe cases where dedicated circuits, special plugs, or professional involvement may be required, raising cost and shrinking the eligible market.
The team places the assumptions. The agent pressure-tests the placement.
Prioritizing stays a team decision. The team maps importance against evidence on their own 2×2, and the agent critiques what they placed. It never builds the map for them.
Takeaway: A second opinion on the riskiest assumptions before anyone spends money.
Ready to move to the Test phase? Name one or two assumptions to test.
Experiments come from a curated library, not improvisation.
The team names an assumption to test. The agent picks a design from the Precoil Experiment Library and writes a brief the team can run, with the assumption copied word for word so nothing drifts.
Takeaway: A cheap first test sized to the risk, with clear signals for right and wrong.
Get the agent and use any idea.
There's no need to assemble a cross-functional team or prepare a formal initiative. Start alone or with a few colleagues and learn how to use EMT Copilot in two focused sessions.
Paid, fixed-fee licensing with guided activation. IT approvals and compatible Microsoft licensing may require separate steps.
EMT Copilot + Guided Activation
01 - Get the EMT Copilot Agent.
Receive the delivered version of the agent with the Extract–Map–Test method and 5)=experiment library for licensed internal use.
02 - 30 minute guided setup.
Get oriented and review the steps to use the agent in your Microsoft 365 Copilot environment.
03 - 90 minute working session.
Bring any idea and work through assumptions, evidence, and experiment options with David J. Bland. No full-day event required.
04 - Keep using the agent.
Continue using the delivered version under the agreed license. Additional workshops and advisory services are optional.
Before you get started.
-
Enterprise innovation teams that already use, or are preparing to use, Microsoft 365 Copilot to evaluate and test business ideas.
-
EMT Copilot uses the structured Extract–Map–Test method and Precoil's curated library of 50+ experiments to connect suggested tests to assumptions and existing evidence.
-
No. Bring a rough idea and attend alone or with a few colleagues. The working session is for learning to use the agent, not running a full-day team workshop.
-
The 30-minute call covers orientation and guided setup. Your organization may need separate administrator approval, compatible Microsoft licensing, security review, or tenant preparation before deployment.
-
Yes. The delivered agent includes Precoil's 50+ experiment library. Internal use and any restrictions on redistribution or integration are defined in the agreement.
-
No. EMT Copilot helps frame assumptions, assess evidence strength, and suggest experiments. Teams still execute tests, interpret real-world results, and decide whether to Commit, Correct, or Cut.
-
EMT Copilot is available through a paid, fixed-fee licensing and activation engagement. There is no public self-serve trial. Licensing scope and commercial details are shared with organizations evaluating deployment.
-
Yes. Facilitated workshops and additional advisory support are available separately, but are not prerequisites for adopting the agent.
Explore EMT Copilot for your organization.
License the agent for internal use, with guided setup and a 90-minute working session. No full-day workshop or cross-functional team required. A paid enterprise offering. Fixed-fee licensing and activation, scoped to your organization's needs. Pricing is shared during a qualified licensing discussion.
