Day 1 · Module 5 45 min pre-work · 60 min session · 50 min challenge

Business Model Types: The Smorgasbord

There are more ways to build a viable business from scientific expertise than most researchers have been shown. This lesson is a menu. You don't have to order everything. But you should at least read it before deciding what you want.

What you'll learn

Survey twelve business model archetypes open to researchers, from consulting, productized services, and SaaS to data, platforms, training, publishing, licensing, non-profits, CROs, government contracts, and hybrids
Answer the three questions every model must: who pays, what job they are hiring you to do, and what the unit of value is
Understand the non-obvious payer: why the beneficiary, the payer, and the decision-maker are often three different people
Shortlist the two or three archetypes that best fit your expertise and situation

Pre-Work (~45 minutes)

This is the longest pre-work in the programme. Take your time. The goal is exposure, not mastery. You are tasting a menu, not committing to a meal.

One warning before you start: most researchers, when they hear “start a business,” imagine one specific model: the startup. Funding round, product, scale, exit. That model is real and sometimes the right one. It is also one of the riskiest and most resource-intensive options available. By the end of this lesson, you should have at least ten other models in mind.

The Three Questions Every Business Model Must Answer

Before the archetypes, three questions that every business model answers, often in non-obvious ways:

Who pays? The payer is not always the person who benefits. Health insurance pays for care that patients receive. Google’s advertisers pay for access to users who don’t pay at all. A grant-funded NGO is paid by donors to serve a community that never writes a cheque. Getting the payer wrong is one of the most common reasons research ventures fail.

What job are they hiring you to do? People don’t buy products or services; they hire them to do a job. A pharmaceutical company doesn’t hire a bioinformatician to “analyse data.” They hire them to reduce the time between target identification and a go/no-go decision. A conservation organisation doesn’t hire an ecologist to “monitor species.” They hire them to produce defensible evidence that a court or regulator will accept. The job-to-be-done is almost always more specific than it appears, and the right model flows from understanding it precisely.

What does the unit of value look like? Are you selling time? A deliverable? Access? An outcome? A subscription? Each answer implies different economics, different risks, and different moats.

The Archetypes

These are not the only business models that exist. They are the ones most likely to be relevant to researchers building something for the first time. Read each one. Note the ones that make you think “I could see myself doing something like that.” Note the ones that make you uncomfortable. Both reactions are useful.

Archetype 1: Expert Services (Consulting)

What it is: you sell your time and expertise directly, typically in engagements defined by a scope of work and a deliverable.

Who pays: the organisation that needs the expertise and does not want to employ a full-time specialist.

What they’re hiring you to do: reduce risk, answer a specific question, or solve a problem faster than they could internally.

Economics: time-for-money at its core. Revenue limited by available hours. Margins can be very high (no cost of goods) but do not scale without hiring. Partners in specialist consulting firms often bill $300-1,000+ per hour. Academic researchers, upon first learning this, are often startled.

Case study: The biosafety consultant. A microbiologist with 20 years of biosafety lab experience left her university position and began consulting for pharmaceutical companies setting up new containment facilities. She charges $2,500 per day. She works 80 days per year. She earns more than she did as a professor with a fraction of the administrative overhead. Her clients hire her because getting biosafety certification wrong costs them months of delay and millions in rework. Her day rate is cheap insurance.

Case study: The statistical consultant. A biostatistician offers power calculation and trial design consulting to small clinical research teams that don’t have internal biostatistics capacity. His clients hire him early in the study design phase because errors there are extremely expensive to fix later. He has built a waiting list by publishing one blog post per month explaining common trial design mistakes. The blog is his marketing.

When this works: you have deep expertise in a narrow domain, you can articulate the cost of the problem you solve, and you are comfortable having direct conversations with potential clients. When it doesn’t: you find every client engagement to be starting from scratch, you don’t enjoy the sales relationship, or you want income that doesn’t depend on your personal availability.

Archetype 2: Productized Consulting

What it is: the same expertise, packaged into a defined, repeatable deliverable with a fixed scope and price. You are still selling expertise, but you are not selling open-ended time.

Who pays: the same organisations as consulting, but they prefer predictable scope and price.

What they’re hiring you to do: the same job, but with the transaction simplified and the risk of scope creep removed.

Economics: higher margins than open consulting because the scope is defined and you get faster at delivery over time. The first version takes 40 hours; the tenth takes 20.

Case study: The environmental impact report service. An ecologist packaged her field methodology into a fixed-scope, fixed-price environmental impact assessment for a specific type of land use application (housing developments adjacent to wetlands). She charges $8,500 per assessment, delivers in 6 weeks, and has a template that handles 80% of the work. She completes 15-20 per year alongside a small amount of bespoke consulting. Clients choose her because they know exactly what they are getting and when.

Case study: The genomic market landscape report. A genomics researcher turned entrepreneur sells a quarterly landscape report on a specific technology area (long-read sequencing applications in clinical settings) to six paying subscribers: two instrument companies, two pharma companies, and two investment firms. He charges $4,000 per report per subscriber, does four reports per year, and earns $96,000 in a predictable, renewable revenue stream from a document he produces anyway to stay current in his field.

When this works: you do similar work repeatedly and could define it precisely enough that a client would pay before they receive it. When it doesn’t: your work is genuinely too variable to package, or the discipline of a fixed scope is incompatible with what you actually do.

Archetype 3: Software as a Service (SaaS)

What it is: you build a software tool and charge users (or their organisations) recurring fees for access.

Who pays: researchers, clinicians, companies, or institutions who use the tool.

What they’re hiring you to do: do a specific computational or analytical job that they would otherwise do themselves badly, slowly, or expensively.

Economics: the economics can be exceptional at scale because the marginal cost of an additional user is near zero. Getting there is capital-intensive (building and maintaining software is expensive) and slow. Most research-derived SaaS companies take 3-5 years to reach meaningful revenue.

Case study: Benchling. Founded by two MIT students who were frustrated by the paper lab notebooks they were expected to use, Benchling built digital lab notebooks and molecular biology design tools. The core product is free for academic labs (community moat) and paid for enterprise (pharma, biotech). The founders understood the job-to-be-done (reduce the friction between designing experiments and running them, and make data searchable) better than existing tools did. It is now valued in the billions.

Case study: Geneious. A small New Zealand company (Biomatters) built a bioinformatics workbench for researchers who found command-line tools inaccessible. They targeted the large population of biologists who needed to do sequence analysis but were not bioinformaticians. That population was large, underserved, and willing to pay for a desktop application. They later expanded to an enterprise version (Geneious Prime) sold to pharmaceutical companies for validated workflows.

When this works: the problem is universal enough that many users share it, the software can be built and maintained without a large team (initially), and you can find the first 100 paying users through your existing network. When it doesn’t: the problem is too idiosyncratic to a small number of users to support the economics, or you don’t have (and cannot attract) software engineering capacity.

Archetype 4: Data as a Service

What it is: your data is the product. You charge for access, for analysis, or for specific extracts.

Who pays: companies, institutions, or governments that need the data to make decisions or build products.

What they’re hiring you to do: reduce the cost and time of acquiring, cleaning, and contextualising data that they cannot generate themselves.

Economics: potentially excellent because data has near-zero marginal cost to copy, but there are meaningful costs in collection, curation, quality control, and legal/consent management. The business is often more expensive to build than it appears.

Case study: Precedence Research / Nature-based markets. A team of conservation scientists built a monitoring database of mangrove ecosystem health across Southeast Asia, initially as a research project. They now sell access to that database to carbon offset verifiers, insurance companies modelling coastal flood risk, and governments developing Blue Carbon policies. The same data answers different questions for each client type. The payers are non-obvious: the beneficiary (coastal communities protected from flooding) is not the payer (the insurance actuary who needs better risk models).

Case study: ZoomInfo / academic analogue. ZoomInfo sells business contact data. The analogue for researchers: a team of epidemiologists built a curated database of rare disease patient registries and biobanks worldwide. Pharmaceutical companies pay substantial annual fees to access it when designing clinical trials. The researchers who built it did so partly as a research infrastructure project. The commercial application came later.

When this works: you have data no one else has, the data answers questions that organisations will pay to have answered, and you have solved (or can solve) the consent, privacy, and legal questions around commercialising it. When it doesn’t: the data is replicable, the consent situation is unresolved, or the market willing to pay is too small.

Archetype 5: Platform and Marketplace

What it is: you build an infrastructure that connects two (or more) groups who need each other, and capture value from facilitating that connection.

Who pays: usually the side of the market that derives more value from the connection, or both sides at different rates.

What they’re hiring you to do: reduce the cost of finding and transacting with the other side of the market.

Economics: can be extraordinary at scale (the platform takes a percentage of every transaction and bears minimal per-transaction cost) but require reaching critical mass on both sides before the business works, which is the hardest part.

Case study: Science Exchange. A marketplace for outsourced research services: companies that need experiments run can find validated contract research providers. Science Exchange handles contracting, invoicing, and quality assurance. The core insight was that buying scientific services was extremely friction-heavy (lawyers, NDAs, custom contracts for each engagement) and a standardised marketplace with vetted providers would reduce that friction enough that both sides would pay for it.

Case study: Kaggle (before acquisition). A competition platform connecting data scientists with companies that had prediction problems. Companies paid to post competitions; data scientists competed for prizes. Kaggle captured the fee from the companies. The platform also became a data science community (community moat) that Google eventually acquired.

When this works: you can identify a clear two-sided market where both sides have friction you can reduce, and you have a credible path to enough supply-side participants (e.g., researchers offering services) to attract demand-side participants (e.g., companies). When it doesn’t: one side of the market won’t pay because they can find each other without you, or the trust and quality assurance problem is too hard to solve at scale.

Archetype 6: Training and Education

What it is: you sell the ability to do what you know how to do: through workshops, online courses, bootcamps, certification programmes, or institutional training contracts.

Who pays: the organisation or individual who wants the capability, or (often) the organisation that funds the individual’s professional development.

What they’re hiring you to do: reduce the capability gap faster or more reliably than self-teaching would allow.

Economics: variable. A single workshop can generate a few thousand dollars. A well-designed online course can generate passive income for years. An enterprise training contract with a large organisation can be $50,000-$500,000. The economics depend enormously on the packaging.

Case study: Statistics in the biosciences (many providers). The gap between researchers who need to use statistical methods and those who can teach them rigorously is large and persistent. Dozens of researchers have built training businesses in R, Python, statistics, and bioinformatics for life scientists. Some do it through workshops (in-person, $500-2,000 per attendee). Some through online courses (Coursera, Teachable, direct). Some through institutional contracts with universities or companies.

Case study: The field methods bootcamp. An ecologist who developed novel methods for rapid biodiversity assessment in remote environments now runs a six-day field bootcamp in New Zealand twice a year. Cost: $3,500 per participant, 12 participants per cohort. Revenue per cohort: $42,000. Expenses: accommodation, equipment, one assistant, her time. She considers it a small business that funds three months of her year and connects her to practitioners who later hire her for consulting work. The bootcamp is also her marketing.

When this works: your expertise is something others want to learn, you can systematise it into a teachable curriculum, and you enjoy teaching. When it doesn’t: your knowledge is too tacit to teach, the audience willing to pay is too small, or the time required is not worth the revenue.

Archetype 7: Publishing and Media

What it is: you build an audience by publishing about what you know, then monetise that audience through subscriptions, sponsorships, affiliate revenue, or by converting readers into consulting or course clients.

Who pays: subscribers (directly), sponsors (for access to your audience), or clients who found you through your content.

What they’re hiring you to do: help them understand a complex domain they cannot navigate alone; or (for sponsors) reach an audience they care about.

Economics: slow to build, but the relationship between time invested and revenue earned can eventually be very good. A newsletter with 5,000 paying subscribers at $10/month generates $600,000 per year in recurring revenue.

Case study: Nat Friedman / AI reading. Not a researcher by training but a useful model: technologists who write clearly and thoughtfully about complex topics build audiences that generate enormous downstream value. In the research community: several science communicators have built Substack newsletters with tens of thousands of subscribers. The best ones convert a fraction of that audience to paid subscribers, and a fraction of paid subscribers to consulting clients or course buyers.

Case study: The genomics newsletter (hypothetical but plausible). A genomics researcher writes a weekly newsletter explaining what is happening in clinical genomics to a non-specialist audience of clinicians, investors, and policy people. 12,000 subscribers. 8% are paid at $15/month ($14,400/month). Three sponsors (sequencing company, genomics software company, CRO) pay $3,000/month each for a banner and one mention ($9,000/month). Total: $23,400/month, or roughly $280,000/year, from writing that takes 6-8 hours per week. The researcher retains their academic position.

When this works: you can write or speak clearly about something non-specialists care about, you have the discipline to publish consistently, and you can build an audience without institutional support. When it doesn’t: your area of expertise does not have a non-specialist audience willing to pay, or you find writing for a general audience deeply unsatisfying compared to writing for peers.

Archetype 8: Licensing and Royalties

What it is: you create something once (a patent, a copyrightable method, a software library, a branded curriculum) and license others the right to use it, receiving royalties on each use.

Who pays: the company or institution that uses the licensed asset.

What they’re hiring you to do: the right to use something they could not create themselves as efficiently.

Economics: potentially excellent if the licensed asset is valuable and widely used, but most licensing income is much smaller than people expect. University technology transfer offices report that the vast majority of licensed patents generate minimal royalties; a small number generate enormous ones. The median is modest.

Case study: CRISPR. The CRISPR patents (contested between Broad Institute and UC Berkeley) are worth hundreds of millions in licensing fees from the companies that have built products using CRISPR. This is the licensing success case that everyone hears about. It required a decade of litigation, institutional resources that most researchers don’t have, and a technology that happened to be foundational to an entire industry. It is not a representative example.

Case study: A validated environmental monitoring protocol. A more accessible licensing model: a research team developed and validated a rapid water quality monitoring protocol that could be used without specialist equipment. They published the method, then licensed the branded, certified version of the protocol to testing companies operating in markets where certification matters. Royalties are modest ($15,000-40,000 per year) but ongoing and require no active work after the initial certification process.

When this works: you have created something with genuine standalone value that others can use independently, the legal situation around the asset is clear, and you can manage (or partner to manage) the licensing relationships. When it doesn’t: the asset is not genuinely reusable without your involvement, or the market is too small to generate meaningful royalty income.

What it is: a mission-driven organisation that earns revenue through grants, donations, government contracts, and earned income, and uses that revenue to pursue a defined social or environmental mission.

Who pays: foundations, government agencies, individual donors, and sometimes beneficiaries who can afford to pay.

What they’re hiring you to do: produce social, environmental, or public health outcomes that the market would not produce efficiently on its own.

Economics: structurally different from for-profit businesses. Revenue is often less predictable (grant cycles, government budgets) but overhead can be lower and the model can survive without turning a profit. New Zealand has specific legal structures (Charitable Trust, Incorporated Society) that provide tax advantages and donor trust.

Case study: BERL (Business and Economic Research Ltd). A New Zealand research consultancy structured as a not-for-profit that produces economic analysis for government, iwi, and community organisations. They compete for contracts that for-profit consultancies also seek, but their structure conveys independence and public interest credibility that can be decisive in certain contexts. Revenue comes from contract research, not donations.

Case study: A conservation genomics trust. A group of conservation scientists established a charitable trust to fund and coordinate environmental DNA (eDNA) monitoring of New Zealand freshwater systems. The trust receives government funding, foundation grants, and industry contributions from companies with environmental compliance obligations. The scientists involved do work they would otherwise do in a university, but with more control over research direction and more direct connection to the communities that benefit.

When this works: the mission genuinely cannot be sustained by a for-profit model (the beneficiaries cannot pay, or the social return exceeds the commercial return), you can build the governance and administrative capacity a non-profit requires, and the funding landscape supports what you are trying to do. When it doesn’t: the mission could be served by a for-profit model (non-profit status is not required to do good work), or the grant dependency creates instability that undermines the mission itself.

Archetype 10: Contract Research Organisation (CRO)

What it is: you sell research capacity: the ability to run experiments, generate data, and produce results for clients who cannot or do not want to maintain that capacity internally.

Who pays: pharmaceutical companies, medical device companies, agribusiness, government agencies; any organisation that needs research done and is outsourcing it.

What they’re hiring you to do: run high-quality, validated, documented experiments faster and cheaper than they could do internally.

Economics: project-based revenue with relatively high overhead (equipment, staff, quality systems). Margins can be thin for commodity services but are much better for specialised capabilities that few others offer.

Case study: A specialist plant pathology CRO. A plant pathologist with deep expertise in a specific crop disease cluster left her university position and set up a small CRO offering validated disease challenge assays for agrichemical and plant breeding companies. Her lab can run specific tests that larger CROs don’t have the biological materials or expertise to run reliably. She is booked six months ahead and charges a premium for specialisation. Her moat is the combination of validated protocols, biological reference materials, and a decade of assay-specific expertise.

Case study: A computational CRO. A team of bioinformaticians offers validated, documented computational analysis pipelines to small biotech companies that have data but not the in-house capacity to analyse it for regulatory submission. The “CRO” model here is software-light but methodology-intensive: what clients are buying is the documented validation of the pipeline, not just the output, because regulators require it.

When this works: you have the equipment, methods, and quality systems that clients need and cannot build efficiently themselves, your specialisation is narrow enough that competition from large CROs is limited, and you can build the project management and compliance capacity that CRO clients expect. When it doesn’t: the regulatory overhead is too high for your scale, or your specialisation is not narrow enough to command a premium.

Archetype 11: Government and Institutional Contracts

What it is: you win competitively tendered contracts from government agencies, research institutions, or international bodies to do defined work.

Who pays: government agencies, public research funders, international organisations.

What they’re hiring you to do: deliver expertise, analysis, or research capacity that they cannot maintain internally but need for a defined purpose.

Economics: can be excellent (government contracts are often large and multi-year) but are slow to win (procurement timelines are long), require significant proposal and compliance overhead, and are subject to political and budget cycles.

Case study: MPI regulatory science contracts (NZ). The New Zealand Ministry for Primary Industries regularly contracts specialist scientists for work it cannot do internally: novel biosecurity threat assessments, risk modelling for new pathogens, environmental impact reviews for new agricultural practices. Researchers who understand MPI’s operational context and who have maintained relationships with programme managers win a disproportionate share of this work.

Case study: SBIR/STTR (US equivalents). The US Small Business Innovation Research programme provides grants to small businesses for R&D with commercial potential. Researchers who understand both the technical requirements and the grant mechanics have built small companies largely on SBIR funding, buying years of runway to develop products before needing external investment. New Zealand’s equivalent includes Callaghan Innovation R&D grants and various MBIE funding schemes.

When this works: you understand the procurement context and relationships in a specific agency, you have the compliance and reporting capacity for government contracts, and your work genuinely aligns with the agency’s statutory obligations. When it doesn’t: the procurement timeline is incompatible with your cash flow needs, or you find the compliance and reporting burden incompatible with how you work.

Archetype 12: Hybrid and Portfolio Models

What it is: most viable research ventures eventually combine two or more of the above archetypes in ways that support each other.

The most common combinations:

Consulting + Training. Consulting builds credibility and case studies. Training monetises the knowledge at scale and converts some trainees into consulting clients. The two reinforce each other.

SaaS + Consulting. The software handles the routine; consulting handles the complex and high-value. The software is marketing for the consulting; the consulting funds development of the software.

Publishing + Consulting. The newsletter or podcast builds an audience. A fraction of that audience becomes consulting clients. The content is both the marketing and a revenue source in its own right.

Data + Software. The data is the moat; the software is the access mechanism. Clients pay for software access; the software is only valuable because of the underlying data.

Non-profit + Earned Income. The charitable trust or NGO structure enables philanthropic funding; a trading arm or consulting subsidiary generates earned income that reduces grant dependency.

Case study: Organisational Mycology. (This programme is itself a business model case study.) The entity behind Growth Medium operates a hybrid model: consulting work funds operations and builds credibility; structured programmes like this one systematise and scale that expertise; relationships built in workshops generate future consulting work. The programme is simultaneously a revenue line, a marketing channel, and a moat-building exercise in community and reputation.

The Non-Obvious Payer: The Most Important Thing In This Lesson

If you read nothing else carefully, read this.

The payer in your business model is often not the person who benefits from your work. This is not unusual or problematic. It is, in fact, one of the most powerful design insights available when building a research-based venture.

Example 1: Conservation science. The beneficiary of biodiversity monitoring is the ecosystem and the communities that depend on it. They cannot pay. The payer might be: a government with environmental obligations, a company with a biodiversity net gain commitment, an NGO funded by donors who care about conservation, or (increasingly) a carbon or biodiversity credit market. Understanding who pays changes what you build and who you sell to.

Example 2: Public health research. The beneficiary of disease surveillance is the population. They don’t pay. The payers are government health agencies, international health organisations, pharmaceutical companies building vaccines, and insurance actuaries modelling risk. The same research can be sold to multiple payers who are hiring it for completely different jobs.

Example 3: Clinical decision support. A diagnostic tool benefits the patient. But the patient doesn’t choose the tool and often doesn’t pay for it. The payer is the hospital or clinic. The purchaser (who decides to buy) is the clinical department head or procurement committee. The user (who operates it) is the clinician. These are three different people with three different jobs to be done and three different things they need to be convinced of. A product that only addresses one of them will struggle.

Example 4: Education. The beneficiary of a training programme is the researcher who attends it. But the payer might be their employer, a professional society, a government workforce development programme, or a research funder who has decided that training is part of what they fund. Each payer has different requirements for what the training must demonstrate.

The exercise in your challenge asks you to map the full triangle: who benefits, who pays, and who decides. Getting all three right is not easy. But not thinking about all three is one of the most common reasons research ventures fail to find a sustainable model.

Reflection Prompts

Which of the twelve archetypes made you think “I could see myself doing something like that”? Which one surprised you?
Who actually pays for the work you currently do (directly or through grants)? Who benefits? Are they the same people?
If you could only choose one archetype to explore further after this programme, which would it be and why?

Session

Facilitator guide. 60 minutes. This session is exploratory, not convergent. The goal is to expand participants’ conception of what is possible, not to make a decision.

Discussion 1: The Model Gallery Walk (15 min)

Post all twelve archetypes on the wall (or display on screen). Give each participant three sticky dots.

Ask each person to place a dot next to the archetype that most resonates with them, most surprises them, or that they want to understand better. One dot per reason (up to three dots per person, but not necessarily one in each category).

After 5 minutes, step back and read the pattern. Where are the clusters? This is a quick picture of where the group’s energy and curiosity sit.

Spend 8 minutes discussing the clusters: why these models? What is appealing? What feels scary? What do people think they’d need that they don’t currently have?

Discussion 2: The Payer Triangle (15 min)

Small groups of 3-4. Take 10 minutes then share back.

Each group takes one case from the pre-work where the payer, the beneficiary, and the decision-maker are different people. Their task: draw the triangle. Name each party. Name what each party needs to be convinced of.

Then: apply the same exercise to someone in the group’s actual work or potential venture. Where are the three parties? What does each one care about?

Share-back: one triangle per group. Note where participants find it hard to name the decision-maker. That difficulty is often the key insight.

Discussion 3: The Same Expertise, Different Model (10 min)

Plenary.

Pick one participant (with their permission) who has agreed to use their own work as a case. Using their expertise as the input, brainstorm how many different business models could be built from it.

The facilitator’s job here is to push past the first obvious answer. Most researchers will immediately say “consulting.” Push: what if you productized it? What if you trained other people to do it? What if the data you generate in consulting engagements became a product? What if you wrote about it and built an audience?

The point is not that they should do all of these. The point is that the expertise has more possible expressions than the first answer suggests.

Discussion 4: The Uncomfortable Model (10 min)

Plenary.

Ask the group: which model in the pre-work made you most uncomfortable, and why?

Common answers and productive responses:

“Licensing feels like I’m trying to extract money from other researchers.” Distinguish between academic licensing (often zero or nominal cost, which is fine) and commercial licensing (charging companies that use your IP to make money). These are different transactions with different ethics.

“The media / newsletter model feels like I’d be dumbing things down.” Push back: the best science communication is not dumbing down, it is translating. The researchers who write clearly for non-specialists often end up with more influence, not less.

“CRO work feels like selling out the independence of research.” Distinguish between CRO work (which is defined-scope, client-directed) and research (which is investigator-directed). Both have value. Many researchers find CRO work satisfying because the applied context is clarifying. The key question is whether you can build a practice that includes enough investigator-directed work to stay intellectually alive.

Discussion 5: The Portfolio Sketch (10 min)

Pairs, then share back.

Ask each pair to sketch a two-year portfolio model for one participant: a combination of two archetypes that would work together and be achievable from their current position.

Ground rules:

Start from what they have now (skills, relationships, data, reputation)
One archetype should be achievable within 12 months; the other can be a 2-year horizon
The two should reinforce each other (client from consulting work becomes training participant; newsletter audience becomes consulting pipeline, etc.)

Share-back: one portfolio sketch per pair, described in three sentences.

Challenge (~50 min)

This challenge produces a Business Model Canvas sketch. It is intentionally rough. A rough sketch that surfaces real questions is more valuable than a polished document that hides them.

Part 1: The Payer Triangle (10 min)

Draw three boxes: Beneficiary, Payer, Decision-Maker.

For the work you are most interested in building, name the entity in each box. If the same entity appears in more than one box, that is useful information (it simplifies your model). If all three are different, that is also useful information (it means you have three different stakeholders to address).

For each box, answer: what does this party primarily care about? What would make them say yes?

Part 2: The Model Shortlist (10 min)

List the three archetypes from the pre-work that interest you most. For each, write two sentences: what would this look like applied to your specific expertise, and what is the main obstacle you can currently see?

Then: cross out the one that is least likely to work in the next 12 months. Circle the one that is most likely to work.

Part 3: The Unit of Value (10 min)

For your leading archetype, describe the unit of value you would sell. Not a description of your expertise. The specific thing a client would pay for and receive.

Examples of well-defined units:

“A three-day on-site assessment resulting in a written report of no more than 20 pages, delivered within four weeks of engagement.”
“Annual subscription access to a database of [specific data], updated quarterly, including an annual briefing call.”
“A five-day workshop for up to 12 participants covering [specific curriculum], delivered at the client’s location.”

If you cannot describe the unit of value in one sentence, that is a signal that the model needs more definition before it can become a business.

Part 4: The Non-Obvious Question (15 min)

Write answers to these three questions:

Who is paying for your expertise right now, even if indirectly? (Your grant funder, your employer, the taxpayer, a foundation.) What are they trying to accomplish by funding you?
If you moved to a direct commercial model, which of the twelve archetypes most closely resembles what that funder is already paying for? Could you offer the same value directly to a different payer?
What would you charge? Be specific. What is a price point that feels uncomfortably high to you? Research suggests that discomfort at your own price is often a signal you are getting closer to the right one, not that you are overcharging.

Part 5: Bring It Forward (5 min)

Write one sentence about the model you want to explore after this programme, at a level of specificity that would let someone else give you useful feedback on it.

Not: “I’m interested in consulting.” Instead: “I want to offer paid statistical design reviews to small clinical research teams in New Zealand and Australia, priced at $2,500-4,000 per engagement, targeting teams who are 3-6 months from submitting a trial protocol.”

That sentence is what you will bring to your 90-day planning session on Day 2.

Key Concepts Reference

Archetype: a repeatable pattern for how a business creates and captures value. Most viable ventures combine elements of more than one archetype.

Payer vs. beneficiary: the entity that pays for a product or service is often distinct from the entity that benefits from it. Mapping both (and the decision-maker who sits between them) is essential to understanding why a model does or does not work.

Unit of value: the specific, describable thing a client pays for and receives. Defining this precisely is a prerequisite for pricing, scoping, and selling.

Productized consulting: consulting services packaged into a defined, repeatable deliverable with a fixed scope and price, allowing faster delivery and higher margins than open-ended time billing.

Jobs-to-be-done (JTBD): a framework for understanding why customers use a product or service, framed as the job they are hiring it to do. The same expertise can be hired to do very different jobs by different customer types, which is why the same research can support multiple business models. JTBD is covered in depth in the Day 2 curriculum.

Portfolio model: a business that deliberately combines two or more archetypes that reinforce each other. Common in research ventures because different revenue streams (consulting, training, publishing, licensing) can share the same underlying expertise while diversifying risk.

Freemium / open core: a model in which a core product is available free (to build community and network effects) and a premium version or enterprise tier is paid. Common in research software. Requires careful design of the line between free and paid to avoid giving away too much value or too little.