Systematic Ideation for Startups & Venture Theses

20 Mental Models for Idea Generation — Part 1

Ali Afridi


Update: A visual reference guide to this post can be found here.

As thesis-driven early-stage VCs, one of the activities that we’re frequently engaged in is startup ideation. Whether it’s discussing new opportunities with entrepreneurs, developing concepts that can serve as investment theses, or for businesses our firm can incubate, the activity can be perpetual in venture.

As we dig into different markets and industries to identify opportunities, there are several mental models I find myself referring back to for ideation. These models are not novel — many have been written about in business books and blog posts — however, my goal here is to compile and present these in a way that enables more systematic ideation.

I’ve personally found that laying the models out in the manner presented below has enabled me to more proactively source insights by knowing what to look for and to develop some pattern recognition on the types of approaches that can work based on different market and firm-level signals.

An inspiration for this activity is the codification of investment ideation that has happened in public market investing. The clear identification of strategies has enabled investors to develop top-of-the-funnel screens and repeatable processes around ideation (more info on pages 76–77 of this guide). Although startup ideation tends to be much more complex and far-reaching, there are some similar models that can be leveraged. As private market data becomes increasingly structured, there may be a similar ability to start developing screens to identify potential new venture opportunities.

A few quick notes before diving into the models:

  1. These mental models are not mutually exclusive, and certainly not collectively exhaustive.
  2. I’ve used curly brackets {} at various points in this post to represent variables that can be swapped out. I’ve described how to do this in areas where they appear.
  3. The models are meant to spark ideas and enable brainstorming for top-of-the-funnel concepts. Each will require further diligence to determine whether the opportunity to pursue the concept exists.
  4. This is the first version of a post I anticipate I’ll revamp and refine over time. I welcome any feedback or discussion around the models or frameworks listed or others I’ve missed.

Categorizing the Mental Models by Key Input Insight

Since 20 models are a lot to remember, I’ve proposed an initial structure on a way to classify and organize the models by the primary insight they capitalize on. This is meant to enable one to keep an eye out for the key insights, and then leverage this post as a reference doc for ideation afterwards.

The 20 Mental Models Broken Out By Key Input Insight

One thing worth noting is that although I have grouped these models by the primary insight that feeds into them, market shifts & catalysts can play a major role across many of these. Shifts in an industry could be enabling the growth of an emerging strong market player, could be turning previously well-situated market players into ones ripe for disruption, or could be creating opportunities through changing customer behaviors or expectations.

The 20 Mental Models for Startup Ideation

  1. Solving Problems & Unmet Needs
  2. Juxtaposition
  3. Disrupting Institutions & {Share Donator} 2.0
  4. Fast Followers & Clones
  5. Unbundling
  6. Bundling
  7. Disintermediation / Cutting out the Middleman
  8. Forecasting
  9. Backcasting
  10. Concepts from Fiction

Models 11–20 can be found in Part 2 of this post.

1 — Solving Problems & Unmet Needs

Solve an existing problem that individuals or businesses would pay to have solved, or that can be monetized in an indirect way.

Doug Clinton’s post (linked below) breaks out 3 of the types of problems around which venture-scale businesses can be built: 1) Cost, 2) Convenience, and 3) Moral.

Example Use: Build a business that reduces the cost burden of secondary education. Start a business that increases access to affordable healthcare.

Input insights to look out for: Problems faced by individuals, groups of people, or businesses.

Additional reading: Great Startups Start With a Problem by Doug Clinton.

2 — Juxtaposition

Juxtapose a business or business model that is working in one market and apply it to a different category or market with similar dynamics.

Juxtaposition generally takes the form of “{x} for {y}” where:

  • {x} can be replaced by a specific business or business model
  • {y} can be replaced by a different industry, category, or technology platform or paradigm.

Note: {y} can also be replaced with a different geography. I’ve broken out this approach further in a separate model (#4 — Fast Followers & Clones)

Example Use: Business in a Box for {y}, Managed Marketplace for {y}, Shopify for {y}, Veeva for {y}, Superhuman for {y}, DTC Brand for {y}, {x} for Mobile, {x} for the Cloud, etc.

Input insights to look out for: Business models or products that are doing well and an understanding of the market dynamics that are enabling the business to succeed. Alternatively, start with a new tech platform or ecosystem and juxtapose business models from previous platforms/ecosystems onto it.

Additional reading:

3 — Disrupting Institutions & {Share Donator} 2.0

Develop a business that directly goes up against an existing institution or large business that is ripe for giving up market share or profits: the Share Donator (a term coined in Quality Investing).

Porter’s Generic Strategies give a good basis for the techniques by which a business can compete against a Share Donator: creating a significantly better product / service (often referred to as a 10x or 100x better product), offering a comparable product at a lower cost, or by going after a market segment the player does not currently service. I’ve laid out the going after an underserved market segment as a separate model (#16 — Expand A Market).

Example Use: Disrupting Banks (ex. Nubank), Disrupting Universities (ex. Lambda School), WebEx 2.0 (ex. Zoom), PeopleSoft 2.0 (ex. Workday), Procare 2.0 (ex. Brightwheel), YPO 2.0 (a wide number of emerging startups).

Input insights to look out for: Companies that offer sub-optimal products or charge excessively in relation to the value they provide. Those that are disliked by key stakeholders or that have failed to innovate as industries or technologies evolve are also good targets. Quality Investing further lays out ignored divisions of large companies and companies with entrenched cost or management structures as further targets.

Additional reading: Taking the wrong lesson from Uber and How to Build an enduring, multi-billion dollar business by Sarah Tavel. Your Product Needs to be 10x Better than the Competition to Win by Mark Suster. Quality Investing (Chapter 1E specifically talks about Share Donators).

4 — Fast Followers & Clones

Develop a business that is similar in value prop and business model to another that has been observed to be doing well.

This approach works especially well in non-winner-take-all-markets, regulated markets that make it slower to scale geographically, or in CapEx or service-intensive businesses where capital (financial or human) is a constraint to growth. Often the fast follower will target a different geography so as not to be a direct competitor on Day 1 although this is not always the case.

Examples: Publicly shared examples include Gilt (inspired by Vente Privee in France) and most of Rocket Internet’s portfolio (ex. Alando which was inspired by eBay and CityDeal which was inspired by Groupon). Despite its prevalence, few companies will acknowledge when they leverage this strategy unless they target a different country.

Input insights to look out for: Businesses that are growing fast and/or performing well.

Additional reading: Rocket Internet: What It’s Like to Work at a Startup Clone Factory by Sam Parr, Rocket Internet — A detailed look An (sic) analysis about Rocket Internet by Christoph Gerber and Rocket Internet’s 2015 Annual Report (pages 20–25 specifically dive into the shared infrastructure they developed to quickly launch & scale new concepts).

5 — Unbundling

Launch a business that picks apart an existing company’s offering or user-base.

Unbundling can be thought of in many ways, but 4 common approaches are:

  1. Unbundling horizontal platforms by verticals or categories (ex. Unbundling Craigslist by category)
  2. Unbundling product/service suites by constituent offerings (ex. Unbundling Banks by launching products better suited for each financial service they provide)
  3. Unbundling by customer segment (ex. Unbundling Zoom to create verticalized offerings for specific industries; Unbundling Tinder by ethnicity, religion or language)
  4. Unbundling a firm’s value chain (ex. Unbundling insurance companies by launching businesses that can do the task of an internal team or division more efficiently)

A-C above are generally disruptive to existing companies in the market. However, D can be an enabler. I’ve laid out D as a separate model: #18–Enabled Outsourcing.

Example Use: Unbundling Reddit, Unbundling OLX, Unbundling Upwork, Unbundling Banks, etc

Input insights to look out for: Companies servicing a wide range of industries, categories, products, or customer sizes where targeted offerings would be more appealing to a sub-segment of users.

Additional reading: The Spawn of Craigslist by Andrew Parker, The Magic of Liquidity: Web Marketplaces Still Have a Long Way to Go by David Haber, Vertical or Horizontal by Josh Breinlinger, Disaggregation of a Bank by Zander Pease, Banking is Under Attack by Tom Loverro, The Unbundling of Excel by Tom Tunguz, The Startups Unbundling FedEx, UPS, & the Logistics Industry by Michael Dempsey, Platforms vs Verticals and the Next Great Unbundling by Jeff Jordan and D’Arcy Coolican, The Unbundling of Harvard Has Begun by Brett Goldstein and Unbundling Zoom by JJ Oslund.

6 — Bundling

Combine products, features, and/or services into a single company that previously required going to multiple vendors.

A common usage of bundling in the startup world over the past several years has been bundling financial services and/or insurance alongside internet platforms or software products (also called embedded financial services or insurance).

Example Use: Bundling different types of documents into a single interface (ex. Coda & Notion), Bundling payroll software with cash advances (ex. Gusto), Bundling benefits with Human Resource Information Systems (ex. Zenefits), Bundling Video Conferencing with games & activities (ex. Icebreaker)

Input insights to look out for:

  • For bundling products or services: Keep an eye out for products or services that are frequently purchased in tandem or where a large number of users have similar workflows they have to go to multiple vendors for.
  • For bundling software w/ financial services: Look for software or internet platforms that facilitate transactions, have access to proprietary data for underwriting, or are embedded in a workflow through which financial services/insurance can be distributed.

Additional reading:

7 — Disintermediation / Cutting out the Middleman

Leverage technology to cut out an intermediary present in a transaction or supply chain.

Examples: Cutting out wholesalers & retailers and selling products direct (large number of DTC companies), Disintermediating Business Process Outsourcing (BPO) firms (UpWork, LiveOps)

Input Insights to look out for: Supply Chains or Industry Value Chains where activities performed by certain players (often-times matchmakers between supply & demand) can be efficiently replaced with technology solutions

Additional reading: “Cutting out the middleman” (LooseThreads) and Tech savvy manufacturers set for £13bn boost as they cut out wholesalers and shops to sell direct to consumers (Barclays).

8 — Forecasting

Forecast what an industry will look like by extrapolating current catalysts, shifts, or trends. Build a business that capitalizes on or accelerates those shifts. Exhibit 1 (Page 9) of Sharpening Your Forecasting Skills highlights the characteristics of ‘superforecasters’ and the methods that can be leveraged for forecasting.

In addition to extrapolating current trends, exploring the second or third-order effects of trends can also help identify opportunities earlier on. Qasim Mohammad’s post on the second-order effects of the DNVB economy gives a demonstration of how this can be leveraged.


  • First Order: As e-Commerce penetration across the globe grows, build marketplaces and platforms that enable consumers to shop online.
  • Second-Order: As the marketplaces gain market-share and mature, build companies that are better positioned to operate on these marketplaces or that assist merchants on them.

Input insights to look out for: Market catalysts or shifts that are changing how markets operate. At Equal Ventures, we leverage the PEST framework to break out and monitor such shifts.

Additional reading: Sharpening Your Forecasting Skills by Michael Mauboussin & Dan Callahan, On Inflection Points by Michael Dempsey, Get ready for the second order effects of the new DNVB economy by Qasim Mohammad and Superforecasting: The Art and Science of Prediction by Philip E. Tetlock & Dan Gardner.

9 — Backcasting

Envision what an activity, industry, or the world in general will look like at a point in the future. Develop businesses that get the world to that point or are relevant in the envisioned world.

There are many techniques that can be leveraged to invent a picture of the future, to the point that it has its own field of study (Futures studies).

Example: Envision a world where every individual has equal access to any opportunity. Develop concepts that could work in that world or help the world get there.

Input insights to look out for: Various, depends on the type of technique being leveraged for creating a picture of the future

Additional reading: How to build a breakthrough… the secret of Backcasting by Mike Maples Jr. explores backcasting by tracking inflections, Thinking in Bets by Annie Duke, Futures techniques (Wikipedia).

10 — Concepts from Fiction

Take concepts from fiction, oftentimes science fiction, and bring them into the real world.

This model is really just one way to leverage backcasting (above) but given the impact it has had on numerous innovators I find it helpful to break it out separately.

Examples: Search engines, online social networks, non-fiat crypto-currencies, automated delivery dispatchers and dozens of other startup & technology concepts were written about in popular cyberpunk & science fiction novels that predated the innovations by many years or even decades. One of the most talked about science fiction concepts that founders & VCs across digital communications, social networking, and gaming reference as a continuous inspiration is the Metaverse from Neal Stephenson’s Snow Crash.

Input insights to look out for: Concepts described in works of fiction (movies, magazines, TV shows, books, comics, etc).

Additional reading: The importance of science fiction to entrepreneurship by Ben Narasin, The Sci Fi Guru Who Predicted Google Earth Explains Silicon Valley’s Latest Obsession by Joanna Robinson, How Star Trek Inspired Amazon’s Alexa by Khari Johnson and The S1 Club | Unity is Manifesting the Metaverse by Mario Gabriele.

Models 11–20 can be found in Part 2.

A huge thank you to Richard Kerby, Rick Zullo, Simran Suri, Chelsea Zhang, Ahmed AlRawi, Joseph Milla, Arjun Kumar, Amrit Singh, and Chase Bonhag for contributing insights and reviewing drafts of this post.