---
title: "Narrative Pillars Examples: How to Pick Your Three Themes"
description: "Narrative pillars examples from crypto and AI, how to choose three themes your comms ladder up to, and a 10-announcement stress test to check they hold."
author: "Shilika Jain"
date: "2026-10-28T00:52:50.255+00:00"
tags: ["web3 pr", "pr strategy", "founder pr", "thought leadership", "content strategy"]
canonical: "https://www.shilikajain.com/blog/three-narrative-pillars-startup-communications"
---

# Narrative Pillars Examples: How to Pick Your Three Themes

By [Shilika Jain](https://www.shilikajain.com/authors/shilika-jain) - 10/28/2026

Narrative pillars examples from crypto and AI, how to choose three themes your comms ladder up to, and a 10-announcement stress test to check they hold.

---

# Narrative Pillars Examples: How to Pick Your Three Themes

Narrative pillars are the three recurring themes every piece of your communication ladders up to: each announcement, op-ed, podcast answer and KOL brief should reinforce at least one of them. Good examples are short, arguable claims about the world, such as "exchanges are becoming banks" or "self-custody is the next fintech", rather than product features. You know you've picked well when your next ten announcements each map cleanly to a pillar and no single pillar hogs them all.

Below are real examples, the criteria I use to choose pillars, and the stress test I run before a team commits to them.

## What are narrative pillars, and how do they differ from a thesis?

A thesis is the one big idea you stand for. Pillars are the three supporting arguments that make the thesis believable, each one a theme you can return to again and again with fresh stories.

If your thesis is the roof, pillars are what hold it up. I've written separately about [picking your one thesis](/blog/how-to-pick-your-one-thesis-web3-ai-founder). This post assumes you have one, or are close, and need the layer underneath.

Pillars are not:

- **Product features.** "Fast transactions" is a spec, not a theme.
- **Audiences.** "Developers, institutions, retail" is a segmentation, not a story.
- **Content formats.** "Education, community, news" describes what you post, not what you argue.

A pillar should be something a reasonable person could disagree with. If nobody would push back, it won't generate stories.

## What are some narrative pillars examples from crypto?

The clearest examples I can offer come from my own public work. When I tore down Tangem's marketing in June 2026, I suggested they stop selling hardware and start selling ownership. Underneath that thesis, I proposed three narratives:

> Push stronger narratives:
> Exchanges are becoming banks.
> Self-custody is the next fintech
> Hardware wallets should be as easy as Apple Pay.
>
> [Shilika Jain on X, 4 Jun 2026](https://x.com/Shilika_jain/status/2062437697978008001)

Look at what each one does:

| Pillar | What it argues | Stories it can generate |
| --- | --- | --- |
| Exchanges are becoming banks | Centralized platforms carry the risks and control of banks | Every exchange freeze, hack or regulatory action becomes commentary material |
| Self-custody is the next fintech | Holding your own keys is a mainstream financial product, not a niche | Adoption data, consumer research, comparisons with fintech apps |
| Hardware wallets should be as easy as Apple Pay | The category's real problem is usability | Setup challenges, UX critiques, product updates framed as progress |

Three pillars, three different news engines. One feeds off the market's bad news, one off data, and one off the product itself. That spread is what you want. The full reasoning sits in the [Tangem teardown](/blog/marketing-teardown-tangem-sell-ownership).

A second example from the same series: for Limitless, I'd position it as the place where the future gets priced first, which naturally splits into pillars around markets beating experts, real-time reaction to news, and local predictions in specific markets.

Chains show the same pattern at a single-pillar level:

> for most users narratives hit on a subconscious level
>
> [Shilika Jain on X, 14 Mar 2023](https://x.com/Shilika_jain/status/1635668279258742786)

In that post I listed Polygon for low fees and Web2 brands, Arbitrum for scalability, Solana for gaming and Scroll for developer experience. Each is one idea repeated until it stuck. A startup usually needs three, because it doesn't yet have the years of repetition that made those single ideas stick.

## How do you choose your three narrative pillars?

I run candidate pillars through four criteria. A pillar needs to pass all four.

1. **Arguable.** Someone credible could disagree. "Security matters" fails. "Most wallet hacks are a design failure, not a user failure" passes.
2. **Provable by you.** You have, or can build, data, product evidence or founder experience that supports it.
3. **Bigger than you.** The theme would exist even if your company didn't. That's what lets reporters use it in stories that aren't about you.
4. **Durable.** It'll still be worth arguing in a year. Pillars tied to one market cycle expire with the cycle.

Then check the set as a whole. The three should cover different sources of news: one tied to the market or industry, one tied to data or research, one tied to the product. If all three depend on your product shipping, you'll go silent every time engineering slips.

A point on balance I made years ago still applies: you need to sit between market trends and how you want users to see you. Pillars that only chase trends sound like everyone else. Pillars that ignore trends get no traction with press.

## How do you stress-test pillars against your next 10 announcements?

This is the test I run before any team locks its pillars. It takes about an hour and it settles more arguments than any brainstorm.

**Step 1.** List the next ten things you plan to announce or publish. Real items from the roadmap, BD pipeline and content calendar, not hypotheticals.

**Step 2.** For each item, mark which pillar it supports. One mark per item, or two if it genuinely serves both.

**Step 3.** Read the grid against these rules:

| Result | What it tells you | What to do |
| --- | --- | --- |
| Every pillar gets 2 or more items | The pillars are fed by real work | Keep them |
| One pillar gets 0 items | It's an aspiration, not a pillar | Replace it, or plan content to feed it |
| One pillar gets 7 or more | It's your product description, not a theme | Split it or widen the others |
| 3 or more items map to nothing | Either the pillars are wrong or the items are noise | Check the items first, then the pillars |
| Most items need two marks to fit | The pillars overlap too much | Sharpen the boundaries between them |

Here's a worked example for an invented company, an AI agent payments startup, so you can see the grid in practice. Pillars: (A) "Agents will need their own wallets", (B) "Payments are the bottleneck for agent adoption", (C) "Developers shouldn't need to be crypto experts to ship agent payments".

| # | Planned item | Pillar |
| --- | --- | --- |
| 1 | Seed round announcement | A |
| 2 | SDK v1 launch | C |
| 3 | Founder op-ed on why agents can't use cards | B |
| 4 | Integration with an agent framework | C |
| 5 | Data on failed agent transactions | B |
| 6 | Podcast on agent identity | A |
| 7 | Hackathon sponsorship | C |
| 8 | New team hire | None |
| 9 | Docs overhaul | C |
| 10 | Commentary on a major agent launch | A, B |

Result: A gets three, B gets three, C gets four, one item maps to nothing. That's healthy. The hire goes into a recap rather than its own announcement. If C had pulled seven items, I'd have told the team their pillars were really one product pillar wearing three hats.

## How do pillars show up in day-to-day communications?

Once agreed, the pillars become the filter for everything that leaves the company.

- **Press pitches** open with the pillar, then the news. The reporter gets the bigger theme first.
- **Op-eds and podcasts** are each assigned a pillar in advance, so the founder's talking points don't drift.
- **KOL briefs** name the pillar the campaign supports and give two lines the KOL can use in their own words.
- **Quarterly planning** picks one pillar to lead each month. I lay out that structure in the [quarterly narrative arc template](/blog/quarterly-narrative-arc-comms-plan-web3).
- **Recaps and founder letters** group updates by pillar, which makes a list of shipped items read as progress on three arguments.

This is also where pillars meet founder voice. A founder who can talk about all three pillars without notes is a founder reporters will call back, which is a big part of how I approach [founder profiling](/services/founder-profiling).

## When do three pillars not work?

Three is a default, not a law.

- **Very early teams with one product and one audience** often do better with one sharp idea and nothing else. Add pillars once you have the proof to support them.
- **Large ecosystems or foundations** may need four or five, one per major audience. Past five, nobody inside the company can remember them, which means nobody outside will hear them.
- **Teams mid-pivot** should hold off. Pillars written for the old product will fight the new one.
- **Pillars without proof** are just slogans. If you can't point to data, product or founder experience behind each one, they'll fall apart the first time a reporter asks a follow-up.

The cost of pillars is discipline. Someone will bring you a good announcement that doesn't fit any of them. Sometimes the right answer is to run it anyway in a recap. If that keeps happening, it's a sign the pillars need revising, which is fine to do once or twice a year. For the deeper positioning work above the pillars, the [Web3 narrative positioning playbook](/playbook/web3-narrative-positioning-strategy-2026) is the place to start.

## FAQ

### How often should you revisit narrative pillars?
Review them every quarter with the stress test and expect to change one about once a year. Changing all three often means the thesis above them has shifted too, and that's a bigger conversation.

### Can narrative pillars be different for different regions?
The pillars should stay the same, but the stories that prove them can change by market. A pillar about self-custody might lean on exchange failures in one region and on consumer research in another, depending on what local reporters cover.

### Should narrative pillars appear on the website?
Not as a list labelled "our pillars". They should show up in the homepage headline, the about page and the boilerplate in a way readers feel rather than see. If a visitor can guess your three themes after five minutes on the site, you've done it right.

### Who should decide the narrative pillars?
The founder and whoever leads communications, with input from product and BD. Keep the final group small. Pillars chosen by a large committee usually end up broad enough to fit everything, which means they filter nothing.

*Have three themes on a whiteboard and not sure they'll hold? [Book a 30-min session](https://calendly.com/shilikajain/30min/) and we'll run your next ten announcements through them together.*

---

**Book a 30-min session to pressure-test your pillars** - https://calendly.com/shilikajain/30min/

Canonical: https://www.shilikajain.com/blog/three-narrative-pillars-startup-communications