LLenny's Podcast
← All frameworks
MarketingYuriy Timen (Grammarly, Canva, Airtable)

The Three Growth Engines Selection

Diagnose which of virality, SEO, or paid your product can actually grow on before spending on any

Difficulty
Moderate
Time to result
~months to results
Steps
3
Confidence
90%

Yuriy Timen argues most subscription products grow overwhelmingly on ONE of three engines: word-of-mouth/virality, SEO, or paid. Rather than trying all three, you diagnose which one your product is structurally suited to by testing for specific pre-conditions. Virality is preferred if the signals exist, then SEO, then paid as the fallback.

Origin

Yuriy Timen's synthesis from eight and a half years leading growth at Grammarly and advising ~15 companies including Canva, Airtable, and Hims & Hers.

Core principles

  • 01Growth engines have structural pre-conditions — they can't be willed into existence
  • 02Product network effects are present at inception or nearly impossible to manufacture later
  • 03Most admired companies run one dominant engine, not a diversified portfolio
  • 04Match the engine to the product's inherent properties, not to founder preference

How to run it

  1. 1

    Test for virality first

    Check whether the product has inherent network effects — utility increasing as more users join, especially within the user's own team or market segment. Also check for a genuinely beloved brand that can carry give-one-get-one referrals even without network effects.

    Pro tip Collaboration tools (Airtable, Whimsical, Monday) and marketplaces have inherent network effects; a single-player tool like Grammarly does not.

    Watch out Do not try to engineer network effects into a fundamentally single-player product — Timen calls it an uphill battle that rarely works.

  2. 2

    Test for SEO next

    If virality isn't viable, assess whether you have a defensible SEO angle: a unique editorial contribution, a programmatic long-tail angle, or a proprietary data angle. Canva's early engine was long-tail programmatic SEO on 'make' and 'template' keywords.

    Pro tip Long-tail programmatic SEO builds a durable moat that is more defensible than paid.

    Watch out SEO has a 6-month-plus return horizon and may never pay back; it is not for companies needing immediate return.

  3. 3

    Fall back to paid

    If neither virality nor SEO fits, assess paid. It works when single-player LTVs are high (prosumer buyers expensing $120-130/year) and free-to-paid conversion exceeds ~5%, allowing you to pay back ad spend.

    Pro tip Prosumer buyers using a single-player tool for work have LTVs in the hundreds vs 50-60 for pure consumer subs.

    Watch out Everyone can try paid but it doesn't work for everyone — weak funnel economics make it a cash drain.

In the wild

Canva's SEO engine

Canva's biggest initial growth loop was long-tail programmatic SEO targeting 'make' keywords and 'template' keywords — any design project someone might search for, like a wedding invitation template.

Built an incredibly strong, defensible moat off long-tail organic traffic.

Grammarly's non-viral reality

Grammarly is not an inherently multiplayer task, so network effects were never there. It leaned on performance marketing instead during part of its lifecycle.

Grew into a household name via paid rather than forcing virality.

Common mistakes

Chasing virality because every founder wants it

Founders default to 'yes, virality, that's how I'll grow' but network effects are either present at inception or can't be manufactured — pursuing them in a single-player product wastes runway.

Assuming admired companies are diversified

From the outside you assume top companies have highly diversified growth engines; in reality one engine usually drives 80%+ of growth.

Is it for you?

Best for

Founders of subscription products (consumer or prosumer) deciding where to concentrate limited growth resources

Not ideal for

Businesses with no self-serve motion or transactional/one-off products where subscription loop dynamics don't apply

From the transcript

there's kind of like these three engines that you can tap into and i imagine the preference would be word-of-mouth virality and then if that…

13:00

it's very it's very hard to manufacture uh you know product network effects if they aren't there from the get-go

14:00

their biggest uh initial growth loop and i think this is public knowledge uh was their long tail seo strategy

12:30

From the episode

How to grow a subscription business

Yuriy Timen (Grammarly, Canva, Airtable)