# Codag landing-page growth score

Lazyweb publishes a public search API over its indexed pages; the guide is at https://www.lazyweb.com/agent-access.

Source: https://www.lazyweb.com/yc-s26/codag
Source observed: 2026-09-30T02:38:35.784Z
Last checked: 2026-10-20T08:13:55.725+00:00
Indexed: 2026-09-30T06:54:34.959+00:00

---

# Codag landing-page growth score

We provide a Growth Score report for [Codag](https://codag.ai/) as part of Lazyweb’s YC S26 collection. Codag is evaluated as a product for coding agents; its landing-page headline is **“Cut the tool-output tax your coding agents pay.”** The report assesses how effectively the landing page communicates value and earns action across trust, information hierarchy, mobile web, value propositions, and calls to action.

## Overall result

Codag’s **Growth Score is 72 out of 100**, updated **September 20, 2026**. Its weakest category is **Trust**, scored at **62**.

| Category | Score | Rating shown |
|---|---:|---|
| Trust | 62 | Fair |
| Information hierarchy | 68 | Fair |
| Mobile web | 72 | Fair |
| Value props | 78 | Fair |
| CTA | 82 | Strong |

[Fix Trust](https://www.lazyweb.com/plans/checkout?plan=individual&source=growth_score_feature_build_fixes)

## How Codag ranks

The overall comparison ranks these products and sites:

1. [Kilo Code (Context Condensing)](https://www.lazyweb.com/growth-score/site/kilocode.ai) — 79
2. Codag — 72
3. [Not Diamond](https://www.lazyweb.com/growth-score/site/notdiamond.ai) — 62
4. [LLMLingua (Microsoft)](https://www.lazyweb.com/growth-score/site/llmlingua.com) — 55

## Score over time

We include a score-history visualization with series for Codag, Kilo Code (Context Condensing), LLMLingua (Microsoft), and Not Diamond. The accessible page identifies Codag as the highlighted series but does not provide the underlying history values as text.

## Trust — 62, Fair

Trust measures proof that a product is real, competent, and safe to try.

![Codag trust comparison capture](https://zlfyzdmohcskkucuunmk.supabase.co/storage/v1/object/public/sites/growth-score/site/codag.ai.jpg?v=2026-09-20T04%3A30%3A56.215%2B00%3A00)

The trust comparison reads the following signals:

| Product | Trust read |
|---|---|
| Codag | A **“Backed by Y Combinator”** badge appears above the hero headline. |
| Kilo Code (Context Condensing) | A **“Trusted by developers at the world's most innovative companies”** logo strip includes Meta, Amazon, Airbnb, PayPal, Square, and a partially unreadable final logo. |
| LLMLingua (Microsoft) | An author list includes **“Microsoft Corporation”** affiliation and the emails `{hjiang, qianhuiwu, xufang.luo}@microsoft.com`, plus a linked item whose remaining text is not readable in the supplied content. |
| Not Diamond | The page uses the metrics **5%+ accuracy gains**, **30%+ cost savings**, and **2x faster dev cycles**. |

Trust comparison scores:

1. Kilo Code (Context Condensing) — 78
2. Codag — 62
3. LLMLingua (Microsoft) — 62
4. Not Diamond — 42

[Build proof section](https://www.lazyweb.com/backlog?site=codag.ai#opt-trust-signals) with our proof-section builder to choose proof signals and receive copy and placement guidance for a stronger trust section.

## Information hierarchy — 68, Fair

This category asks whether the design helps or inhibits the landing page’s intended message.

![Information hierarchy annotations on the desktop capture](https://zlfyzdmohcskkucuunmk.supabase.co/storage/v1/object/public/sites/growth-score/site/codag.ai.jpg?v=2026-09-20T04%3A30%3A56.215%2B00%3A00)

1. **Duplicate CTA pairs:** “Start free” and “Book a demo” appear twice at equal weight—in the navigation and hero—splitting the primary action four ways.
2. **Left-panel log wall:** The “all tool output” panel renders approximately 20 lines of dense terminal text that competes with the headline during the first scan.
3. **Y Combinator badge:** The light-grey badge sits directly above the headline and has greater contrast, interrupting the scan path to the value proposition.
4. **Approximately 75% compression label:** The core proof metric is small and low contrast between the panels, making it easy to miss.

[View layout plan](https://www.lazyweb.com/backlog?site=codag.ai#opt-information-hierarchy) to turn these callouts into a sequenced keep, move, merge, or remove plan.

## Mobile web — 72, Fair

This category evaluates the phone experience as a first-class experience.

![Mobile web annotations on the mobile capture](https://zlfyzdmohcskkucuunmk.supabase.co/storage/v1/object/public/sites/growth-score/site/codag.ai-mobile.jpg?v=2026-09-20T04%3A30%3A56.215%2B00%3A00)

1. **Crowded header CTA pair:** Two CTAs beside the logo cause “Book a demo” to wrap onto two lines and duplicate the hero buttons below.
2. **Small hero CTA buttons:** “Start free” and “Book a demo” retain desktop proportions side by side instead of stacking as full-width buttons for thumb use.
3. **Truncated terminal log lines:** Desktop-density monospace output clips with ellipses in the phone column, making the demo unreadable.

[View mobile fixes](https://www.lazyweb.com/backlog?site=codag.ai#opt-mobile-web) for a prioritized implementation checklist.

## Value props — 78, Fair

This category evaluates whether the title, subtitle, and hero sell Codag’s true value.

### Codag’s headline

> **“Cut the tool-output tax your coding agents pay”**

What works:

- It names a real cost coding agents incur.
- The hero demo shows approximately **75% compression** in action.
- It uses specific numbers: **4,212 lines versus the failure**.

What could improve:

- There is no subhead explaining the coined phrase.
- “Tool-output tax” needs decoding from the diagram.
- The payoff in tokens, speed, and cost is never stated.

Value-proposition comparison:

1. Not Diamond — 80
2. Codag — 78
3. Kilo Code (Context Condensing) — 76
4. LLMLingua (Microsoft) — 62

## CTA — 82, Strong

This category evaluates CTA copy, size, verb, and perceived cost of clicking.

![Codag CTA comparison capture](https://zlfyzdmohcskkucuunmk.supabase.co/storage/v1/object/public/sites/growth-score/site/codag.ai.jpg?v=2026-09-20T04%3A30%3A56.215%2B00%3A00)

Codag’s visible CTA is **“Start free.”** The report identifies these problems:

- The navigation gives the demo CTA a green fill.
- The hero button is small compared with the headline.
- Emphasis is split with **“Book a demo.”**

CTA comparison:

1. Kilo Code (Context Condensing) — 86
2. Codag — 82
3. Not Diamond — 60
4. LLMLingua (Microsoft) — 52

## Competitors

The report infers competitors from product, audience, and value proposition:

1. [Kilo Code (Context Condensing) — kilocode.ai — 79](https://www.lazyweb.com/growth-score/site/kilocode.ai)
2. Codag — 72
3. [Not Diamond — notdiamond.ai — 62](https://www.lazyweb.com/growth-score/site/notdiamond.ai)
4. [LLMLingua (Microsoft) — llmlingua.com — 55](https://www.lazyweb.com/growth-score/site/llmlingua.com)

## Report actions

- [Build proof section](https://www.lazyweb.com/backlog?site=codag.ai#opt-trust-signals)
- [View layout plan](https://www.lazyweb.com/backlog?site=codag.ai#opt-information-hierarchy)
- [View mobile fixes](https://www.lazyweb.com/backlog?site=codag.ai#opt-mobile-web)

## Observation

This report was fetched at **2026-09-30T02:38:35.784Z**. The Growth Score itself is labeled **updated Sep 20, 2026**.

---

[Read this page’s summary](https://www.lazyweb.com/api/public/v1/page-tldr?url=https%3A%2F%2Fwww.lazyweb.com%2Fyc-s26%2Fcodag&trace=t_bc7e1c4e5b07)

## Ask a follow-up about this page

Need a more specific answer? Ask the exact question you have. It returns a fresh, cited answer from Lazyweb’s knowledge base. No account or API key required.

Open this URL with your own question, URL-encoded. Only q is required. No account or setup.

`GET https://www.lazyweb.com/agent/faq?q={question}&trace=t_bc7e1c4e5b07`

[Tool guide and examples](https://www.lazyweb.com/agent-access) · [OpenAPI](https://www.lazyweb.com/api/public/v1/openapi.json) · [Page TLDR form](https://www.lazyweb.com/agent-tools/page-tldr?url=https%3A%2F%2Fwww.lazyweb.com%2Fyc-s26%2Fcodag). Opening a form does not execute a tool.