# AI Coding Pricing / Cost Hub — Market & Keyword Opportunity Report

Task: t_937a8ac9
Project: aicodingpricing.com
Target: US / English
Date: 2026-05-21
Verdict: 值得做，P0 直接进入 PRD。不是靠泛词 `ai coding cost calculator`，而是靠品牌 pricing / usage / limits / vs 对比词切入；calculator 是转化工具，不是 SEO 主入口。

## 1. Executive judgment

结论：A- / 值得立即做。

原因：
- 已验证 DataForSEO Google Ads US/en 数据：核心词合计 83,000/mo；不是用户摘要里的 97.5K/mo。97.5K 若要成立，需要补充未列出的长尾词。
- P0 可索引入口约 69,200/mo，且 CPC 极高：`claude code vs codex` $166.84、`codex vs claude code` $93.36、`codex usage` $41.54、`codex pricing` $25.97。
- SERP 不是铁板：多个 P0 SERP 已有独立博客、小型内容站、Reddit/HN/GitHub 页面进入前 10，说明新站可进入。
- 官方页面占据 pricing 类头部，但官方页不解决“我该买哪个、会不会超限、Claude/Codex/Cursor 怎么换算”的决策问题；这就是 Hub + calculator + changelog 的机会。
- 最大风险不是建站，而是 freshness：pricing/limits 变动快。必须把 changelog、last checked、source evidence 做成核心能力。

不建议做的方式：
- 不要把首页主词押在 `llm pricing calculator` 或 `ai coding cost calculator`。
- 不要做品牌词域名。
- 不要用 logo/官方视觉冒充关联关系。
- 不要做静态一次性 pricing 表，过期会快速失信。

## 2. DataForSEO validation

Source: DataForSEO Google Ads `keywords_data/google_ads/search_volume/live`, location_code=2840, language_code=en, search_partners=false.

| Keyword | Validated volume/mo | CPC | Competition | KD from input | 12m shape | Priority |
|---|---:|---:|---|---:|---|---|
| claude code pricing | 27,100 | $6.42 | LOW | 13 | rising: 3,600 May 2025 → 49,500 Jan-Apr 2026 | P0 |
| cursor pricing | 12,100 | $2.92 | LOW | 18 | stable/high: 9,900-18,100 range | P0 |
| claude code vs cursor | 8,100 | $17.46 | LOW | 0 | rising then slight pullback: 2,900 → 14,800 peak → 8,100 | P0 |
| codex vs claude code | 8,100 | $93.36 | LOW | 14 | strong rising: 880 → 27,100 Apr 2026 | P0 canonical synonym |
| codex usage | 6,600 | $41.54 | LOW | 3 | breakout: 10 → 40,500 Apr 2026 | P0 |
| cursor vs claude code | 6,600 | $23.12 | LOW | 0 | rising: 590 → 12,100 Apr 2026 | P0 canonical synonym |
| claude code cost | 4,400 | $11.40 | LOW | 49 | rising but harder KD | P1 |
| claude code vs codex | 3,600 | $166.84 | LOW | 11 | rising: 590 → 9,900 Mar-Apr 2026 | P0 |
| codex pricing | 3,600 | $25.97 | MEDIUM | 5 | strong rising: 590 → 14,800 Apr 2026 | P0 |
| claude code limits | 1,300 | $15.76 | LOW | 28 | rising from low base, volatile | P1 |
| claude code usage limits | 880 | $20.11 | LOW | 26 | rising from low base | P1 |
| codex limits | 590 | $50.61 | LOW | n/a | rising: 40 → 2,400 Apr 2026 | P1 |
| llm pricing calculator | 30 | $3.32 | LOW | n/a | flat/no demand | P2 / support only |

Validation note:
- Exact listed keyword sum = 83,000/mo, not 97.5K/mo.
- P0 selected indexable + canonical cluster sum before dedupe = 69,200/mo.
- KD was supplied by upstream input; the Google Ads endpoint validates volume/CPC/competition, not KD.
- DataForSEO batch call returned null for several long-tail terms, but individual exact calls validated them. Use individual calls or smaller batches for final data refresh.

## 3. SERP spot checks and entry judgment

### 3.1 `claude code pricing`

Top SERP observed:
1. support.claude.com — Choose a Claude plan
2. claude.ai — Upgrade to Claude Pro
3. verdent.ai — Claude Code Pricing 2026: Real Costs
4. reddit.com — tracked Claude Code subscription costs
5. platform.claude.com — Claude API Docs pricing
6. ssdnodes.com — Claude Code Pricing in 2026
7. finout.io — Claude pricing for orgs/developers
8. code.claude.com — Manage costs effectively
9. YouTube — free setup angle

Entry judgment: enterable but needs differentiation.

Competitor pattern:
- Official pages dominate factual pricing and API cost.
- Third-party pages already rank (#3, #6, #7), proving non-official SEO entry is possible.
- Current third-party gap: mostly article-style, weak calculator/freshness/source-diff UX.

Attack angle:
- `last checked` table per plan.
- Pro vs Max vs API cost scenarios.
- Claude Code usage window / subscription vs API explanation.
- Real cost calculator: daily sessions, avg context, model, agent count.
- Changelog with official source links.

Risk:
- Official Anthropic pages will hold top spots.
- Need factual disclaimers and no official affiliation.

### 3.2 `codex pricing`

Top SERP observed:
1. developers.openai.com — Codex Pricing
2. chatgpt.com — Codex Pricing
3. reddit.com — Codex pricing discussion
4. news.ycombinator.com — Codex pricing / API token usage discussion
5. uibakery.io — OpenAI Codex Pricing
6. milvus.io — how much Codex costs
7. YouTube — Codex pricing explained
8. eesel.ai — Codex pricing guide
9. userjot.com — Codex pricing article

Entry judgment: enterable, high value, update-driven.

Competitor pattern:
- Official OpenAI pages top 2.
- Community pages rank high (#3/#4), indicating searchers want interpretation, not only official pricing.
- Third-party pages are mostly generic explainers.

Attack angle:
- Codex included-plan matrix: Free/Go/Plus/Pro/Business/Edu/Enterprise.
- Usage credit explainer + `/status` usage check.
- Codex CLI vs ChatGPT auth vs API key billing.
- Pricing changelog and promo expiry tracking.

Risk:
- OpenAI pricing and included usage can change quickly.
- Must show last updated and source snapshots.

### 3.3 `claude code vs codex`

Top SERP observed:
1. Medium/civillearning — tested both for 6 months
2. Reddit — Claude vs Codex discussion
3. nocodesaas.io — honest comparison for non-coders
4. composio.dev — built same app in both
5. build.ms — Codex vs Claude Code
6. datacamp.com — key differences
7. mindstudio.ai — which tool should you use

Entry judgment: highly enterable.

Competitor pattern:
- No official pages dominate.
- SERP is content-heavy: Medium, Reddit, small/mid blogs.
- There is no obvious calculator-first comparison page.

Attack angle:
- Canonical page `/claude-code-vs-codex`.
- Include cost calculator comparison: subscription limits, API key billing, task types.
- Include decision tree: local repo vs cloud sandbox, team workflow, PR review, privacy, cost predictability.
- Update monthly because models/pricing shift.

Risk:
- CPC is likely inflated/volatile, but even if overstated, commercial intent is real.

### 3.4 `claude code vs cursor`

Top SERP observed:
1. Reddit — used both for 30 days
2. ksred.com — Cursor vs Claude Code article
3. codeaholicguy.com — Claude Code vs Cursor
4. haihai.ai — Cursor Agent vs Claude Code
5. builder.io — Claude Code vs Cursor
6. Medium — comparison article
7. qodo.ai — team comparison
8. Indie Hackers — Claude vs Cursor app build

Entry judgment: highly enterable, but needs stronger original utility.

Competitor pattern:
- Reddit and small blogs rank high.
- Builder/Qodo have strong domain authority, but content angle is broad.
- Searcher intent mixes workflow comparison and pricing/limits comparison.

Attack angle:
- Canonical page `/claude-code-vs-cursor`.
- Cost-aware comparison: $20/$60/$100/$200 plan mapping.
- “Which plan should I buy if I code X hours/week?” calculator block.
- Include IDE-first vs agent-first decision matrix.

Risk:
- Cursor official pricing and product packaging change often.
- Need avoid shallow “feature table only” page.

### 3.5 `cursor pricing`

Top SERP observed:
1. cursor.com — official pricing
2. reddit.com — which Cursor plan is worth it / limits
3. cursor.com/students — official students
4. medium.com — pricing worth-it breakdown
5. vantage.sh — Cursor Pricing Explained
6. YouTube — pricing tutorial
7. nocode.mba — Cursor Pricing 2026
8. flexprice.io — Cursor pricing guide

Entry judgment: enterable but more crowded than Codex/Claude comparison pages.

Competitor pattern:
- Official page #1 and official student page #3.
- Multiple SaaS/content competitors already target the keyword.
- Searcher pain is pricing confusion and limit anxiety.

Attack angle:
- Treat as P0 because volume is high, but page must be very fresh.
- Do not just repeat official plan table.
- Add “Pro vs Pro+ vs Ultra break-even” and usage overage explanation.

Risk:
- Lower CPC than Codex/Claude terms.
- More mature SERP; harder than vs pages.

### 3.6 `codex usage`

Top SERP observed:
1. chatgpt.com — Codex Pricing
2. reddit.com — monitoring Codex usage
3. github.com/steipete/CodexBar — usage stats app
4. developers.openai.com — Codex use cases
5. HN — pricing aligns with API token usage
6. Medium/civillearning — Claude Code vs Codex
7. ccusage.com — Codex data source
8. uibakery.io — how to use Codex

Entry judgment: very enterable and strategically important.

Competitor pattern:
- SERP is fragmented: official, Reddit, GitHub app, HN, docs, blog.
- Strong evidence that users want usage monitoring, not just pricing.

Attack angle:
- `/codex-usage` should be a usage explainer + calculator + monitoring guide.
- Include CLI `/status`, plan limits, ChatGPT credits, API key mode, team monitoring.
- Link to `/codex-pricing`, `/codex-limits`, `/claude-code-vs-codex`.

Risk:
- Requires frequent updates due to plan/credit changes.

## 4. P0 / P1 / P2 keyword and page matrix

### P0 — build and index first

| Page | Primary keyword | Secondary keywords | Policy | Why |
|---|---|---|---|---|
| `/` | AI coding cost calculator | AI coding pricing, AI coding tool cost, Claude/Codex/Cursor cost comparison | index | Hub + tool entry; not expected to rank for broad volume immediately |
| `/claude-code-pricing` | claude code pricing | claude code cost, Claude Code Pro/Max/API pricing | index | 27.1K/mo; official + third-party SERP; main capture page |
| `/codex-pricing` | codex pricing | OpenAI Codex pricing, Codex API cost, Codex credits | index | 3.6K/mo, CPC $25.97; update-driven |
| `/codex-usage` | codex usage | Codex usage limit, Codex status, Codex credits | index | 6.6K/mo, CPC $41.54; fragmented SERP |
| `/claude-code-vs-codex` | claude code vs codex | codex vs claude code, Claude vs Codex pricing | index canonical | Highest commercial value; no official SERP lock |
| `/codex-vs-claude-code` | codex vs claude code | same as above | 301 or canonical to `/claude-code-vs-codex` | Avoid duplicate competing pages |
| `/claude-code-vs-cursor` | claude code vs cursor | cursor vs claude code, Claude Code Cursor pricing | index canonical | 8.1K/mo + high CPC; SERP enterable |
| `/cursor-vs-claude-code` | cursor vs claude code | same as above | 301 or canonical to `/claude-code-vs-cursor` | Avoid duplicate competing pages |
| `/cursor-pricing` | cursor pricing | Cursor Pro+ pricing, Cursor Ultra pricing, Cursor usage pricing | index | 12.1K/mo, mature SERP but large demand |

### P1 — build after P0 copy freeze or include as sections first

| Page | Primary keyword | Policy | Why |
|---|---|---|---|
| `/claude-code-usage-limits` | claude code usage limits | index | 880/mo, high CPC; can initially be section inside Claude pricing, then split |
| `/claude-code-cost` | claude code cost | index or canonical section to `/claude-code-pricing` initially | 4.4K/mo but KD 49; use as supporting page if enough unique copy |
| `/claude-code-limits` | claude code limits | index | 1.3K/mo; freshness-sensitive FAQ page |
| `/codex-limits` | codex limits | index | 590/mo but CPC $50.61; small high-intent page |
| `/codex-pricing-changelog` | Codex pricing changelog | index | update-tail capture, supports freshness moat |
| `/claude-code-pricing-changelog` | Claude Code pricing changelog | index | update-tail capture, supports freshness moat |
| `/pricing-changelog` | AI coding pricing changelog | index | Hub for source diffs/newsletter |
| `/api-vs-subscription` | API vs subscription for AI coding tools | index | Decision support; internal-linking page |
| `/codex-vs-cursor` | Codex vs Cursor | index after P0 | Likely relevant but not validated in this SERP pass |

### P2 — backlog / expansion only

| Page | Policy | Why |
|---|---|---|
| `/github-copilot-pricing` | index later | Strong official/brand competition; not part of validated P0 SERP check |
| `/github-copilot-vs-cursor` | index later | Good category fit; validate volume/SERP first |
| `/claude-max-vs-pro` | index later | Useful support page; likely lower volume but high conversion |
| `/chatgpt-pro-vs-claude-max` | index later | Adjacent buyer-intent page; validate before PRD expansion |
| `/llm-pricing-calculator` | noindex initially or canonical to `/` | Only 30/mo; not worth standalone SEO page at launch |

## 5. Competitor map

| Cluster | Official competitors | Third-party competitors observed | Small/independent entry? | Entry verdict |
|---|---|---|---|---|
| Claude Code pricing | support.claude.com, claude.ai, platform.claude.com, code.claude.com | verdent.ai, ssdnodes.com, finout.io, Reddit, YouTube | Yes: Verdent/SSD Nodes rank | Enterable with better freshness + calculator |
| Codex pricing | developers.openai.com, chatgpt.com | UI Bakery, Milvus, Eesel, UserJot, Reddit, HN | Yes: several content sites rank | Enterable with source-tracked pricing + usage explainer |
| Claude Code vs Codex | none dominant | Medium, Reddit, nocodesaas, Composio, build.ms, DataCamp, MindStudio | Yes: Medium/small blogs rank | Strongest SEO opening |
| Claude Code vs Cursor | none dominant | Reddit, ksred, codeaholicguy, haihai, Builder, Medium, Qodo | Yes: several blogs rank | Strong opening if content is original and cost-aware |
| Cursor pricing | cursor.com | Reddit, Medium, Vantage, NoCode MBA, Flexprice, YouTube | Yes but crowded | Worth doing, harder than comparison pages |
| Codex usage | chatgpt.com/developers.openai.com | Reddit, GitHub CodexBar, HN, ccusage, UI Bakery, Medium | Yes: GitHub/ccusage rank | Very enterable; product utility angle strong |

## 6. Product and SEO positioning

Recommended positioning:

"AI Coding Pricing is a fresh, source-linked cost hub for Claude Code, OpenAI Codex, Cursor, and Copilot. Compare subscription tiers, usage limits, API costs, and break-even scenarios before you upgrade."

Must-have differentiators:
1. Freshness system
   - `Last checked` timestamp per provider/page.
   - Source URL per pricing row.
   - Changelog entries for pricing/usage/limit changes.
   - Newsletter capture: “Get alerts when AI coding prices or limits change.”

2. Calculator system
   - Inputs: tool, plan, coding hours/week, agent sessions/day, team size, local/API mode, heavy model usage.
   - Outputs: expected monthly cost, plan recommendation, upgrade threshold, risk of hitting limits.
   - Comparison mode: Claude Code vs Codex vs Cursor.

3. Fair-use compliance
   - Brand pages are factual and comparative.
   - Prominent disclaimer: not affiliated with Anthropic/OpenAI/Cursor/GitHub.
   - No unauthorized logos unless legally cleared; use text names or neutral icons.

4. SEO-Copy Freeze before design
   - Freeze final H1/H2/H3, FAQ, schema, target word count, internal links, and canonical policy before visual design.

## 7. Suggested first IA

P0 internal-link structure:

- `/` links prominently to:
  - `/claude-code-pricing`
  - `/codex-pricing`
  - `/cursor-pricing`
  - `/claude-code-vs-codex`
  - `/claude-code-vs-cursor`
  - `/codex-usage`
  - `/pricing-changelog`

Each pricing page should link to:
- Official source page
- Related usage/limits page
- Relevant vs page
- Calculator block on homepage or embedded calculator
- Changelog page

Each vs page should include:
- Pricing summary table
- Usage/limits comparison
- Workflow fit matrix
- Cost calculator block
- FAQ schema
- canonical self URL

## 8. SERP risks

High risks:
- Official pricing pages will own top positions for exact pricing facts.
- Pricing/limits update frequency creates staleness risk.
- Brand/fair-use risk if copy or UI implies official affiliation.
- CPC values for Codex comparison terms may be volatile; treat as directionally high commercial intent, not guaranteed ad monetization.

Medium risks:
- Cursor pricing is already crowded with strong content competitors.
- `claude code cost` has KD 49 in input; should be P1, not P0 standalone dependency.
- Duplicate vs synonyms can cannibalize unless canonical/301 is strict.

Low risks:
- MVP implementation complexity is manageable if calculator starts deterministic/static.
- SERP already accepts independent pages and community content.

## 9. Recommended action

Proceed to PRD with this scope:

P0 launch package:
- Home hub + calculator
- Claude Code pricing
- Codex pricing
- Codex usage
- Cursor pricing
- Claude Code vs Codex
- Claude Code vs Cursor
- Pricing changelog
- Canonical/301 handling for reversed vs slugs

Do not expand to Copilot/ChatGPT Pro/Claude Max pages until P0 copy and freshness system are locked.

## 10. Sources checked

Data:
- DataForSEO Google Ads search volume live API, US/en, exact keywords listed above.

SERP:
- DataForSEO Google Organic SERP live advanced API, US/en desktop, depth 10, checked for: `claude code pricing`, `codex pricing`, `claude code vs codex`, `claude code vs cursor`, `cursor pricing`, `codex usage`.

Official/current pricing source pages extracted:
- https://support.claude.com/en/articles/11049762-choose-a-claude-plan
- https://code.claude.com/docs/en/costs
- https://developers.openai.com/codex/pricing
- https://cursor.com/pricing

Representative third-party pages extracted:
- https://www.verdent.ai/guides/claude-code-pricing-2026
- https://www.ssdnodes.com/blog/claude-code-pricing-in-2026-every-plan-explained-pro-max-api-teams/
- https://uibakery.io/blog/openai-codex-pricing
- https://www.vantage.sh/blog/cursor-pricing-explained
- https://www.nocode.mba/articles/cursor-pricing
- https://flexprice.io/blog/cursor-pricing-guide
- https://composio.dev/content/claude-code-vs-openai-codex
- https://www.datacamp.com/blog/codex-vs-claude-code
- https://www.mindstudio.ai/blog/claude-code-vs-codex/
- https://www.builder.io/blog/cursor-vs-claude-code
- https://www.qodo.ai/blog/claude-code-vs-cursor/