Product

One growth engine. Your team and your agents work it together.

LogNorm finds what's holding growth back, ranks it into a weekly plan, and hands the work to people and to Claude Code, Codex, Cursor or Claude. Then it re-checks every fix and measures what moved.

Works withClaude CodeClaude CodeCodexCodexCursorCursorClaudeClaudeand any MCP client

The LogNorm platform at a glance

Signals
Site & GEO auditSearch ConsoleKeywordsCompetitorsAI answers
LogNormfinds · ranks · verifies
Ranked movesWeekly planCompany Brain
Your team
Plan, review, publish
Claude CodeCodexCursorClaudeAgents
Fix, write, research
Proof
Fixes verified liveMeasured at 28 and 90 days

Every shipped move is measured in Search Console and AI answers, and the ranking learns from it.

Why it matters

Growth stalls on the backlog, not the ideas.

01

The backlog never gets done.

Broken pages, missing metadata, content gaps, competitors outranking you, ChatGPT recommending someone else. Hundreds of items, and nobody has the hours.
02

Agents can do the work, but they fly blind.

Claude Code and Codex can fix a site and write a post. Alone they don't know what's broken, what people search for, or what worked last time.
03

LogNorm is the missing layer.

The data, the ranked priorities, the playbooks and one shared place to work. Every claim, comment, draft and fix shows up under the agent's name.
Humans and agents, in parallel

You plan. Agents ship. LogNorm proves it.

Nobody waits on anybody. The same weekly plan drives your team, your agents and the measurement.

You and your team
Decide, review, publish
  1. 01Pick this week's moves from the ranked plan
  2. 02Review drafts, PRs and comments on the dashboard
  3. 03Publish to your CMS in one click
Your agents
Claude Code, Codex, Cursor, Claude
Claude CodeCodexCursorClaude
  1. 01Claim a move and fix it in your repo
  2. 02Draft posts with research, images and a cover
  3. 03Research competitors and run audits on request
LogNorm
Finds, ranks, verifies, measures
  1. 01Audits the site and tracks 3 AI engines
  2. 02Ranks every signal into Moves
  3. 03Re-checks fixes live, measures at 28 and 90 days
3 AI engines
ChatGPT, Gemini and Google AI Overviews, tracked in your country.
12 destinations
Publish to WordPress, Ghost, Webflow, Shopify, Notion and more.
146 actions
Dashboard actions agents can take, plus 53 agent tools.
Proof built in
Fixes re-checked on the live site. Outcomes at 28 and 90 days.
How it works

The loop, step by step.

Six steps, repeated every week. Here is what happens at each one and which part of LogNorm does it.

  1. 1.Discover

    Discovery runs when you ask and on the schedules you set. A site and GEO audit, Search Console, keyword research, competitor analysis and AI answer tracking each look for a different kind of gap, and each gap they find becomes a Move with its evidence attached.

  2. 2.Rank

    Instead of scoring each move on its own, LogNorm compares open moves in pairs and fits one order from the results. You can read why any move sits where it does, and your own “rank higher” or “rank lower” is kept on every pass.

  3. 3.Plan

    Tell LogNorm how much your team can ship in a week. It proposes a plan from the top of the ranking, you accept, swap or snooze, and the plan becomes the list everyone works from.

  4. 4.Create

    Content moves become briefs built from what currently ranks, then drafts written from your Company Brain. A live scorecard grades the draft and checks its claims against what your company has actually said.

  5. 5.Publish

    Publish to WordPress, Ghost, Webflow, Contentful, Sanity, Strapi, HubSpot, Shopify, Storyblok, Notion, GitHub or a webhook. Images and metadata go with the page, and the move is marked shipped.

  6. 6.Measure

    Every shipped move keeps a before and after from Search Console and AI answer checks. Results at 28 and 90 days show what worked, and that evidence shapes how the next moves are ranked.

The platform

Explore each part in depth.

Every part works for your team in the dashboard and for your agents over MCP.

Put your growth backlog to work.

Start free and get your first ranked moves while you finish onboarding. Then hand them to your agents: they claim, fix and write while your team reviews.

Already use an agent?

Paste this into Claude Code, Codex, Cursor or Claude. It connects itself; you click Allow once.

Claude CodeClaude CodeCodexCodexCursorCursorClaudeClaudeand any MCP client