BetaSEOAnalyticsSaaS

Climbr

SEO keyword intelligence for agencies

Import thousands of keywords from Google Search Console, classify them with AI, and surface the patterns that matter - built for agencies managing multiple clients.

1 min read

Visit product

The problem

SEO agencies drown in keyword data. Google Search Console gives you thousands of queries per client, but no structure. Which keywords are branded? Which are transactional? Where are the content gaps? Answering these questions manually doesn’t scale past a handful of clients.

What Climbr does

Climbr imports your keyword data from Google Search Console and classifies it automatically. A hybrid engine combines deterministic rules (fast, predictable) with GPT-based classification (handles the ambiguous ones). The result: structured keyword intelligence across all your clients, without the manual spreadsheet work.

Key capabilities

Hybrid classification

Rules handle the obvious cases - brand terms, exact-match categories, known patterns. GPT handles everything else, with confidence scores so you can review edge cases. You get speed where it’s easy and intelligence where it’s hard.

Deduplication at scale

Large keyword datasets are full of near-duplicates: singular/plural, word order variants, URL-encoded noise. Climbr’s L1/L2 pipeline normalizes, canonicalizes, and fuzzy-matches keywords before classification - so you’re not paying to classify the same query three times.

Multi-tenant architecture

Each client gets fully isolated data with tenant-scoped queries throughout the stack. Designed for agencies managing 10+ accounts from a single dashboard, with per-client analytics and exports.

Search Console analytics

Automated GSC data sync with historical tracking, query-level metrics (clicks, impressions, CTR, position), and exportable reports for client deliverables.

Under the hood

  • Backend: FastAPI with PostgreSQL, tenant-scoped data isolation at the query level
  • Classification: Rules engine + OpenAI GPT with configurable confidence thresholds
  • Frontend: React + TypeScript with filterable tables, bulk operations, and CSV export
  • Infrastructure: Railway (API + database) + Vercel (frontend)

Status

In beta. Originally built as a client project for Be Frank (Copenhagen SEO agency), now being generalized into a standalone SaaS product.

All products

More from B3dmar

  • LiveOpen SourceGTM

    GTM:OS

    Open-source GTM workflow toolkit for AI assistants

    8 skills and 3 agents that turn a markdown pipeline file into a daily operating system - morning briefs, lead qualification, prospect research, and personalized outreach. No CRM required.

    PythonClaude Code SkillsCodexClaude Cowork
  • LiveOpen SourceAuth

    b3dmar-auth-core

    Authentication primitives for FastAPI services

    A shared library for JWT handling, password hashing, RBAC, and token revocation - extracted from production services and designed to drop into any FastAPI project.

    PythonFastAPIRedisArgon2id+2
  • LiveDeveloper ToolsAI

    sp3cmar

    Structured development framework for AI coding agents

    A skill library and review framework that brings discipline to AI-assisted development - specs before code, reviews before merge, accountability throughout.

    PythonClaude Code SkillsGitHub Actions

Want this kind of engineering on your team?

The same practices behind these products are available as consulting. Tell us what you are building and we will tell you honestly whether we can help.

Get in touchAll products