# Build brief — a focused alternative to Profound

> **Verdict:** Partly, if you narrow it · **Buildability:** 46/100 · **Category:** SEO Marketing
> **Source:** https://www.canitbevibecoded.com/profound
> Independent editorial assessment from Can It Be Vibe Coded? Not affiliated with, endorsed by, or derived from Profound. Verify current pricing and capabilities before acting.

## Context

**Profound** — Tracks how ChatGPT, Perplexity, and AI Overviews describe your brand, and what AI crawlers do on your site. It currently costs $99/mo.

The tracking loop is genuinely straightforward to build: send a fixed prompt set through four model APIs on a schedule, count brand and competitor mentions, normalize the citations, and parse your access logs for AI crawler and AI referral traffic. That gets you most of the dashboard for one brand. The gaps are the honest part. API answers are not the answers the ChatGPT app or Google AI Overviews actually serve, you cannot measure what real people ask AI, and a number with no history behind it tells you nothing on week one.

This brief describes a focused, single-operator replacement for the part of Profound that is genuinely reproducible. It is deliberately narrower than the product it replaces, and it says so in writing. Build the useful core; do not pretend to have rebuilt the rest.

## What you are building

Run a fixed prompt set daily through the answer engine APIs, score brand and competitor mentions plus cited sources, and parse server logs for AI crawler and AI referral traffic.

- Automate a bounded research or reporting workflow using permitted data sources.
- A responsive interface with real empty, loading, success, and error states.

## Requirements

### Functional

- Each provider's web search or grounding tool.
- Server access logs.
- Durable per-run storage.
- A scheduler for daily runs.

### Data and integrations

- OpenAI, Anthropic, Gemini, and Perplexity API keys.
- An API budget that scales with prompts x engines x days.

Each of these needs a real account, credential, or quota. Set them up before writing feature code.

### Non-functional

- Accessibility: semantic markup, labelled controls, visible focus, and reduced-motion support.
- Security: server-side secrets, validated input, and no credentials in the client bundle.
- Reliability: retries with backoff on external calls, and a clear failure state when a provider is down.
- Portability: the operator can export their data and leave without losing it.

## Implementation brief

Build me a local AI answer engine visibility tracker for one brand. Requirements:

- Node 22, TypeScript, SQLite via better-sqlite3, a CLI, and a plain server-rendered
  dashboard. Local only, no accounts, no telemetry.
- brand.json holds my brand name, aliases, domain, and competitor names. prompts.json
  holds up to 40 buyer questions.
- `track run` sends every prompt through OpenAI, Anthropic, Gemini, and Perplexity with
  each provider's web search or grounding tool enabled. Keys live in .env.
- Store one immutable row per run, prompt, and provider: raw answer, cited URLs, model
  id, latency, and error text. Never overwrite an existing run.
- Cap concurrency at 3 per provider, retry twice on 429 and 5xx with backoff, and keep
  failed cells visible in the report instead of dropping them.
- Detect brand and competitor mentions case-insensitively using the alias list, and
  record the first-mention character offset as a crude prominence proxy.
- Score the sentiment of each brand mention in one cheap structured pass over stored
  answers, after the run, never inline.
- Normalize citations to hostname plus canonical path, strip tracking parameters, then
  compute owned-domain citation share and a top 25 sources table.
- `track serve` renders visibility per provider over time, share of voice against each
  competitor, the sources table, and the prompts where competitors appear and I do not.
- `track crawlers --log access.log` parses server logs for GPTBot, OAI-SearchBot,
  ClaudeBot, PerplexityBot, Google-Extended and friends, plus referral hits from
  chatgpt.com and perplexity.ai, and reports which URLs they touched.
- Keep the bot user agent list in an editable JSON file. Log lines matching nothing get
  counted as unknown agents, not silently discarded.
- `track export` writes runs, mentions, and citations to CSV.
- Fixture tests for mention detection, URL normalization, and log parsing.
- Out of scope: real consumer surface answers, prompt volume estimates, content
  generation agents, teams, and hosted scheduling. Do not scrape the consumer web UIs.
- README: setup, per-run cost estimate, a cron line for daily runs, and a plain note
  that API answers only approximate what users actually see.

## Delivery standard

- Inspect the repository first, then write a short implementation plan before writing code.
- Deliver the smallest complete end-to-end workflow first; every primary control must work against persisted data.
- Use real validation and storage; never substitute fake dashboards, decorative controls, hard-coded success states, or mock integrations.
- Include responsive layouts plus genuine empty, loading, success, validation, and failure states.
- Keep secrets server-side in environment variables, provide .env.example, and never commit credentials or user data.
- Add structured logs around every external call and return actionable errors without leaking sensitive details.
- Write unit tests for the core logic and one automated test of the main user journey.
- Finish with a README covering setup, architecture, data location, backups, tests, deployment, and known limitations.

## Acceptance criteria

- [ ] A clean install starts the app using only the README and .env.example.
- [ ] The primary journey works from first visit through saved result, reload, edit, export, and deletion where applicable.
- [ ] Invalid input, missing configuration, provider failure, and an empty database each have a usable state.
- [ ] The interface works at 390px and 1440px, is keyboard navigable, and shows visible focus on every control.
- [ ] Tests, type checking, linting, and a production build all pass with no ignored failures.
- [ ] No part of the interface implies a live integration, security guarantee, or scale capability that was not actually built and verified.

## Non-goals

Do not build these, and do not claim to have replaced them:

- Prompt volume data: what people actually ask AI is not measurable from outside.
- The real consumer surfaces, since AI Overviews and the ChatGPT app have no API that matches what users see.
- Months of history and competitor baselines, without which a single week's visibility number means nothing.
- Upkeep as engines, crawler user agents, and citation formats keep changing.
- The agent, recommendation, and product visibility layers stacked on top of the tracking.

## What you still own after launch

- Secure credentials, rotate secrets, and handle provider rate limits.
- Run migrations, backups, restores, and dependency updates.
- Test the critical journey after every model, API, or hosting change.
- Monitor failures and fix the edge cases a first prompt will miss.
- Maintain every third-party integration as APIs and OAuth rules change.

## Risk

**Operational risk.** The code is achievable; dependable data, integrations, and ongoing operations are the real cost.

Editorial confidence in this assessment: medium. No reviewed project implementation is linked yet.

## Existing alternatives

Before building, compare these checked options:

- [Elmo](https://www.elmohq.com/) — A real self-hosted AI visibility dashboard; crawler-log analysis is the conspicuous thing it does not replace

## Prior art

Working open-source software you can read, fork, or borrow from before starting:

- [Elmo](https://github.com/elmohq/elmo) — MIT-licensed self-hosted AEO/GEO tracker: runs your prompts across the major answer engines and records mentions, competitors, and cited sources
- [llm-brand-tracker](https://github.com/sarahkb125/llm-brand-tracker) — Small research-grade toolkit for monitoring brand visibility in LLM search; useful as a starting point, last commit mid-2025

---

Generated by [Can It Be Vibe Coded?](https://www.canitbevibecoded.com) · Full report: https://www.canitbevibecoded.com/profound