Valker

CASE STUDY · WEB APP · SAAS · AI

Inciteful

Search is becoming an answer. This measures whether you are in it.

A working SaaS platform that tracks whether AI assistants name a brand, and generates the content to change the answer when they do not.

  • Node.js
  • PostgreSQL
  • REST API and CLI
  • WordPress and Shopify publishing
  • Stripe
Visit the live site

The brief

People increasingly ask an assistant instead of searching, and an assistant returns one answer rather than ten links. That changes the question a business has to ask from where do I rank to am I the name it says. Inciteful is a full product built to answer it: it tracks whether ChatGPT, Google AI Overviews and Perplexity name a brand for the questions its buyers actually ask, shows which sources those engines trust instead, and generates content built to be cited.

What we did

  • Full SaaS product built from scratch: tracking, reporting, content generation and billing
  • Citation tracking across the three surfaces that matter, checking whether a brand is named for a given buyer question and recording when that changes
  • Source intelligence showing which domains the assistants actually cite for those questions, so the gap is visible rather than guessed at
  • Content generation aimed at being quotable, with each page scored on the structure an assistant can extract
  • White-label client reporting, so an agency can show the work under its own brand
  • Per-client workspaces, because agencies run this across a book of business rather than one site
  • Native publishing into WordPress and Shopify, with a REST API and CLI for everything else
  • Deliberately honest measurement: the product records what changed and does not promise a citation, because nobody can

The numbers

3 surfaces
ChatGPT, Google AI Overviews and Perplexity tracked
White-label
Agencies report under their own brand
Full product
Tracking, generation, reporting and billing

What we said no to

  • Promising rankings or citations. The product measures change and says so plainly, because guaranteeing an AI answer is not something anyone can honestly sell.
  • A single-brand tool. Agencies run this across many clients, so per-client workspaces were the starting point rather than an upgrade.
  • Locking content inside the platform. Publishing goes out natively or through the API, because content nobody can move is content nobody will trust.

Common questions

  • You have to ask it repeatedly and record the answers, because the result varies between runs. Inciteful automates that: it checks a set of real buyer questions across ChatGPT, Google AI Overviews and Perplexity on a schedule, and records whether the brand was named and which sources were cited instead.

  • SEO wins a position in a list of links. AI citation wins the answer itself, and the answer usually names one or two businesses rather than ten. Being on page one of Google does not mean an assistant will name you, which is why it needs measuring separately.

  • No, and any tool claiming otherwise is overselling. Model output is non-deterministic and changes between runs and versions. What can be done honestly is measuring whether a brand is named today, tracking whether that improves, and building the structure and evidence that make a business easier to cite.

  • Because agencies run it across a whole book of business, not one brand. Each client needs its own tracked questions, its own competitor set and its own report, and the report needs to carry the agency's branding rather than the tool's.

Stack

  • Node.js
  • PostgreSQL
  • REST API and CLI
  • WordPress and Shopify publishing
  • Stripe
NEXT PROJECTThe Church Tattoo

Want something like this?

30 minutes, no pitch deck.

WhatsAppGet a quote