SignalForge is GEOforge's measurement module. It tracks your AI Share of Voice in ChatGPT, Google AI Overviews and Google AI Mode, and reports every number with its margin of error.
What SignalForge is
SignalForge is the measurement layer of GEOforge. On the Dominate plan it runs every tracked prompt 40 times a week on each of ChatGPT, Google AI Overviews and Google AI Mode, which gives 1,200 answers per engine across 30 prompts and a margin of error of about ±1.5 points on a brand's weekly Share of Voice. It opens the closed loop by finding gaps and closes it by re-measuring the lift.
Where it sits in the closed loop
Measure, ground, create, cite, then re-measure. SignalForge closes the loop by proving what moved Share of Voice, and the next turn starts from that evidence.
Ask an AI engine the same question twice and you will often get different brands. A single check has a margin of error of ±9–13 points, wider than almost any real weekly movement.
SignalForge asks enough times for the number to mean something, and shows the uncertainty instead of hiding it. Confidence ranges use the Wilson score interval, the standard method for rare yes/no rates.
You generate, edit and monitor 30 prompts that represent the questions buyers ask AI when they are looking for a solution like yours.
The SoV Measurer runs each prompt 40 times a week on each engine and grades every answer for mentions and citations of your brand and up to 10 competitors.
Share of Voice is reported weekly, per engine, with a 95% confidence range, so you can tell real movement from noise.
The Crawl Tracker shows which pages AI crawlers visit, which earn AI citations and which drive referral traffic, using Cloudflare, GA4 and Google Search Console.
Gaps and losing prompts go to BaseForge and ContentForge. After content and citations ship, SignalForge re-measures and attributes the lift.
ChatGPT, Google AI Overviews and Google AI Mode, on every plan.
Because AI answers vary between runs. Forty runs on each of 30 prompts gives 1,200 answers per engine each week, which brings the margin of error on Share of Voice to about ±1.5 points at typical visibility levels. A single check is off by ±9–13 points.
SignalForge is the read layer, but it does not stop at a dashboard. Its gaps feed BaseForge and ContentForge, CiteForge acts on the sources it finds, and SignalForge then re-measures whether those actions moved Share of Voice.
Start a 14-day trial with every module included and no credit card, or book a demo to see SignalForge, BaseForge, ContentForge and CiteForge working as one closed loop.