Skip to main content

Documentation Index

Fetch the complete documentation index at: https://agents-docs.hockeystack.com/llms.txt

Use this file to discover all available pages before exploring further.

​
Overview

The Signals tab on a company’s detail page gives you a data-driven view of how engaged an account is, how strong their intent signals are, and what specific signals the AI has identified. It combines AI-generated summaries, scoring, and individual AI Property research results into a single view.

​
Accessing signals

  1. Navigate to the Companies page and click on a company.
  2. Open the Signals tab.

​
Signal Summary

At the top of the Signals tab, a summary card shows three AI-generated assessments:
AssessmentWhat it measures
FitHow well this company matches your target profile based on research and engagement, scored 0–10
TimingHow ready this company is to buy right now based on intent signals and momentum, scored 0–10
ContextA narrative explanation of the account’s current state and relevant factors
Each assessment includes a score visualization and a description explaining the reasoning. The summary is generated from the company’s AI Property signals, engagement data, and intent signals. The summary is cached and refreshes when underlying signal data changes.

​
Intent score

When scoring is enabled for your workspace, an intent score section shows the company’s current score with momentum indicators. Your workspace uses one of two scoring modes:
  • Delphi score — A proprietary intent score computed by HockeyStack’s scoring pipeline.
  • LLM intent score — An AI-generated intent score based on behavioral signals.
The active scoring mode is configured at the workspace level. Only one score type is displayed per account.

​
Momentum

Momentum indicators show how the score is trending:
  • Day-over-day — Score change compared to the previous scoring event.
  • Week-over-week — Score change compared to approximately 7 days ago.
Momentum helps you identify accounts with rising intent (good timing for outreach) versus declining intent (may need a different approach).

​
AI Property signals

Below the summary and score, individual AI Property results are displayed as signal cards. Each card shows:
  • Property name — The question that was researched.
  • Answer — The structured output (Yes/No, a dropdown selection, a score, or freeform text).
  • Description — A 1–3 sentence explanation of how the AI arrived at the answer.
  • Citations — Links to the evidence sources the AI used during research.
For Score type properties, a score meter visualization displays the numeric value on the 0–10 scale. For Dropdown type properties, the selected value is highlighted from the set of possible options.

​
How signals are generated

Signals come from multiple sources:
SourceWhat it produces
AI PropertiesStructured research results from your configured properties (see AI Properties)
Engagement dataActivity signals from your CRM, email, calls, and website tracking
Intent dataThird-party intent signals when available through connected integrations
Intent scoringIntent score and momentum indicators showing score trends over time
AI Property signals are generated when company context is built — either manually, by the overnight worker, or by auto AI research. Engagement and intent signals update as new data flows in from your integrations.

​
How signals connect to other features

FeatureConnection
Signal SummarySynthesizes all signals into Fit, Timing, and Context assessments
Prospecting task generationThe AI references signals when deciding task priority and content
Company AgentThe agent has access to all signals when answering questions
AI Properties configurationAll company-scoped AI properties are automatically researched during context generation

​
Constraints

  • Signal Summary generation requires at least one AI Property to have a researched answer for the company.
  • Intent score and momentum are only available when scoring is enabled for your workspace.
  • Signals are read-only on this page. To change what gets researched, edit your AI Properties in the AI Agent Builder.

​
FAQ

The summary requires AI Property data. If no properties have been researched for this company, generate context from the Agent tab or wait for the overnight worker to process the account.
The summary refreshes when underlying signal data changes. If a new AI Property answer is generated or engagement data updates significantly, the summary recalculates.
The signals shown depend on which AI Properties you have configured. Create new properties in the AI Agent Builder to add new signal types. See AI Properties for details.
Momentum requires at least two scoring events to compute a comparison. If the account was recently added or scoring was recently enabled, momentum data may not be available yet.
The property has not been researched for this company yet. Generate context from the Agent tab or enable auto AI research for the property to cover this company’s sales view.