Signal Ark assigns a score to every detected event before it reaches your dashboard. Raw signals — an unscored job posting or a news mention — aren’t useful on their own. The scoring model converts them into a ranked list of opportunities so your team spends time on the signals most likely to lead to a conversation, not the ones that happened to appear most recently.Documentation Index
Fetch the complete documentation index at: https://docs.signalark.app/llms.txt
Use this file to discover all available pages before exploring further.
The five scoring dimensions
Every signal is evaluated across five criteria. The results are combined into a single composite score from 0 to 100.Intent and budget relevance
Intent and budget relevance
Does this signal indicate actual buying intent, or is it ambient noise? Signals that directly suggest a company is in an active buying cycle — a Series B funding announcement, a job posting for a specific technology, a competitor review — score higher than signals that are weakly correlated with intent, like a generic LinkedIn company update.Budget relevance also factors in here. A funding round from Crunchbase scores higher than a blog post, because it’s evidence that capital is available to spend.
Actionability
Actionability
Can a sales rep use this signal to craft a specific, relevant outreach message? A signal is highly actionable when it gives a rep a concrete reason to reach out — for example, “I saw you’re hiring three DevSecOps engineers” or “Congratulations on the Series B.” A vague or indirect signal with no clear outreach angle scores lower on actionability, even if it suggests intent.
Freshness
Freshness
How recently did the event occur? Signals lose value as time passes. A funding round announced yesterday is more actionable than the same round from three months ago. Signal Ark applies a time-based decay function (described below) to reduce the freshness score as signals age.
ICP fit
ICP fit
How well does the account associated with this signal match your Ideal Customer Profile? ICP fit is calculated separately from the signal itself — it’s a property of the account, not the event. A strong signal from a weak-fit account is scored lower overall than the same signal from an Ideal-tier account.This dimension ties signal scoring directly to your ICP configuration. Keeping your ICP profiles accurate is the most effective way to improve the quality of your signal feed.
Compounding (stacking)
Compounding (stacking)
Is this an isolated event, or part of a broader pattern? When multiple signals occur at the same account within a short window — especially if they come from different sources — Signal Ark boosts the composite score to reflect the higher confidence that something is happening at that account.Ten job postings from a single job board are less meaningful than three signals spread across a job board, a news article, and a GitHub activity spike. Source diversity is weighted explicitly in the compounding calculation.
Priority bands
The composite score from 0 to 100 places each signal into one of four priority bands:| Band | Score range | What it means |
|---|---|---|
| High priority | 80–100 | Imminent buying intent. These signals trigger GTME play evaluation and surface in your Monday 8 prioritization dashboard. |
| Worth outreach | 60–79 | Solid rationale for contact. Good candidates for enrollment in a standard outreach sequence. |
| Monitor | 40–59 | Weak or early-stage signals. Signal Ark keeps them on the radar to see if intent compounds over time. |
| Low priority | Below 40 | Filtered out of standard views to reduce noise. Still accessible via the full timeline if needed. |
Signal decay
Intent is time-sensitive. A company that announced a funding round last week is in a very different mindset than one that raised 90 days ago. Signal Ark applies an automated decay curve to every signal score based on its source type, reducing the effective score as time passes. Decay is modeled as a half-life: after one half-life period, the signal’s freshness component is reduced by half. After two half-lives, it’s reduced by three-quarters, and so on.- Fast decay
- Medium decay
- Slow decay
Half-life: 3–7 daysSources: website visits, social mentions, Reddit threads, Twitter/X posts.These signals reflect real-time activity that loses relevance quickly. A tweet about evaluating a new tool is very actionable today and largely irrelevant in two weeks.
How scores are displayed
In the signal timeline and account views, you’ll see each signal’s composite score alongside a breakdown of the five dimensions. This breakdown helps you understand why a signal ranked where it did — and gives reps the context they need to personalize outreach.Signal timeline
See how scores are visualized in the account timeline view.
ICP scoring
Learn how your ICP profile configuration affects signal scores.