Every insider purchase gets a score out of 100, built from three things: how much they spent, what else was going on, and how recently they bought.
Three more things are measured and shown on each signal but left out of the score — your own history with that insider, what the recent headlines say, and how the stock is trading today. None of them is a fact about the purchase, and all of them change on their own. That is why a signal reads the same for everyone and does not drift over time.
How much money the insider put in. Scored on a curve from $50,000 up to $50 million. Past $50M it maxes out — buys that large are rare enough that separating them further tells you nothing.
| Condition | Points | Label |
| ≥ $50 million | 20 pts | Saturated — top of the scale |
| ≈ $16 million | 17 pts | Very large |
| ≈ $5 million | 13 pts | Large |
| ≈ $1.6 million | 10 pts | Substantial |
| ≈ $500,000 | 7 pts | Moderate |
| ≈ $160,000 | 3 pts | Small |
| ≤ $50,000 | 0 pts | Below the ingest floor |
How this insider’s earlier buys have done in your watchlist. Shown for context, not counted — it is measured from your trades, so it would be different for every user.
| Condition | Points | Label |
| Avg return ≥ 150% | 20 pts | Exceptional |
| Avg return ≥ 80% | 17 pts | Strong |
| Avg return ≥ 30% | 14 pts | Solid |
| Avg return ≥ 0% | 10 pts | Modest |
| Avg return < 0% | 4 pts | Poor |
| No prior trades | 10 pts | Neutral (default) |
Who else was buying, and when. Other insiders in the same stock, the same insider buying repeatedly, and whether the purchase landed just before an earnings date. All of it is fixed by the filings, so none of it changes later. Adds up, capped at 20.
| Condition | Points | Label |
| CEO base score | 6 pts | Open-market purchase |
| Political base score | 8 pts | STOCK Act disclosure |
| Cluster buying | +4–12 pts | Scales with cluster size: 2×/3×/4×+ insiders |
| Insider accumulating | +3–6 pts | Same insider, 2–3 or 4+ buys in 90 days |
| Bought ahead of earnings | +3 pts | Purchase 1–21 days before an earnings date |
How recently they bought. Insider signals have a shelf life — last week means more than two years ago.
| Condition | Points | Label |
| ≤ 30 days ago | 20 pts | Very fresh |
| ≤ 90 days ago | 16 pts | Recent |
| ≤ 180 days ago | 12 pts | Moderately aged |
| ≤ 365 days ago | 7 pts | Aging |
| ≤ 730 days ago | 3 pts | Old |
| > 730 days ago | 1 pt | Historical |
Keywords picked out of the stock's recent headlines, plus whether earnings fall in the next two weeks. Shown for context, not counted. We read the last few headlines, so a catalyst that shows today can be gone next week — that is a fact about the news cycle, not about the purchase.
| Condition | Points | Label |
| Near legislative vote | +5 pts | Legislation / bill / committee in headlines |
| Record-size purchase | +3 pts | Largest / record / first time in headlines |
| Pre-event timing | +3 pts | Ahead of / days before in headlines |
| M&A catalyst | +3 pts | Merger / acquisition in recent headlines |
| Earnings catalyst | +2 pts | Earnings / guidance / beat in headlines |
| Earnings within 2 weeks | +2 pts | An earnings date in the next fortnight |
| Regulatory catalyst | +2 pts | FDA / approval / trial in headlines |
| Defense / government contract | +2 pts | Defense / Pentagon / contract in headlines |
| AI / tech catalyst | +2 pts | AI / datacenter in headlines |
| Upcoming announcement | +1 pt | Partnership / launch in headlines |
How the stock is trading today, plus analyst sentiment. Shown for context, not counted — it describes today’s market, not the filing, so it would change the score hour to hour. Useful for timing an entry.
| Condition | Points | Label |
| Day change ≥ +3% | 18 pts | Strong day |
| Day change ≥ +1% | 15 pts | Positive day |
| Day change ≥ 0% | 12 pts | Flat |
| Day change ≥ −1% | 8 pts | Slight pullback |
| Day change ≥ −3% | 5 pts | Selling pressure |
| Day change < −3% | 2 pts | Heavy selling |
| Analyst upside >20% | + 3 pts | Bonus: strong analyst consensus |
| Analyst upside >10% | + 1 pt | Bonus: moderate analyst upside |
| Analyst consensus<0 | − 2 pts | Penalty: analysts bearish |
| No live data | 8 pts | Neutral pending quote |
Insider's cost is what they actually paid. If they bought more than once in the last 90 days it is the average across all of those buys, weighted by size, so one filing does not misrepresent the whole position.
Analyst consensus is shown when analysts have published one more than 5% above the current price, with the number of ratings behind it. It is their figure, not ours. InsiderEdge does not forecast prices, so when there is no consensus the field is simply empty.
Risk level is a fixed 8% below the current price — a consistent reference across every signal, not a prediction.