Calculator Inputs
Use the responsive three, two, and one column input layout below.
Signal Visualization
The chart compares normalized signal strength across all factors.
Example Data Table
| Session | CTR % | Dwell Sec | Scroll % | Repeat Visits | Actions | Intent Score | Class |
|---|---|---|---|---|---|---|---|
| A102 | 8.4 | 124 | 88 | 4 | 3 | 82.60 | High Intent |
| B245 | 3.2 | 54 | 42 | 1 | 0 | 31.75 | Low Intent |
| C390 | 6.9 | 101 | 74 | 3 | 2 | 71.40 | High Intent |
| D411 | 1.8 | 28 | 24 | 0 | 0 | 16.55 | Very Low Intent |
| E527 | 5.6 | 89 | 67 | 2 | 1 | 58.90 | Moderate Intent |
Formula Used
This calculator uses a weighted behavioral scoring model. Each raw signal is normalized to a 0 to 1 range. The final score becomes a 0 to 100 intent value.
The probability output uses a logistic transform:
Why weights matter
Weighted scoring gives more importance to deeper actions. Micro conversions often predict stronger intent than passive browsing. Recency and repeat visits also improve practical ranking quality.
Normalization logic
Percentage metrics are divided by 100. Dwell time is capped at 180 seconds. Repeat visits are capped at five. Recency scores decrease as inactivity grows.
How to Use This Calculator
- Enter behavioral metrics from your analytics or model features.
- Use percentage fields on a 0 to 100 scale.
- Enter dwell time in seconds.
- Enter repeat visits and micro conversion counts.
- Set recency as hours since the last meaningful session.
- Score query specificity, source quality, and device fit.
- Click the calculate button to generate intent results.
- Review the score, class, confidence, and action priority.
- Use CSV or PDF export for reporting and sharing.
Frequently Asked Questions
1. What is a signal intent score?
A signal intent score estimates how strongly a visitor appears ready to act. It combines measurable behavior, such as clicks, dwell time, and conversions, into one practical score.
2. Is this a machine learning model?
This page uses a weighted scoring framework, not a trained model. It mimics common feature engineering logic and helps teams test assumptions before building a production classifier.
3. Why are micro conversions weighted heavily?
Micro conversions often show stronger intent than simple page views. Actions like signups, downloads, or add-to-cart events usually represent more deliberate interest.
4. Can I change the weights?
Yes. You can edit the weight values in the code. Teams often tune them using historical conversion data, feature importance, or business priorities.
5. Why does recency improve the score?
Recent activity usually indicates fresher intent. A user who engaged a few hours ago is often more relevant than one who last visited several days earlier.
6. How should I interpret the probability output?
The probability converts the score into an easier decision signal. It is not a guaranteed conversion rate. It simply expresses score strength on a smoother scale.
7. Can this calculator support lead routing?
Yes. You can map score bands to actions. High scores can go to sales. Mid scores can enter nurture flows. Lower scores can receive awareness campaigns.
8. Is this suitable for ecommerce and SaaS?
Yes. The logic works across ecommerce, SaaS, education, and content funnels. You only need to align the signals and thresholds with your own business behavior.