> ## Documentation Index
> Fetch the complete documentation index at: https://docs.allyy.io/llms.txt
> Use this file to discover all available pages before exploring further.

# Generating predictions

> Apply a trained model to a population to produce scores

Training teaches a model the patterns in your history. **Generating predictions** (also called *scoring*) applies that trained model to a population today, producing a score for every supporter in it.

## What a prediction is

Each supporter gets a score whose meaning depends on the model type:

| Model type                                       | Score meaning                                                        |
| ------------------------------------------------ | -------------------------------------------------------------------- |
| Propensity / churn / engagement (classification) | A **probability** from 0 to 1 — e.g. 0.82 = "82% likely to respond". |
| Expected amount / lifetime value (regression)    | An **amount** — e.g. an expected gift of €45.                        |
| Ask amount                                       | A suggested **set of ask values** for that supporter.                |

## Running a prediction

<Steps>
  <Step title="Choose the model">
    Pick a trained model. Allyy uses the model exactly as it was trained — the same features and logic — so predictions are consistent with the reported performance.
  </Step>

  <Step title="Choose the population to score">
    Point the model at the supporters you want scored — typically a **list** (e.g. everyone eligible for an upcoming appeal). Only supporters present in your mapped data can be scored.
  </Step>

  <Step title="Set the scoring date (optional)">
    By default Allyy scores as of **now**. You can also score as of a past date to back-test how the model would have ranked a previous campaign.
  </Step>

  <Step title="Run">
    Allyy scores the population and stores the results, ready to view, turn into a decision, or export.

    <Note icon="camera">**Screenshot to add** — a completed prediction run with example scores.</Note>
  </Step>
</Steps>

<Note>
  Predictions are **descriptive of likelihood, not instructions**. A score ranks supporters; deciding *how many* to contact and *with what* is the job of a [decision](/predictions/decisions).
</Note>

## Keeping predictions fresh

Scores reflect the data at the moment they were generated. Re-run scoring before each campaign — or schedule it with a [workflow](/automation/workflows) — so you're always acting on current behaviour.

## Next

<CardGroup cols={2}>
  <Card title="Turn scores into action" icon="bullseye" href="/predictions/decisions">
    Combine and optimise scores into a ranked contact list.
  </Card>

  <Card title="Check the model first" icon="chart-line" href="/dashboards-evaluation/index">
    Use the evaluation dashboards to confirm the model ranks well before you rely on its scores.
  </Card>
</CardGroup>