---
schema_version: "1"
id: 652462152f58174c6e65312cb39c2aa6ab4abf5e2c06b6e0c686a5932b495fe5
title: Improve the Quality of Numeric Predictions (Beta)
collection: admin
version: current
locale: en-us
portal_url: https://docs.salesforce.com/admin/current/en-us/sales/264-0-0/custom_ai_prediction_builder_quality_tips_numeric.md
canonical_url: https://help.salesforce.com/s/articleView?id=sales.custom_ai_prediction_builder_quality_tips_numeric.htm&release=264.0.0&type=5
source_site: help.salesforce.com
content_type: documentation
content_hash: aef285013dbf8722f16119de6980692ff92ce90aab2383a0e84b8659fc1999f5
description: When reviewing your prediction’s scorecard, you get a prediction quality score. If you’re not satisfied with the quality, here are some steps you can take to try to improve it.
products:
  - Sales
  - Sales Enablement
release: "264"
---
# Improve the Quality of Numeric Predictions (Beta)

When reviewing your prediction’s scorecard, you get a prediction quality score. If you’re not satisfied with the quality, here are some steps you can take to try to improve it.

If your prediction quality is Great….great! No need to try to improve it, but check your scorecard again from time to time to keep an eye on the quality. If the quality is Good when it’s running on live data (not estimated), you’re also probably all set. If, on the other hand, your prediction quality is OK, Too Low, or Too High, read on.

![Prediction Quality score](https://sf-zdocs-cdn-prod.zoominsoftware.com/tdta-sales-custom_ai-264-0-0-production-enus/40dfbcd7-3bfe-4d20-9d53-164eb414358b/custom_ai/images/einstein_prediction_builder_scorecard_quality.png)

First, notice whether the Prediction Quality title says “(Estimated)” or not. If so, it means that either you haven’t enabled your prediction yet, or it’s enabled but not yet running on live data. Either way, the quality score is estimated, based on a subset of your example data. It’s typical for the quality to go down after the prediction starts running on live data. For example, if your prediction quality is Great but estimated, check the scorecard again in about a month to see how it’s performing with live data.

- Collect more data. Certainly not all quality problems are due to the amount of data, but in some cases you can improve the quality simply by waiting until your dataset is more full.
- Try changing the way your data is segmented. For example, make sure your segment includes only records that are relevant to your prediction. If you have two distinct datasets that would each benefit from having its own prediction, create a separate prediction for each segment.
- Make sure your example set includes a representative range of numbers that you want to predict. Look for possible sources of bias in your data that might influence your prediction.
- Make sure you’re not excluding any fields that could provide additional, useful data.
- If your prediction quality is Too High when estimated (before it’s running on live data), it’s likely too good to be true. Review the fields included in your prediction. Do any of them look wrong, or contain data that couldn’t have been known until after the event being predicted? If so, exclude those fields. This type of problem is called hindsight bias. Also look over the Top Predictors on the scorecard. Do they make sense? If any of them look wrong, consider excluding those fields.
- Consider whether your use case is a good fit for a numeric prediction (Beta). Are you truly predicting a numeric value? Or is it just a number that represents a multiple choice value? For example, a field that assigns a number to each country as part of a picklist, isn’t truly a numeric prediction.