Understanding Match Score Thresholds

A match score is a number from 1 to 100 that shows how likely it is that a match found by Demandbase is correct. It helps assess the quality of a matched record and is used by the Data Integrity, Enrich, and Refresh products to return accurate matches based on your input.

Important: A match score is not a direct percentage. For example, a score of 60 doesn’t mean there’s a 60% chance the match is correct. More accurately, a score of 60 corresponds to an accuracy rate of about 90%.

To set match score thresholds see Set Match Score Thresholds.

How Are Match Scores Calculated?

Match scores are based on both positive (additive) and negative (subtractive) factors. The following additive factors can raise or lower the match score threshold:

• The number of data fields you provide: Fewer input fields usually result in a lower match score. This is because the accuracy of a match depends on how many fields Demandbase can compare with its own data.
• The number of potential matches: A larger number of possible matches can lead to a lower match score. When there are many alternatives, it's harder to identify the correct one, which reduces confidence in the match.

If a company has many potential matches in the database, the match score is reduced in proportion to the number of similar companies that could dilute the accuracy of the result.

Why Use Match Thresholds?

When configuring Enrich, account administrators may set a match score threshold—i.e. a minimum match score below which enriched leads will not be sent into the marketing automation system. The appropriate match threshold for your use case will depend on your business requirements.

Since match scores reflect data accuracy, setting the match score threshold is essentially about deciding how many false positives you’re willing to accept in your lead supply.

In general, a lower threshold will result in more leads but with a higher chance of false positives, while a higher threshold improves accuracy but reduces the number of leads.

How to Set Match Score Thresholds

Data Integrity, Enrich, and Refresh allow system administrators to set match score thresholds, giving you control over the quality of data used to update and enrich your records. Demandbase uses these thresholds to group matches into the following categories:

Best Practices for Setting Match Score Thresholds

When setting match score thresholds, it’s important to understand the tradeoff between precision and match rate (recall). They work in opposite directions: increasing precision reduces the number of returned records (lower recall), while increasing recall may reduce precision.

Use the following guidelines to choose a threshold that fits your goals:

• Very high precision, decent recall: Set the threshold to 0.7

• High precision, high recall: Set the threshold to 0.5

• Decent precision, very high recall: Set the threshold to 0.3

Match Score  Benchmarks

This table shows the overall match score (also referred to as "confidence score") for various data matching results, based on a sample size of 100,000 records:

Use Case Example

A Demandbase customer in the tech industry wanted to fine-tune Enrich by setting a match score threshold that would balance accuracy with the level of enrichment.

To support this, Demandbase provided three options. The table below shows how different thresholds—70, 60, and 50—affect accuracy, false positives, and enrichment levels.