What is Age & Gender Inference?

An in-depth look at Age & Gender Inference: what it is, how it works, and what data it uses.


What is Age & Gender Inference? 

Age & Gender Inference enriches missing donor records with an inferred age and gender for every single contact in a nonprofit’s database. 

Inferred Ages & Genders appear as highlighted purple text in Contact & List Views in the Dataro App.

Age & Gender Inference in List View:

Screenshot 2024-02-15 at 1.35.53 PM

Screenshot 2024-02-15 at 1.44.49 PM

Age & Gender Inference in Contact View: 

Screenshot 2024-02-15 at 1.45.58 PM


Age & Gender Inference leverages government, giving, and the (most importantly) the Dataro Data Pool to provide an inferred age & gender for every donor in a Dataro customer’s database,

  • Age is rounded to the nearest 5th year (e.g. 26->25, 48->50). 
  • Gender is listed as Male, Female, or Unknown

What Data is used to infer age & gender?

This comes down to basically three pools of data:

1. Government & Population Data: we are able to look at local name and age distribution to gather data on how likely a person with NAME is likely to be a given age and gender.

Primarily, this looks like utilizing the US Social Security name-age database


2. Dataro Client’s Donor Data
  • From records that are complete, we can gather age & gender trends among a client’s donors (e.g. 25% of known donors in a client’s database are 20-30, so it could follow that 25% of ALL of their donors are 20-30)
  • Individual giving behavior can also reflect trends in age and gender (e.g. method of giving, amount given, # of donations, and how long a donor have been giving can be used to infer age)


and finally..

3. Dataro’s Data Pool:  Dataro’s Age & Gender inference is the only tool with age & gender enrichment that is built with fundraising data as a foundation.

While our client's data is missing ~75% of ages, our Data Pool has age for ~25% of records (>25M individuals). This provides more than sufficient data to train a machine learning model. 

How does Age & Gender Inference work? What factors are taken into account? 

We use a lot of factors to infer age & gender. You'll see them below. 

  • Name Age Distribution based on Government & Population Data (e.g. XXX amount of Collins were born in 1993)

  • Organization Age Distribution (typical ratios of donor ages for that org)

  • Donor email, phone number, and address
  • Salutation
  • Individual Giving Behavior 
    • method/channel of giving,
    • amount given, 
    • # of donations, 
    • how long the donor has been donating 
  • Name/age lookup from the Dataro Data Pool


Are Inferred Ages Exact? How accurate is the Age Model?

Inferred Ages are not exact. We round up to the nearest 5th year.

Clients can treat age as an indicator that a donor is likely within a 5 year range on either side of an Inferred Age.

On average we are predicting within +/- 11.2 years of the donors true age. 


What if there isn’t enough data to infer gender?

If our model is not reasonably confident, the Gender field will read Unknown in the donor record. 


Are inferred ages & genders reflected in a customer’s CRM?

No. Inferred ages & genders are only reflected in the Dataro App (for now).


What other Prediction models are trained using Inferred Age & Gender?

Nearly all Dataro Predict models could benefit from Inferred Age & Gender. 

At present, we are using Inferred Age & Gender to train Gifts in Will & Major Giving models.