Data entry ops · Production

Ancestry uses Labelbox to achieve weekly ML model iteration cycles for genealogical data

The problem

Ancestry's data science team needed more efficient ways to decode census data and train ML models faster. Domain experts had strong knowledge of historical documents but limited insight into model behavior, making it hard to connect their expertise to the labeling process. Getting data labeled and reviewed was extremely slow.

First attempt

On other labeling platforms, the process was opaque — teams had to wait for all labels to come back before reviewing or giving feedback, with no ability to intervene or clarify mid-process.

Workflow diagram · grounded in source
1
Census data labeling need
trigger
“looking for more efficient ways to decode census data and to train their ML models faster, shifting from solely building models to taking a more data-centric approach”
2
Model-assisted labeling
ai_action
“leveraging the latest in labeling automation and collaboration including model-assisted labeling”
3
Domain expert review
human_review
“Finding easier ways to involve their domain experts during the labeling and review process was essential because it helped unlock the insights of these subject matter experts who have a wealth of knowledge in looking at historical documents”
4
Real-time labeler quality evaluation
validation
“The ability to evaluate how labelers were annotating data with analytics and in-depth metrics further helped the Ancestry team to be able to communicate in real-time and correct labels as needed”
5
Inline collaborative feedback
feedback_loop
“our team is able to collaborate more efficiently by dropping a pin on any image [asset], where we'll have a question and we can respond to our labelers in line. For other platforms, the labeling process is a complete black box as we have…”
6
Model training and testing
output
“train and test new models in record time”
Reported outcome

With Labelbox, Ancestry achieved a weekly model iteration cycle and can now maintain training data quality, save time collaborating with domain experts, and train and test new models in record time.

Reported metrics
Model iteration cycleweekly iteration cycle
Model training and testing timerecord time
Collaboration time with domain expertssave time
Reported stack
LabelboxBoost labeling services
Source
https://labelbox.com/customers/ancestry-customer-story
Read source ↗

Frequently asked questions

What did this team achieve with this AI workflow?

With Labelbox, Ancestry achieved a weekly model iteration cycle and can now maintain training data quality, save time collaborating with domain experts, and train and test new models in record time.

What tools did this team use?

Labelbox, Boost labeling services.

What results were reported?

Model iteration cycle: weekly iteration cycle; Model training and testing time: record time; Collaboration time with domain experts: save time (source-reported, not independently verified).

What failed first in this deployment?

On other labeling platforms, the process was opaque — teams had to wait for all labels to come back before reviewing or giving feedback, with no ability to intervene or clarify mid-process.

How is this data entry ops AI workflow structured?

Census data labeling need → Model-assisted labeling → Domain expert review → Real-time labeler quality evaluation → Inline collaborative feedback → Model training and testing.

Related data entry ops cases