Predict every behavior
Remarkable customer experiences produce remarkable customers.
Likelihood to buy, convert, churn, and more
How likely is the customer to exhibit a behavior or take an action that you’ve seen before?
# LEAD SCORING
# First register your data to produce event streams
curl https://api.faraday.ai/datasets --header 'Authorization: Bearer $YOUR_API_TOKEN' --json '{
"name": "Signup",
"identity_sets": {
"shipping": {...}
},
"options": {
"type": "hosted_csv",
"upload_directory": "signup_data_files"
},
"output_to_streams": {
"signup": {
"data_map": {
"datetime": "created_at"
},
"value": "total"
}
}
}'
curl https://api.faraday.ai/datasets --header 'Authorization: Bearer $YOUR_API_TOKEN' --json '{
"name": "Transaction",
"identity_sets": {
"shipping": {...}
},
"options": {
"type": "hosted_csv",
"upload_directory": "transaction_data_files"
},
"output_to_streams": {
"transaction": {
"data_map": {
"datetime": "created_at"
},
"value": "total"
}
}
}'
# Now organize your customer data into cohorts
curl https://api.faraday.ai/cohorts --header 'Authorization: Bearer $YOUR_API_TOKEN' --json '{
"name": "Leads",
"stream_name": "signup"
}'
curl https://api.faraday.ai/cohorts --header 'Authorization: Bearer $YOUR_API_TOKEN' --json '{
"name": "Customers",
"stream_name": "transaction"
}'
# Next, declare your prediction objectivesfalse
curl https://api.faraday.ai/outcomes --header 'Authorization: Bearer $YOUR_API_TOKEN' --json '{
"name": "Likelihood to convert",
"attainment": "$CUSTOMERS_COHORT_ID",
"eligible": "$LEADS_COHORT_ID"
}'
# And finally complete your pipeline to deploy
curl https://api.faraday.ai/scopes --header 'Authorization: Bearer $YOUR_API_TOKEN' --json '{
"name": "Lead scoring",
"population": {
"include": [
"$LEADS_COHORT_ID"
]
},
"payload": {
"outcome_ids": [
"$LIKELIHOOD_TO_CONVERT_OUTCOME_ID"
]
}
}';
Dynamic prediction
Patterns that predict behavior in the early stages of a customer journey eventually stop working and new ones take over. Only Faraday uses time-linked model ensembles to tell the true story.
Algorithmic freedom
Faraday will detect and select the best tool for the job, from time-tested classics to cutting-edge GenAI.
Everything you need to predict customer behavior, built-in
All the power, flexibility, and data you need to ship fast.
Built-in consumer data
Get more accurate predictions—and avoid the dreaded cold-start problem—with 1,500+ consumer attributes on nearly 240 million adults.
AI safety
Explainability, privacy, bias management, and transparency: find your balance between power and fairness.
Automated feature engineering
In addition to its built-in consumer data, Faraday can find patterns in whatever first-party data you have for even more accurate predictions.