Machine learning to transform time tracking
From time tracking and resource planning to roadmapping to project and program management, get the data you need to optimize how your team works.
OPPORTUNITY
Kickdrum’s prototype detected missing timesheet entries and eliminated manual data input using natural language processing.
RESULTS
Accurate timesheets drive customer value
“The quality of timesheet entries is at the base of everything else in the suite of Tempo products, and us improving it will drive up the value our customers get from using our product.” - VP Product, Tempo
Saving engineers time with automated timesheet suggestions
The Tempo real-time activity event stream captures a users Jira interactions, calendar, code editing, and source control events. Using advanced NLP and temporal correlation techniques, matching tickets are presented to the user as suggestions for tracking time spent.
Extensive research and experimentation to enhance accuracy
Kickdrum explored several NLP and ML approaches, including LSTM, LDA, XGBoost, SVM, Naive Bayes, and cosine similarity using trained feature vectors and embeddings. The system achieved over 70% accuracy based on the users’ acceptance rate of timesheet suggestions.
“Kickdrum’s expertise was invaluable in bringing machine learning to our product.”
Mark Lorian
CEO, Tempo Software
