¹ Dot is able to complete 80% of adhoc requests that would take an analyst less than 60 minutes to complete.

² The USD savings represents how much would have been spent on hiring analysts to serve this amount of questions. In most companies, the main advantage of Dot is that more stakeholders can get served, improving decision making across the company. These benefits are harder to quantify, but should usually be even higher (a decision that is made because of data can regularly affect revenue by up to 10%).

³ 3% of all employees work directly in analytics. 7% of all employees are power users who know SQL or have a deep understanding of the data model. Therefore 90% are stuck with static dashboards that can't answer a long tail of questions.

Formulas

1. Total questions per year = Number of employees × Questions per month × 12
2. Hourly rate = Yearly salary ÷ (230 days × 8 hours)
3. Human cost per question = Work time × Hourly rate
4. Total human cost = Total questions × Human cost per question
5. Total AI cost = Total questions × AI cost per question ($1)
6. ROI (USD) = Total human cost - Total AI cost
7. ROI (%) = (Total human cost - Total AI cost) ÷ Total AI cost × 100
8. Time reduction % = (Total human time - AI response time) ÷ Total human time × 100