Skip to main content
New user request Sign in Visit humanforce.com

Ask AI Best Practices

Ask AI lets you explore your Humanforce Analytics data using everyday language. The way you phrase a question shapes the answer you get back, so a few small adjustments to your prompts can make Ask AI much more useful.

This guide covers the prompting habits that produce the most accurate, relevant answers.

Before you start

  • You need access to Humanforce Analytics on the Analytics PRO product tier.
  • You need access to at least one dataset (such as Talent, HR, Workforce Management (Coming Soon!), Benefits (Coming Soon!), or Learning (Coming Soon!)) to query.

Use natural, everyday language

Ask AI is designed to understand questions the way you would ask a colleague. Write prompts the same way you would speak them.

  • Ask "What were our total hires last quarter?" rather than "SELECT COUNT hires WHERE quarter = Q1".
  • Use the words your team uses day-to-day. Ask AI will match those words to the columns and measures in your dataset.

If Ask AI doesn't understand a term, rephrase the question using a more common word, or use the actual column name from the dataset.

Be specific with column names and values

Ask AI works best when your prompt points to something concrete in the data.

  • Use real column names or values when you know them. For example, "Show turnover by department" is clearer than "Show people leaving by team" if Department is the column in your dataset.
  • Spell out the metric you want. "Average tenure in months" is more precise than "How long do people stay?".

The clearer your reference to the underlying data, the more accurate the answer.

Filter your queries

You can ask Ask AI to focus on a slice of the data rather than the full dataset. Filters work well for time periods, locations, departments, or any other dimension in your data.

Example prompts:

  • "What was my Active FTE for August 2025?"
  • "Show me Days to Hire for Sydney based positions this year."
  • "What is my annualised attrition rate for the Sales department right now?"

Combine multiple filters in a single prompt to narrow the result further.

Rephrase when Ask AI misreads your question

After you send a prompt, Ask AI summarises how it interpreted your question. Check this summary before reading the answer.

If the interpretation doesn't match what you meant:

  1. Read the summary to see which part Ask AI misunderstood.
  2. Rephrase the prompt, swapping ambiguous words for the actual column names or values.
  3. Submit the revised prompt.

A small change in wording often produces a noticeably different answer.

Ask follow-up questions in the conversation

Ask AI keeps the context of your conversation, so you can refine your analysis step-by-step instead of writing one long prompt.

Example flow:

  1. Start with a broad question: "What is my annualised attrition rate?"
  2. Add a follow-up: "Break this down by Location."
  3. Refine further: "Break down the Sydney location by Departments."

Each response shows the changes from the previous step, so you can see exactly how the result was adjusted. Follow-ups are useful when you want to drill into a result without starting over.

Related articles

  • Feature Overview: Humanforce Analytics
  • Using Dashboards

Related to: