Skip to content
English
  • There are no suggestions because the search field is empty.

Getting Started — What to Ask Your Time Doctor AI Connection - Beta

Note: The AI connection (MCP) is available on all Time Doctor subscription plans and can be accessed by Owners, Admins, and Managers.

TL;DR:

Ask the connected AI assistant plain-language questions about Time Doctor data — hours, productivity, meetings, attendance, and more. Starter prompts below make the AI explain its own capabilities. Set team KPIs once so follow-up answers are measured against real targets. Answers are always limited to what the assigned Time Doctor role can see.

 

Start With These Prompts

Not connected yet? See Connect Time Doctor to Your AI Tools first.

These prompts make the connection explain itself. Try these before memorizing anything else:

  • "What Time Doctor tools (insights) are available, and what can each one tell me?" — lists everything the connection can access.
  • "Help me get started with my Time Doctor data. Ask me about my goals and build my first report." — the AI asks what's being managed, then builds queries around it.
  • "What am I not measuring that could be useful? Look at the data available on my plan and suggest insights I haven't asked about." — surfaces reports that may not be obvious, such as meeting load or schedule adherence.

 

Set Team KPIs First

The AI gives sharper answers once it knows what "good" looks like for a team. Set the context once at the start of a chat:

"For context: my team works 8-hour days, Monday to Friday, Eastern time. Idle above 15% is worth a look, and the target is 6+ productive hours a day. Keep this in mind for my questions."

Every follow-up ("who's off target this week?") is then measured against those numbers instead of generic ones.

 

The Four Levels of Questions

Level 1 — Reports, conversationally. Anything normally pulled from a report, asked in plain language.

"Top 10 apps and sites used by the Support team last month, and flag anyone well above the team average."

 

Level 2 — Cross-referenced insights. Questions that combine data the web app shows on separate screens.

"We have no-meeting Fridays. For Marketing in June, who actually kept Fridays meeting-free, and who just had lighter meeting loads?"

 

Level 3 — Visual reports the product doesn't build. On-the-fly charts and one-page summaries.

"Chart the team's scheduled shifts this week with actual tracked hours overlaid."

 

Level 4 — Time Doctor data plus other data. Combine with anything else the AI assistant can see.

"Compare closed deals from the CRM (or an attached file) against the sales team's tracked activity. Which work patterns line up with more wins?"

 

Reusable Question Patterns

Most useful questions follow one of five patterns. Swap in specific people, teams, numbers, and dates.

Pattern

Formula

Example

Threshold

Who is above/below [number]

"Who tracked under 6 hours a day last week?"

Comparison + why

[Team A] vs. [team B] + what's driving the gap

"Productivity summary for Ops vs. Sales this quarter, with what's driving the gap."

Schedule deviation

Who worked outside [schedule/hours]

"Who worked outside their scheduled hours or late at night in the last 90 days?"

Cross-check

Was [event A] explained by [event B]

"Was Ana on a call during her idle block on Tuesday?"

Data hygiene

What's missing or unassigned

"Show users with no group assigned."

Trends

How is [metric] changing over [period]

"How do hours tracked trend across the week? Is productivity up or down vs. last month?"

 

Sample Prompts to Try

Topic

Start simple

Go deeper

Hours tracked

"How many hours did the Sales team log this month?"

"Compare each person's hours this pay period vs. last, and flag anyone with gaps in logged hours."

Productivity

"What's the Support team's productivity % this week?"

"Drill into the three teams with the lowest productivity this quarter and explain what's driving it."

Apps & websites

"Which apps and websites get the most time?"

"Break down time by app category company-wide. Are people using AI tools, and which ones?"

Idle & activity

"What's the team's idle time this week, and who had the most?"

"Where is tracked-but-idle time concentrated: which people, teams, or days of the week?"

Meetings

"How much time went to meetings, and what percentage of work was that?"

"Show meeting time by team and flag anyone whose meeting load looks like a burnout risk, aggregated, not by individual."

Projects & tasks

"Which projects took the most time this month?"

"Give project and task time totals for the last pay period, formatted for export."

Attendance & schedules

"Who didn't track any time today? Who's late?"

"List schedule issues and conflicts for the team to catch before payroll runs."

Leave & time off

"Who was on leave this month, and how much has Carlos taken?"

"Show leave for the team, pending vs. approved, and flag approved leave that conflicts with scheduled shifts."

People & teams

"List the people on the Sales team."

"List everyone Ana manages, and show any users with no group assigned."

Unusual activity

"Show unusual activity flags for this month."

"Give a per-user unusual activity breakdown for Ana, with the days worth reviewing."

Trends

"Show productivity over the last month. Is it trending up or down?"

"Summarize this quarter vs. last — hours, productivity, and meeting load — in a format suitable for a board update."

 

Note on attendance: "Who's late?" works when Settings / Schedules is set up; otherwise, include a start time in the question, or connect the AI to an external tool with schedule details.

 

These are starting points, not limits. Combine topics, phrase questions naturally, and follow up on any answer to drill in further.

 

Tips for the Best Answers

  • Name a date or range: "today," "last week," "June," or "between April 1 and 30."
  • Name the person or team.
  • Mention a timezone for day boundaries; answers default to the timezone set in the Time Doctor account.
  • If the AI doesn't use Time Doctor automatically, name it directly: "Using Time Doctor, show hours tracked this week."
  • Start broad, then narrow: "How is the team spending time?" then "Break that down by task for Ana."

 

Understanding Data Limits

To keep responses fast and reliable, the AI connection works within a few limits:

  • Group size: questions about a group cover up to 200 members at a time.
  • Date range: up to about one year per request; roughly two years of history is kept.
  • Amount of data: results return in batches. Detailed data, such as Web & App Usage, can be hundreds of rows per person per day.

 

For large analyses (for example, 500 users over a month), avoid requesting everything at once. Request a smaller batch (50 users, one week), let the AI return it, continue batch by batch, then ask it to combine everything into one analysis. If a request returns incomplete, narrow the range, reduce the user count, or ask for a summary ("top 10 apps") instead of every row.

 

 


 

 

Should there be any inconsistencies or concerns regarding the article, contact support@timedoctor.com for prompt assistance.