Skip to content

Insights + MCP

Your training data, in your AI assistant.

Connect Closelly with a compatible assistant, such as ChatGPT or Claude, and discuss your organization's courses, learners, and progress.

Compare results, identify outstanding work, and prepare follow-up without starting every analysis from scratch.

  • Which courses need attention?
  • Who still has work outstanding?
  • What should I bring to the next meeting?

Example questions

MCP in action

From a question to a better-informed decision.

See how to check course progress, explore outstanding work, and prepare a summary for your team through conversation.

Demonstration recreated with fictional data. The recommendations shown do not execute actions automatically.

Read the video description and transcript

Description Duration: 60 seconds. Spanish audio and embedded subtitles. Demonstration using fictional data. The main video recreates ChatGPT connected to Closelly through MCP: it is not an actual recording and does not show client data. Fictional courses with completion rates of 80%, 70%, and 38% are compared. In Consultative Selling, 24 of the 62 learners who have not completed the course have not logged in to the platform for 7 days or more: 10 have not started and 14 are in progress. The follow-up plan is a suggestion; the assistant does not carry out actions. The opening and closing screens name ChatGPT, Claude, Gemini CLI, Copilot Studio, and other MCP-compatible AI agents, depending on their capabilities, permissions, and configuration. Transcript Your training. In your ChatGPT. Connect Closelly through MCP and access your organization's authorized data from your own assistant. Ask which course needs attention. In this example, Onboarding reaches eighty percent; Safety, seventy; and Consultative Selling, thirty-eight. ChatGPT turns the data into a clear comparison. Look deeper: twenty-four Consultative Selling learners have not completed the course and have not logged in to the platform for seven days or more. Ten have not started and fourteen are in progress. They account for twenty-four of the sixty-two learners who have not completed the course. Prepare the follow-up: contact these learners, review their barriers, and measure again in a week. The data guides you; you decide. Closelly, connected to ChatGPT and other MCP-compatible assistants. Your training data, in your conversation. Request a demo.

Practical examples

Less time searching. More clarity to act.

Which courses need attention?

“Compare completion across these courses and show me where more learners have not started.”

Get a comparative view and explore further wherever you need.

Who still has work outstanding?

“In this course, which learners have not started and which are still in progress?”

Distinguish different situations to prepare more relevant follow-up, according to the available data and permissions.

What should I bring to the next meeting?

“Summarize the state of training and highlight outstanding items we should review.”

Turn the queried information into a clear summary to discuss with your team.

AI INSIGHTS · WITHIN CLOSELLY

Understand what is happening. Decide where to reinforce learning.

Analyze progress, results, and learning difficulties from the back office, with dashboards and AI-assisted interpretation.

Closelly administration dashboard with courses, progress, and insights.
Closelly back-office view
  • Identify difficulties

    Recognize topics and questions that need reinforcement.

  • Understand progress

    Review participation, progress, and results by course.

  • Explore interactions

    Look more closely at questions asked of LENA, according to the enabled modules and permissions.

A PRACTICAL EXAMPLE

Which concepts cause the most difficulty, and which content should we review?

Review available results and recurring questions to identify content worth reinforcing. AI helps organize the analysis; the team validates the conclusions.

Insights: within Closelly.

Analyze with dashboards and AI-assisted interpretation.

MCP: from your AI assistant.

Ask questions, compare information, and prepare analyses with authorized Closelly tools and data.

Two complementary ways to use your data. The MCP connection has its own scope and does not automatically include all Insights features.

Access and control

Connect what you need. Keep control of access.

The connection uses an authorized account and queries the information enabled for its organization. Available tools and access to personal data or exports depend on the corresponding permissions.

Before making decisions, you can review the period, filters, and scope of the information used.

  • Authorized account
  • Enabled data
  • Access permissions

What would you like to ask your training data?

Bring a specific case. We will show you how to approach it with MCP and what you can analyze within Closelly using Insights.