Back to Playbook
Workflow

Analyse My
Data

Clean, combine, and analyse project datasets using AI.

Turn messy files and folders into clean datasets, insights, risks, and next actions.
Primary action

Start with the right prompt

Use this when...

You have project spreadsheets, tracker exports, folder extracts, or multiple files that need cleaning before decisions are made.

File upload, folder selection, and link paste are interface placeholders. Use an approved Copilot or Microsoft 365 environment, then copy a prompt below.
Your brief (optional)
Data source
Analysis type
Copy prompt and use in Copilot
Guided Workflow

From Raw Data to Decisions

Use the same flow for Excel files, exported logs, tracker lists, and project folders.

01

Understand Data

  • Identify datasets
  • Detect structure
  • Highlight inconsistencies
02

Clean & Combine

  • Standardise columns
  • Merge datasets
  • Remove duplicates
03

Generate Insights

  • Trends
  • Risks
  • Key findings
Copy-ready Prompts

Run the Analysis in Copilot

This is the working part of the page. Copy the prompt, attach approved files, and review the outputs before sharing.

Prompt

Project Data Analysis

You are a senior project data analyst supporting a Project Manager. Context: I am reviewing project cost and schedule performance across multiple datasets. Your task is to analyse the attached file(s) and produce decision-ready insights. Steps: 1. Understand datasets (structure, columns, inconsistencies) 2. Propose how datasets should be combined 3. Clean and standardise data 4. Identify anomalies, missing values, duplicates 5. Analyse trends and risks 6. Provide: * 5 key insights * 3 key risks * 3 recommended actions Rules: * Do NOT assume joins * Do NOT invent data * Flag anything unclear Output: * Insights * Risks * Actions
Use It in Practice

Example Scenario

A practical view of the input, output, and delivery value.

Example Input

Three project files: cost forecast, schedule export, and change log for a rail design package.

  • Different work package names across files
  • Missing forecast values for two months
  • Duplicate change records from tracker exports

Example Output

  • Insight: forecast variance is concentrated in two work packages
  • Risk: schedule slippage may affect cost confidence
  • Action: confirm package mapping before reporting trends

Before vs After

Before

  • Manual file checks
  • Unclear joins
  • Slow insight drafting

After

  • Clear data issues
  • Faster trend view
  • Better PM decisions
Expected Outputs

What You Should Receive

Cleaned dataset

Combined data

Key insights

Identified risks/issues

Validation Checklist

Check Before Use

Next Actions

Move From Insight to Delivery

Take the cleaned evidence into the next project workflow.

Draft Report

Turn key findings into a concise report structure and evidence-backed narrative.

Open workflow

Create Presentation

Convert the insight set into a slide storyline for project or client discussion.

Open workflow

Review Project Risks

Use identified anomalies and issues to update the risk view and next actions.

Open workflow