Making Life Easier with Chat GPT, Copilot and Gemini

Business are introducing Microsoft Copilot, Chat GPT and Gemini into their day to day lives

Microsoft Copilot is a great example of how anyone can make a difference to their daily productivity by using Copilot to assist with mundane tasks within the Office application suite and BI tools such as Power BI.

Chat GPT is perhaps one that people are most familiar with and is not only useful for asking questions and finding out information, but can also be used support many day to work tasks such as generating summaries, analysing charts, creating code.

Gemini is great for uploading photos, voice recordings and text to carry out analysis, summaries and content generation for you.

AI search engines have gone live such as Google AI Mode. Let’s see what Google AI Mode has to say about Copilot, Gemini and Chat GPT!

Using Microsoft Copilot To Summarise Email Chains

Have you tried using Copilot to summarize a lengthy email chain in Outlook? Brilliant time saver!

The following screenshot shows a concise summary of this email which happened to be a news article about closing the AI gap with Copilot in Windows:

I asked Copilot to draft a summary of the whitepaper and it came up with this:

Using Copilot to Summarise a Word Document

Here is an example of Copilot panel from Word. Whilst writing an article for our December newsletter I asked Copilot to summarise my document to see if I was hitting the right notes:

Using Copilot to Summarise a Power BI Dashboard Page

The following screen shot shows a demo Power BI report and a supporting Copilot panel where Copilot has been asked to summarise the report page:

We are going to be talking about Microsoft Copilot in our January newsletter and introducing some new Copilot training courses.

Using Chat GPT to create a tutorial from a recorded Teams call transcript

It is very common for training, coaching and mentoring sessions to be given via Teams meetings.

Meetings can be recorded and the transcripts of the meetings can then be uploaded to Chat GPT and a supporting training reference document can be created from the transcript.

Using Chat GPT To Build a Power BI Report From an Image

I gave Chat GPT an image of a P&L Chart and asked for assistance in identifying required measures and the DAX behind them:

The full transcript of what Chat GPT came back with is detailed below.

Chat GPT Transcript

Below is a clean breakdown of what measures you need for the chart and recommended DAX that follows standard P&L modelling best-practice in Power BI.

Because I can’t see your data model, I’ll use generic table/column names that you can adapt:

• ‘Fact P&L’

  = your transaction table

• ‘Account’

  = dimension containing Account Group (Sales, Purchases, Direct Expenses, Overheads)

• ‘Date’

  = date dimension

• ‘Budget’

  = budget table (optional)

✅ 1. Core Measures Needed

Your P&L chart shows five key items:

1. Sales

2. Purchases

3. Direct Expenses

4. Overheads

5. Total

…and each compared to Previous Year, Budget, and Forecast, depending on slicers.

So you need:

A. Base Measures

1. Total Amount

Total Amount :=

SUM ( 'Fact P&L'[Amount] )

B. Account Group Measures

For each group, filter the Account table.

2. Sales

Sales :=

CALCULATE (

[Total Amount],

'Account'[Account Group] = "Sales"

)

3. Purchases

Purchases :=

CALCULATE (

[Total Amount],

'Account'[Account Group] = "Purchases"

)

4. Direct Expenses

Direct Expenses :=

CALCULATE (

[Total Amount],

'Account'[Account Group] = "Direct Expenses"

)

5. Overheads

Overheads :=

CALCULATE (

[Total Amount],

'Account'[Account Group] = "Overheads"

)

C. Total (Net Profit before tax)

Total P&L :=

[Sales] + [Purchases] + [Direct Expenses] + [Overheads]

(If Purchases/Expenses are positive in your model but visually negative, you may need to switch the sign.)

✅ 2. Time Intelligence Measures

Your chart compares:

✔ Current period ✔ Previous year ✔ Variance

6. Previous Year Total Amount

Total Amount PY :=

CALCULATE (

[Total Amount],

SAMEPERIODLASTYEAR ( 'Date'[Date] )

)

…and repeat for each group:

Sales PY := CALCULATE ( [Sales], SAMEPERIODLASTYEAR('Date'[Date]) )

Purchases PY := CALCULATE ( [Purchases], SAMEPERIODLASTYEAR('Date'[Date]) )

Direct Expenses PY := CALCULATE ( [Direct Expenses], SAMEPERIODLASTYEAR('Date'[Date]) )

Overheads PY := CALCULATE ( [Overheads], SAMEPERIODLASTYEAR('Date'[Date]) )

Total P&L PY := CALCULATE ( [Total P&L], SAMEPERIODLASTYEAR('Date'[Date]) )

7. Variance to Previous Year

Total Amount Var PY :=

[Total Amount] - [Total Amount PY]

And for every group:

Sales Var PY := [Sales] - [Sales PY]

Purchases Var PY := [Purchases] - [Purchases PY]

Direct Expenses Var PY := [Direct Expenses] - [Direct Expenses PY]

Overheads Var PY := [Overheads] - [Overheads PY]

Total P&L Var PY := [Total P&L] - [Total P&L PY]

✅ 3. Percentage of Sales (for bottom chart)

Your second chart shows spending as a % of sales.

8. Purchases % of Sales

Purchases % of Sales :=

DIVIDE ( [Purchases], [Sales] )

9. Direct Expenses % of Sales

Direct Expenses % of Sales :=

DIVIDE ( [Direct Expenses], [Sales] )

10. Overheads % of Sales

Overheads % of Sales :=

DIVIDE ( [Overheads], [Sales] )

✅ 4. Budget / Forecast Measures (if applicable)

11. Budget Amount

Budget Amount :=

SUM ( 'Budget'[Budget Amount] )

12. Variance vs Budget

Variance to Budget :=

[Total Amount] - [Budget Amount]

Forecast follows the same pattern.

✅ Summary of Measures You Need

If you want, I can also:

✅ Build a ready-to-use P&L matrix ✅ Provide waterfall-ready measures ✅ Design the exact bar chart configuration from the screenshot

Just tell me which part you want next!

Using Gemini to Identify Ingredients and Calories From an Image

I had a lovely Carbonara on Saturday in an Italian restaurant and wanted to know what the likely ingredients and calorie values were likely to be. I uploaded a photo to Gemini and asked it to analyse it for ingredients and calories.

Here is the request and the initial analysis

Here is the identified ingredients:

And here is the nutritional information:

Using Gemini to Identify Objects and How To Use Them

I uploaded a picture of a tape dispenser to Gemini and asked what it is and how I should use it:

It correctly identified it and gave me instructions on how to use it:

Using AI to Analyse, Summarise and Advise Needs Care

AI does not always get things right. When you use AI tools you will see small disclaimers at the bottom of the results pages.

From Google AI Overview:

From Gemini:

From Chat GPT:

From Microsoft Copilot:

An Example of a Gemini Mistake

I uploaded a rather cute phot of my dogs to Gemini and asked Gemini to identify the breed.

My dogs are actually Beagle crosses, but Gemini identified them as Labradors based on analysing the photo I uploaded.

I tried a number of photos of them sitting, standing, but Gemini could not identify them as Beagles.

When I uploaded this photo Gemini was able to identify some other possibilities which included Beagles:

This is the transcript Gemini offered:

Based on their physical characteristics visible in the image, the dogs appear to be Labrador Retrievers.

Here's why they match the breed standard:

Coat: They have the short, dense, and sleek coat that is characteristic of a purebred Labrador Retriever.

Color: Their coloring is a very pale cream or off-white, which is a recognized shade of a Yellow Labrador. This specific, very light color is sometimes referred to as "white" or "cream" Labrador.

Build: While resting, their overall body and head structure—a strong, sturdy build and a broad skull—align well with the Labrador breed.

Ears: They have the signature wide, floppy ears that hang close to the side of the head, a trait shared by many sporting and hound dogs, including Labs.

Other Possibilities

It's also possible they could be a different breed or a mix, such as:

Lemon Beagles: These dogs can also be white and tan/yellow, but Beagles are a much smaller breed (20-30 lbs) than Labradors (55-80 lbs), and the dogs in the photo appear larger.

Beagle-Labrador Mix (Labbeagle): This crossbreed is possible and can result in a dog with a color similar to a Labrador but potentially a smaller size.

However, the size and overall structure lean most strongly toward the Labrador Retriever breed.

Would you like to compare the typical adult size of a Labrador to a Beagle?

Summary of Making Life Easier With AI Tools

We are all going to use AI tools more in 2026. There are definitely benefits in using tools such as Chat GPT, Gemini and Copilot in our everyday home and work lives:

• Speed

• Timesaving

• Questions & Answers

• Summary Documentation

• Instructions

• Identification

• Research

These are to name but a few.

So be brave and jump in to the world of AI tools if you haven't already done so, but be mindful of the following:

• Be careful what you share and upload - think of confidentiality and security

• Sanity check results and be certain that the results are accurate and trustworthy, especially if they are to drive important decision making.

• Remember that the more detail you share the more accurate the results will be

If you need help with your AI journey, whether it be learning how to use Copilot, Chat GPT and Gemini, or planning at a more strategic level for rolling out AI tools in a secure and safe way with proper data governance controls, do get in touch. We offer both AI training courses and AI consultancy services to get you started.