RavenPrompts
Home

Find your category

Create and communicate

  • Photography and image
  • Videos
  • Social media
  • Content creation
  • Marketing

Business and organisation

  • Sales
  • Business
  • Finance
  • Productivity

Learn and develop

  • Education
  • Programming

Life and leisure

  • Food
  • Sports
  • Travel
View all categories →
Prompt libraryFavoritesBlogPricingSuggest a prompt
← Back to the blog
Article

Duplicates in Excel: what to check before deleting a row

A nine-row case for choosing a record key, separating repeated data from conflicting amounts, and giving AI a review you can verify.

RavenPrompts editorial team · November 9, 2026

A repeated order number can be a valid line

You combine two exports and see the same order number several times. It looks like a quick cleanup, until you remember that an order can contain several items. Comparing the order number alone could discard a valid line. Today’s Reel introduces that distinction; here we work through the review.

Our nine rows were created for this case and contain no customer data. Each has a source row ID, order, line, item, quantity and amount in euros. The first output we want is a list of issues with reasons. A shorter table is not yet the goal.

Choose the key before counting matches

For this case, the combination of order and line number identifies an order line. P-2103 has lines 1 and 2, both valid. Another dataset might also need a store, year or revision. That rule comes from how your records are created; AI should not guess it.

Keep the source file and work on a copy. Add review columns alongside the received values. For these identifiers, we have agreed to ignore outer spaces and letter case: “ P-2101 ” is compared as P-2101, and p-2102 as P-2102. Do not apply that rule to case-sensitive codes or remove leading zeros without checking what they mean.

Nine rows, four different situations

R01 and R03 share order P-2101, line 1, item CAR-01, two units and €48. After the agreed normalisation, the displayed data repeats. Flag both rows to establish whether the same import was included twice. A match alone does not authorise deletion.

R02 and R05 share P-2102 and line 1, but their amounts are €18 and €20. This is a conflict to check against the source record. Choosing the larger or later amount, or averaging them, would change the data without explaining what happened.

R04 and R06 belong to P-2103, with lines 1 and 2 and amounts of €27 and €24: their complete keys differ. R07 shows €18 but has no order number, so it remains incomplete. R08, P-2104 line 1, shows €9. R09, P-2105 line 1, has zero units and €0. Zero is a recorded value, not a gap to fill.

Give AI a review task with a traceable result

Explain the columns, the chosen key, the permitted formatting changes and what must remain untouched. Ask for a review showing source row IDs, matching fields, differences and the next check. Identifying R02 and R05 lets you return to the source; “duplicates cleaned” does not tell you what changed.

RavenPrompts’ existing prompt for detecting duplicate Excel records can help frame this task. Start with a description of the structure or an authorised sample without personal information. Before sharing a work file, check your organisation’s and tool’s rules. Review any suggested formula or script and try it on the copy before applying it more widely.

Detect duplicate Excel records: view the prompt ↗

Choose an operation that preserves what you need

Microsoft distinguishes highlighting, filtering and removing duplicates. Conditional formatting lets you inspect matches without deleting them. Remove Duplicates removes records using the selected columns and keeps the first occurrence. The first row is not necessarily the correct business record. Microsoft recommends keeping a copy before removal.

COUNTIFS can count rows meeting criteria across several columns. Formula names and separators must match your Excel installation. Check the covered range as well: a calculation limited to the sample must not be described as covering the entire worksheet.

Microsoft — Find and remove duplicates ↗

Microsoft — Filter for unique values or remove duplicate values ↗

Microsoft — COUNTIFS function ↗

Reconcile the review before changing the source

This sample has four flagged rows in two pairs: one pair repeats data and the other has conflicting amounts. One row is incomplete and four have neither issue. That accounts for all nine rows. Four flagged rows do not mean four deletions: both occurrences in each pair are included.

Check that P-2103 retains both lines, R07 remains pending and R09 keeps its zeros. On a copy, change a key and add a row at the end, then check whether the result and covered range respond as intended. Record each decision, its reason and a reference to the original entry.

Recent Excel developments still need your judgement

The official Excel Blog described new Copilot data-preparation assistance on 12 October 2026 and updated the post on 16 October. That is recent context, not an announcement made today. Availability and limits depend on the account and rollout. You can practise this review with tools you already have access to.

Microsoft’s Copilot documentation warns that results can be inaccurate and need review. AI can help describe a rule or explain an issue; checking it still means returning to your records. When you close the file, you should be able to see what matched, what remains unknown and why each row was changed.

Excel Blog — Beyond Analysis: Copilot Gets to Work in Excel (12 October 2026; updated16 October) ↗

Microsoft — Frequently asked questions about Copilot in Excel ↗

Frequently asked questions

Why not compare only the order number?

An order can contain several valid lines. This case uses order and line together; another dataset needs its own confirmed record key.

Can I delete every flagged row?

No. The count includes both occurrences in each pair. Some are conflicts requiring a source check, and some repetitions may be valid. Review the source and document the decision first.

What if two codes look identical but do not match?

Check spaces, data types and identifier rules while preserving the original. Microsoft notes that TRIM alone does not remove every space character, including nonbreaking spaces. Avoid indiscriminate normalisation.

In this article

01A repeated order number can be a valid line02Choose the key before counting matches03Nine rows, four different situations04Give AI a review task with a traceable result05Choose an operation that preserves what you need06Reconcile the review before changing the source07Recent Excel developments still need your judgement

Your map of AI tools

Microsoft Copilot↗

Independent editorial content. RavenPrompts is not affiliated with or endorsed by these tools.

RavenPrompts

You know which AI to choose. Now put it to work.

Find a RavenPrompts prompt you can adapt to your task and take to your AI tool.

Browse the prompt library ↗View pricing

Related reading

Comparison

Compare two options with AI: spot what is missing before you choose

A tidy table is not enough to make a decision. Learn to compare two desks against your space and budget limits, without mistaking missing information for an advantage.

Read article ↗
Article

From a voice note to clear writing with AI: start by talking

When you know what you want to say but struggle to write it, try saying it aloud first. Learn to check the dictated text and turn it into a clear message without adding promises or losing details.

Read article ↗
Article

An AI training diary: organise your notes without making up data

Turn your post-workout notes into a record you can check easily. A simple example of keeping repetitions and minutes separate, leaving missing details blank and reviewing the AI’s answer.

Read article ↗
Article

How to check a date with AI before you share it

A screenshot says the workshop has moved. Before forwarding it, find the original notice and check which year it refers to. AI can help you work out where to look.

Read article ↗
Article

Plan a games afternoon with AI in a small flat

Five people, a 90 cm table and an hour to play. See how to ask an AI chat for a plan that fits your space and adjust it if the group changes.

Read article ↗
Article

How to fix an AI reply without starting over

When an AI reply goes off track, point out what to keep, which detail to fix and what the next draft should look like. Here's how that works with a book swap poster.

Read article ↗
Article

How to reschedule an appointment with AI and confirm the new time

A worked example for offering two real slots, checking the time zone and keeping one calendar invitation once the change is agreed.

Read article ↗
Article

Your CV with AI: clearer, without inventing a different life

A useful CV explains what you can do. AI can help organise and review it; the experience, dates and decisions remain yours.

Read article ↗
Article

Before you paste text into AI, check which details it needs

Use a print quote to separate job requirements from private details, ask for a draft with placeholders and check it before you send it.

Read article ↗
Article

Tracking unpaid invoices with AI and a balance you can check

Work through partial payments, a duplicate record and an unidentified receipt while keeping the supporting documents in view.

Read article ↗
Article

Your first AI chat, from rough notes to an email you can check

Start with a real task in ChatGPT. Draft a request for a quote, spot an invented commitment and fix it with a follow-up.

Read article ↗
Article

An AI study plan that fits your actual week

A science example for balancing explanation, practice and review, narrowing the goal when time is short and checking what you have learnt against your course material.

Read article ↗
Article

What is an AI prompt? Simple examples to get started

A prompt is what you ask an AI to do. That part is simple. The interesting bit is explaining what you need without writing a novel or learning a new vocabulary.

Read article ↗
Article

A thousand photos, no folders: getting organised with help from AI

Choose a way to find your memories, try it on ten photos and keep control of what you share. You can ask for ideas without handing over your gallery.

Read article ↗
Article

Gemini and your documents: check before sharing

A packing list for a day out is a simple way to understand the update: the content comes first, then the file.

Read article ↗
Article

The meeting is over. What happens to all those notes?

Granola, Fathom and Otter are changing what we expect from a summary: finding agreements, following a project’s thread and bringing a little more context to the next conversation.

Read article ↗
Article

With so many AI tools, choosing one has become a task of its own

Finding a source, understanding notes, imagining an image and giving a text a voice are different jobs. Looking beyond the familiar names helps explain where each tool might fit.

Read article ↗
Comparison

ChatGPT, Claude and Gemini: three ways to develop an idea

The choice changes when a conversation becomes a document, an ongoing project or a task spread across emails and files.

Read article ↗
Questions and answers

When an idea becomes an image, a voice or a song

AI brings the first version of an idea closer, opening new questions about its meaning, its origins and where it will end up.

Read article ↗
RavenPrompts

Prompts to create, learn and work better with artificial intelligence.

Find your next great prompt.

Follow us

  • Instagram
  • X
  • YouTube

Explore

  • Home
  • Prompt library
  • Blog
  • Free prompts
  • New prompts
  • Suggest a prompt
  • All categories
  • My favorites

Categories

  • Marketing
  • Social media
  • Productivity
  • Programming
  • Education
  • Photography and image
  • Travel

RavenPrompts PRO

  • Plans and pricing
  • My account and purchases

Help and contact

  • Frequently asked questions
  • Contact us
  • Purchase terms
  • Legal notice
  • Privacy
  • Cookie policy

© 2026 RavenPrompts. All rights reserved.

Made to take your ideas further.

Your cookies, your choice

We use necessary cookies for access, security and remembering preferences. You can allow speed measurement and visitor analytics separately. We do not use advertising cookies or cross-site tracking. You can also allow Metricool visitor measurement separately. Cookie policy

Cookie settings

Each optional measurement requires your permission. Earlier choices do not authorise the new visitor analytics. You can withdraw either permission at any time. Metricool is off by default. Your earlier Vercel permissions do not enable it.

They allow sign-in, protect forms and save your choices. Your language is remembered when you select it.

Always active

With your permission, Vercel Speed Insights measures loading, responsiveness and visual stability on some public pages. It does not collect prompts you enter or form contents.

Optional · off by default

With your permission, standard Vercel Web Analytics counts visits to public pages and provides aggregate device, browser and country statistics. We do not send searches, forms or prompts.

Optional · off by default

With your permission, Metricool counts visits to public routes and estimates the country from the IP address. We send no searches, forms, prompts or referring pages.

Optional · off by default

We do not use cookies or pixels to personalise ads or track your activity across sites.

Not currently used

We remember this choice for one year in this browser. You can change it at any time using the cookie button or the footer.

Cookie policy
Home
Prompt library
Blog
Free prompts
New prompts
Suggest a prompt
All categories
My favorites
Marketing
Social media
Productivity
Programming
Education
Photography and image
Travel
Plans and pricing
My account and purchases
Frequently asked questions
Contact us
Purchase terms