I wanted to ask for a raise.
But I did not want to walk into the conversation with nothing more than:
“I have been working hard, and I think I deserve more.”
Working hard matters.
But it does not automatically create a convincing business case.
So I decided to use AI to help me prepare.
I gave it information about the role, responsibilities, recent achievements, market salary range, and the additional work being handled.
Then I asked it to build my negotiation strategy.
The first answer was terrible.
It was generic, overly polished, and full of empty statements such as:
“I consistently exceed expectations and bring significant value to the organization.”
That sounded confident.
But it proved nothing.
So I changed the way I used AI.
And that was when it became useful.
AI Could Not Build the Case Without Evidence
My first mistake was asking AI to create a raise request before giving it enough proof.
AI can improve your language.
It cannot invent a credible reason for your employer to pay you more.
A strong raise request needs evidence such as:
Revenue generated
Costs reduced
Time saved
Projects completed
Targets exceeded
Customers retained
Processes improved
Additional responsibilities
People trained
Problems solved
Instead of asking:
Help me ask for a raise.
I used a more detailed prompt:
Act as a salary-negotiation coach. Help me build a business case for a raise based only on the evidence I provide. Identify my strongest achievements, remove weak claims, and tell me what information is missing. Do not exaggerate or invent results.
I then added my responsibilities and achievements.
The quality of the response changed immediately.
AI Showed Me Where My Case Was Weak
One of the most useful things AI did was challenge my evidence.
I had written:
I took on more responsibility.
AI asked:
Which responsibilities?
When were they added?
Were they previously handled by someone more senior?
What outcome did you produce?
How did this work help the company?
Can the impact be measured?
That exposed a problem.
I knew I was doing more.
But I had not documented it clearly enough to prove it.
So I created a simple table:
Responsibility or result | Before | My contribution | Business impact |
|---|---|---|---|
Weekly reporting | Took four hours | Built a faster process | Reduced preparation to one hour |
Client onboarding | No standard process | Created a checklist | Reduced repeated questions |
Team support | Manager handled training | Trained three new employees | Improved onboarding consistency |
Campaign performance | Declining engagement | Tested new approach | Increased clicks by 22% |
The raise request became stronger because it was no longer based on effort alone.
It was based on impact.
The First Script Sounded Too Aggressive
AI initially gave me an opening similar to:
Based on my performance and expanded responsibilities, I believe my compensation should be increased to reflect my current value.
There was nothing technically wrong with it.
But it did not sound like a natural conversation.
It also moved to the request before establishing the evidence.
I asked AI to rewrite it using a calmer structure:
Contribution → Evidence → Request → Discussion
The revised version was much better:
Over the past year, my responsibilities have expanded beyond the original scope of my role. I have taken ownership of the weekly reporting process, trained three new team members, and introduced a new onboarding system that reduced repeated customer questions.
I would like to discuss adjusting my compensation to reflect the level of responsibility and results I am now delivering. Based on my research, I believe a salary in the range of [amount] to [amount] would be appropriate.
I would appreciate your perspective and would like to discuss what would be possible.
This sounded confident without becoming confrontational.
AI Helped Me Find the Right Salary Range
AI can help organise salary research, but it should not be treated as the final source of truth.
Salary ranges depend on:
Location
Industry
Company size
Experience
Job level
Technical requirements
Management responsibility
Bonuses and benefits
Current market conditions
I gathered salary information from several sources and then asked AI to help me compare it.
My prompt was:
Compare the salary ranges below for similar roles. Identify the most realistic range based on five years of experience, additional project responsibility, and strong performance. Explain the assumptions behind your recommendation and highlight any information that is missing.
The useful part was not the number AI suggested.
It was the reasoning.
It helped me separate three figures:
Target: The salary I would confidently request.
Expected outcome: A realistic figure the employer might approve.
Minimum acceptable outcome: The result below which I would need another benefit, commitment, or review date.
Keeping these figures separate prevented me from opening with the lowest amount I would accept.
I Used AI to Challenge My Own Argument
Most people use AI to agree with them.
That is not always helpful.
I asked it to act like a skeptical manager:
Review my raise request as a manager with a limited salary budget. Identify every reason you might reject or delay it. Challenge my evidence and ask the difficult follow-up questions I should prepare for.
It raised objections such as:
These achievements are already part of your job.
The company cannot increase salaries right now.
Your performance is strong, but not yet at the next level.
Other employees have taken on additional responsibilities too.
We need to wait until the annual review.
There is not enough budget.
We cannot approve that percentage increase.
This was more valuable than receiving another polished script.
It forced me to prepare for the real conversation.
How I Prepared for Common Objections
“There is no budget right now.”
I understand that budgets may already be allocated. Could we agree on the specific results I would need to achieve and schedule a compensation review for a defined date?
“These responsibilities are already part of your role.”
Some of the responsibilities overlap with my original position, but the scope has expanded. I am now also responsible for [specific examples], which were not included when my current salary was agreed.
“You are not ready for the next level.”
I appreciate the feedback. Could we define the specific skills, results, or responsibilities I would need to demonstrate to be considered ready?
“We can review it later.”
That makes sense. Could we agree on a date now and document the expectations that will be used during the review?
The goal was not to win an argument.
It was to avoid leaving the meeting with a vague promise.
AI Helped Me Rehearse the Conversation
Reading a script silently does not prepare you to say it confidently.
So I used AI for role-play.
I asked:
Act as my manager during a salary review. Ask one question at a time, challenge my answers, and do not approve the raise too easily. After each response, tell me where my answer is weak, vague, defensive, or unsupported.
The role-play revealed several habits:
I spoke too quickly.
I overexplained.
I softened the request too much.
I focused on effort instead of results.
I became defensive when challenged.
I did not always answer the exact question.
After several rounds, my responses became shorter and clearer.
Instead of saying:
I have been doing a lot more recently, and I feel that I have shown I am committed and willing to help wherever needed.
I could say:
My responsibilities expanded in three measurable areas: reporting, onboarding, and team training. The reporting change alone saves approximately three hours each week, while the new onboarding process has reduced repeated support requests.
One answer describes effort.
The other demonstrates value.
What AI Could Not Do
AI helped me prepare.
But it could not understand every part of the situation.
It did not know:
The company’s real financial position
Internal salary bands
Office politics
How my manager communicates
Whether a promotion was already being considered
How difficult replacing me would be
Whether the company had recently frozen salaries
What had been discussed privately by leadership
It could also generate language that sounded more confident than the evidence justified.
That is why every recommendation needed human judgement.
AI was useful as a coach, editor, and practice partner.
It was not the decision-maker.
So, Did AI Get Me the Raise?
AI did not get the raise.
It made the case harder to dismiss.
The biggest result was not a clever sentence.
It was that the preparation became more structured.
I entered the conversation knowing:
What I had achieved
Which results mattered most
What range I would request
How I would explain my value
Which objections might come up
What I would ask for if the answer was no
What follow-up commitment I needed
That changed the tone of the conversation.
Instead of asking the company to reward loyalty or effort, I was discussing responsibilities, impact, market value, and future contribution.
That is a much stronger position.
The Prompts I Would Use Again
Build the business case
Act as a salary-negotiation coach. Analyse the responsibilities and achievements below. Identify the strongest evidence supporting a raise, remove weak claims, and list any missing information I should collect.
Turn achievements into proof
Rewrite these achievements using clear business outcomes. Do not invent numbers. Where the impact is unclear, ask me a question that would help make it measurable.
Prepare the opening
Create a 60-second opening for a raise conversation using this structure: contribution, evidence, salary request, and invitation to discuss. Make it confident, professional, and natural.
Challenge the request
Act as a skeptical manager with a limited budget. Give me the five strongest reasons for rejecting or delaying this raise and help me prepare calm, evidence-based responses.
Practise the conversation
Role-play my salary-review meeting. Ask one question at a time. Challenge vague claims and give feedback on clarity, confidence, evidence, and tone.
A Better Way to Use AI for Salary Negotiation
Do not ask AI to create a magic script.
Use it to strengthen the entire process:
Document your achievements.
Identify measurable business impact.
Research your market value.
Build a clear request.
Challenge your own argument.
Prepare for objections.
Practice speaking out loud.
Decide what you will request if salary cannot move.
Document the next steps after the meeting.
The tool matters less than the evidence you give it.
Final Thought
AI will not make an employer value work you cannot prove.
But it can help you recognize where your argument is weak.
It can help you organize your evidence, improve your language, prepare for objections, and walk into the conversation with greater confidence.
The biggest lesson was simple:
Do not use AI to sound more impressive.
Use it to make your real value easier to understand.
Want to Build Career Proof People Can See?
Inside the Threads Income Lab, you will learn how to share your knowledge, build a relevant audience, and turn your experience into visible proof that can create career and income opportunities.
You do not need to become an influencer.
You need a clear message, useful content, and a repeatable system.
How to Ask for a Raise and Prove Your Value
Build a stronger salary case using measurable achievements, market research, and confident communication.
5 AI Skills That Will Get You Promoted in 2026
Develop practical AI skills that help you save time, improve results, and become more valuable at work.
How to Build a Career Development Plan
Create a clear roadmap to develop your skills, increase your income, and prepare for your next opportunity.
What Would You Ask AI Before a Salary Conversation?
Are you preparing to request a raise, negotiate a promotion, document your achievements, or respond to an employer who says there is no budget?
Send me the situation you are facing.
Your question may shape a future OwnerPath guide on salary growth, workplace negotiation, AI career tools, and proving your value at work.