Hi Reader,
I was talking to one of my bootcamp participants last week, and they asked me a great question: "Do you use AI tools in your content strategy work, and are they increasing how much you make?"
I do, and they are, so I explained a bit about how I use them, where they save me time and effort, and how that increases my profit margin.
Then they asked a follow-up question: "But couldn't you just charge less because it takes you less time? Then you could do more strategies and make up the difference."
Um, no. For so many reasons. My clients don't pay me for my time, I don't bill hourly, I don't want to do more work, and I want to increase my profit margin, not reduce it.
Imagine you regularly do a project that used to take you ten hours. You charged $100 an hour, so the client paid $1,000. Now you have an AI tool that helps you complete part of the research and work more efficiently. You still have to do the hard stuff, but the entire project takes six hours instead of ten.
If you still charge by the hour, your invoice is now $600. The client received the same completed project and saved $400.
You paid for the tool, learned how to use it, figured out where it could and couldn't be trusted, and became more productive. Your reward for all of that was earning less.
Of course, you also got four hours back.
You could use those hours for another paying project, which would allow you to earn more overall. But that assumes you have another project ready to fill the space.
If you don't, you haven't increased your income. You've completed the project faster and been paid less for it, and I'm not sure why any of us would want that.
When people talk about AI increasing earning potential, they don't always go into how it can do that and what you have to change to make it happen. Increased productivity and increased income are not the same thing. AI creates the possibility of earning more, but your business still needs a way to capture the value created when you use it.
It's not a black-and-white situation
There's solid evidence out there that AI can help people complete certain tasks faster and, in some cases, improve the quality of what they produce.
In a field experiment involving 758 Boston Consulting Group consultants, people using GPT-4 completed tasks more than 25 percent faster, finished more tasks, and received substantially higher quality ratings when the work fell within the tool's capabilities.
The researchers also described what they called a "jagged technological frontier," because AI performed impressively on some tasks and poorly on others. The struggle is real. The advantage in this study came from understanding where the tool helped and where human expertise was a necessity.
That's all well and good, but completing work faster is a productivity result, not an income result.
According to Upwork's marketplace data, freelancers performing AI-related work earned 34 percent more per hour than those whose work did not incorporate AI. But the advantage was not spread evenly across every kind of AI work.
Contracts for generative AI and creative production work increased 90 percent while earnings per contract fell 13 percent. More clients were contracting for the work if it involved AI, but the work itself was becoming less valuable per engagement. Meanwhile, more complex professional services that combined AI with domain expertise grew in both volume and earnings.
That only shows that the market is not placing the same value on every kind of AI-assisted work.
When AI makes a deliverable easier for more, or less qualified, people to produce, the supply of that work grows faster than its value. Clients have more choices, turnaround windows act like Shrinky Dinks in an oven, and there's consistent pressure to lower prices to compete with a supposed "market price." (Just say NO!)
When AI is being directed by someone with meaningful experience, judgment, and accountability, it supports more complex work. In that situation, the value isn't the tool itself. It's the human expertise and judgment that make the tool useful.
We also have evidence of what happens on the more vulnerable side of the market.
Researchers studying freelance writers and designers after the release of ChatGPT and popular image-generation tools found short-term declines in both assignments and earnings.
Writing-related freelancers experienced a 2 percent decline in monthly jobs and a 5.2 percent decline in monthly earnings. Image-related freelancers experienced larger drops.
AI is helping some of us do higher-value work and earn more. It is also increasing competition, lowering the value of some execution work, and encouraging clients to expect more speed for the same money.
In my book, that's a bad direction for the market to take.
Instead of asking if we should be using AI, maybe we need to ask what kind of work we can provide, what clients believe they are buying, and who benefits financially when the work takes less time.
But does it make your business more profitable?
It's not a good idea to assume AI is making your work more efficient and profitable. You need to prove it by looking at what's happening holistically in your business, instead of only focusing on if AI makes one task faster. Saying AI makes your business better without proving it turns the whole thing into a straw man fallacy.
Let's look at facts instead of fallacies.
AI can absolutely improve profitability when it reduces the total cost of delivering work without reducing the fee you receive. If you charge a fixed project fee and complete the work in less time, your effective hourly earnings increase.
If AI reduces unpaid administrative work, research time, or other behind-the-scenes tasks, it can improve the economics of an engagement even when the client-facing work remains largely the same. This is my favorite way to use it.
It can also create additional capacity. You might use the time you save to take on another project, develop a new offer, pursue better clients, or work fewer hours without reducing your income.
All of those can be legitimate gains, although they do not all have the same effect on your bank account.
Remember though that time that could theoretically be used for client work is earning potential. It only becomes additional revenue when you use it for something that generates revenue.
AI can also make the work less profitable.
That happens most obviously when you bill by the hour and increased efficiency results in a smaller invoice.
But it can happen with other pricing models too. A fixed-fee project becomes less profitable if faster production leads to more versions, more revisions, or more deliverables. A retainer, especially one based on a set number of hours, becomes less profitable when every hour you save means more work is added within the same amount of time.
There are less obvious costs as well. AI tools cost money. Learning to use them takes time. Preparing source material, protecting confidential information, checking facts, correcting weak output, and making sure the final work is actually good can consume some or all of the time you appeared to save.
You can also end up producing more simply because you can.
If AI helps you complete twice as many deliverables but the client is paying the same fee, your output has increased without a corresponding increase in revenue.
That might be worthwhile if the extra work improves the result or strengthens the relationship, but it should be a deliberate choice rather than the cost of using a faster tool.
Find out what's happening in your business
1. Choose some of your recent projects or one recurring service you provide regularly. For each one, write down:
- The fee you received
- The total time the work required
- The cost of any tools, contractors, or other direct expenses
- The time spent preparing information for AI
- The time spent reviewing, verifying, correcting, and refining its output
- Any additional work or revisions that were added along the way
- What happened to the time you saved
Be honest about the total time spent on the project. Include meetings, email, research, preparation, revisions, quality control, file management, and the final handoff.
If you count only the time spent on the task AI performed faster, you get a distorted picture of what the project required.
2. Then calculate your net effective hourly earnings: fee received minus direct project expenses, divided by the total number of hours worked.
You do not have to bill the client by the hour to use an internal hourly calculation. It's simply a way to assess the project-level economics and find out whether the service is financially working for you.
3. Compare that number with similar work you completed before introducing AI, or with the minimum effective hourly rate you need the service to produce. Then ask:
- Which parts of the process take less time?
- Which still require the same amount of judgment, communication, verification, or responsibility?
- Has the fee changed? Up or down?
- Have the client's expectations changed? For better or worse?
- What happens to the time you save?
- Who is receiving most of the financial benefit from your increased efficiency?
If your effective hourly earnings have increased, AI is probably improving the economics of the project.
If they have fallen, find out why.
Did the invoice amount go down? Did the scope expand? Are you spending more time fact-checking the tool than expected? Did you lower the price because you assumed the work would be easier? Are you counting unused availability as though it were additional income?
You may also discover that your effective hourly earnings have stayed roughly the same while your working hours have gone down.
That may be a success, depending on your business goals. Profitability is important, but it is not the only useful measure. Earning the same amount with less stress, fewer late nights, or a better work/life balance can make your business more sustainable even if your revenue doesn't go up.
| How is AI affecting the profitability of your client work? |
|
|
|
|
|
Make sure you keep some of the gain
Once you know where the money and time are going, you can decide what needs to change.
Set an internal minimum. Decide what a project or service needs to bring in after expenses and total working time. If the effective rate repeatedly falls below that number, you need to set a higher rate, adjust the scope, or fix your process. You might also want to take a serious look at if you want to continue offering that service. Breaking even or losing money are not good for a business.
Protect fixed-fee scopes. Define the deliverable, number of revisions, expected timeline, and what happens when the client requests additional work.
Keep retainers from blowing up. Define the responsibility or capacity the client is purchasing. As with fixed-fee projects, scope is important. Any meaningful expansion of the work should be quoted and invoiced separately.
Revisit hourly rates and minimum fees. Hourly billing may make the most sense when the work is unpredictable or the client controls the volume. However, if AI has permanently reduced the time required for repeatable work, your existing rate might need to go up to stay profitable. You might also consider setting a minimum project fee, minimum billing block, or fixed fee for repeatable work.
Separate faster production from rush delivery. AI helping you work more efficiently does not mean every client automatically receives an fast turnaround. If a client needs work sooner than your normal turnaround time, treat that as rush work and price it accordingly.
Don't forget the hidden costs. Tool subscriptions, training, verification, privacy precautions, and quality control should all be part of your profitability assessment. They also need to be factored into your rates.
Decide what the saved time is for. If you want the additional capacity to increase your income, you need a plan for using it. If you want it to shorten your workweek, don't allowing it to fill with more client work. Use it wisely.
And if a particular AI-assisted process repeatedly creates more revisions, risk, or client management than it saves, stop using it.
Adopting a tool does not obligate you to use it everywhere.
Use it intentionally
You need to know whether your business gets to keep the value that speed and efficiency create.
I am not in any way saying AI is a magic wand that brings in more money. Every dollar still requires your hard work and input.
"Earning potential" is too often treated as though the only worthwhile use of increased efficiency is producing more. More projects, in less time. I don't agree. Mostly because more work doesn't always mean more profit. It just means more work. That's hustle culture, and it's harmful for a lot of reasons.
If the benefit you want from using AI is to protect your income while reducing your workload, create more room for strategic thinking, or to stop spending evenings on administrative work, it might be working for you.
If, however, the client receives a faster turnaround, more deliverables, and a smaller invoice while you receive additional pressure, AI is not improving your business. Something in the offer, pricing, scope, or process has to change.
Whether it makes your business more profitable, more sustainable, or simply more demanding is a business decision you need to make for yourself. And that means you need to measure if it's actually producing in the "more" you want.
Cheers,
Treasa
