As I slurped on my English Breakfast Tea over the weekend, I read some job posts that made me seriously laugh out loud. It appears that agencies are now hiring the next new wave of specialised skillsets.

Yes, this time it’s the new role of Prompt Engineers. But please give me a break! Our industry is already full of us shouting from the rooftops about how we are transforming through tech. I think when clients see this there will be a huge rolling of the eyes, they have enough on their plate with simply trying to get contact reports done in time. The last thing they need is another useless resource added to the rate card.

A Morning Brew and a Rant Brewed Stronger

One challenge with AI is figuring out how to ask for what you want in a way that gets what you want. For example, suppose you want an image of five footballers in a 4X4 in the middle of the desert debating VAR. What’s the best way to construct your prompt? Are you looking for a realistic or artistic feel? Any specific teams? What style of footballers, past or present? While complex nuances can be added to a prompt, which are most important for the brief? How do you structure the prompt to maximise the match between the output and your goals?

We’d all be happy to have a prompt engineer at the ready to accelerate our success with generative AI, but isn’t it lazy that we can’t even be bothered to embrace the tools and use them ourselves in our existing roles? And let’s not forget the bureaucratic existence of IT, who says we can’t even have these tools in place because ‘it’s not global policy and signed off yet.’ There are practical realities that will dampen the ability to turn that enthusiasm into a viable long-term role.

A major constraint on the viability of prompt engineering as a career is cost/benefit analysis. The loopholes I had to jump through just to get a freelancer with finance were enough to send me crazy. No prompt engineers existed even a year ago, and very few, if any, companies have one in place today. Each prompt engineer will be a net new budget expense that will have to be justified based on the value they can provide. I can see the FD gleefully rubbing his hands in anticipation of that debate.

The Rise of Yet Another “Role”

The marketing industry has a long history of inventing fancy job titles that sound impressive but often describe tasks that existing roles already cover. Remember when “Social Media Ninjas” and “Growth Hackers” were all the rage? How about “Customer Journey Architects” or “Brand Evangelists”? Each wave brought a flurry of business cards with impressive titles, but rarely represented genuinely new skill sets.

Prompt engineers are just the latest iteration of this phenomenon. At its core, the role involves understanding how to communicate effectively with AI tools – something that should be part of every modern marketer’s toolkit rather than a specialised position. It’s like creating a “Google Search Specialist” role in 2005 instead of just teaching everyone how to use search engines effectively.

What’s particularly frustrating is how these titles create artificial barriers. By labelling something as requiring specialised expertise, we discourage regular team members from developing these skills themselves. Instead of democratising technology, we create priesthoods around it – exactly the opposite of what we should be doing with accessible tools like generative AI.

The tech industry has perfected the art of repackaging existing skills with shiny new labels. This isn’t just annoying – it’s actively harmful to organisations trying to build genuine capabilities. When we create specialised roles around emerging technologies, we often end up with isolated pockets of knowledge rather than organisation-wide competency.

There’s also an undeniable performative aspect to these hiring trends. Companies want to be seen as innovative, and what better way than hiring for a cutting-edge role that signals “we’re on top of the latest tech”? It’s innovation theater rather than actual innovation – focusing on appearances rather than outcomes.

I’ve sat in too many meetings where executives proudly announce they’ve hired a specialist for some emerging technology, only to discover months later that this person sits in isolation, unable to drive meaningful change across the organisation. Real innovation happens when technologies are democratised and integrated into everyone’s workflow, not when they’re siloed into specialised roles.

Most of us today are not equipped to create prompts for generative AI applications right now. We are just getting by, but back in the day, we never needed “search engineers” either. We are getting by, and it isn’t worth the extra cost to have someone help us. Ask any CEO before his next quarterly finance meeting whether this is on his recruitment list. I don’t think so…

Historically, desk support roles are hard to staff. The people who know their stuff and are expert enough to answer a broad range of user questions will often want to apply their skills in a way outside of answering questions at a help desk. How many people will find it exciting to sit day after day and wait for the next order from a user to create a prompt?

The Economics Don’t Add Up

Let’s talk about the financial reality of prompt engineering as a career path. In today’s business environment, every new role needs to demonstrate clear ROI. When I look at prompt engineering through this lens, the numbers simply don’t add up.

First, there’s the salary consideration. Companies are advertising prompt engineering positions with surprisingly high salaries – often comparable to experienced developers or strategic marketing expertise. For a role that essentially involves crafting effective queries, this represents a significant investment.

Then there’s the question of output value. What’s the measurable impact of having a dedicated prompt engineer versus training existing team members in prompt crafting? In most cases, the incremental improvement in AI outputs doesn’t justify the cost of a specialised role. A marketing manager who becomes 80% as effective at prompt writing as a dedicated engineer, but also brings their domain expertise to the table, will likely deliver more overall value.

The economics become even more questionable when you consider the rapid pace of AI advancement. Today’s carefully crafted prompts may be unnecessary in tomorrow’s more intuitive systems. We’re investing in specialised skills that could be obsolete within a year or two as AI interfaces become more sophisticated and user-friendly.

The Learning Curve Isn’t That Steep

Another reason prompt engineering doesn’t make sense as a dedicated role is that the learning curve simply isn’t steep enough to justify specialisation. I’ve watched people with no technical background become proficient at prompt writing within weeks, not years.

The fundamental skills involved – clear communication, understanding context, iterative refinement – are things most professionals already possess. What’s needed is application of these skills to a new domain, not years of specialised training.

Compare this to genuinely specialised roles like machine learning engineers or data scientists, where years of education and experience are necessary to develop the required expertise. Prompt engineering just doesn’t have the same depth of specialised knowledge to justify dedicated positions.

This isn’t to say there’s no skill involved – there absolutely is. But it’s a skill that should be distributed across teams rather than concentrated in a single role. It’s more akin to learning PowerPoint or Excel than learning to code.

The Future Is Integration, Not Specialisation

The most successful organisations won’t be those with dedicated prompt engineers, but those that effectively integrate AI capabilities across their entire workforce. This means training everyone to be competent with these tools rather than creating bottlenecks through specialisation.

Think about how we approach other technologies. We don’t have “Excel Engineers” or “PowerPoint Specialists” – we expect a baseline competency across the organisation, with perhaps a few power users who can help with complex cases. The same model makes far more sense for generative AI.

What’s more valuable: one prompt engineer who crafts perfect prompts, or fifty team members who each craft good prompts that incorporate their specific domain expertise? The distributed model wins every time, both in terms of scalability and in terms of integrating relevant subject matter knowledge.

Prompt Engineering is going nowhere as a career. Most people won’t need much support as they gain more experience and expertise. It will be hard for a prompt engineer to add enough incremental prompt efficiency and quality to warrant the costs, and those who are best at it will often prefer another job. Perhaps there will be a few prompt engineers in place at some of the largest organisations that make significant use of generative AI.

However, the numbers will be limited, and I wouldn’t recommend betting your career on the field. I simply don’t see prompt engineers becoming a thing, so agencies please stop using it as propaganda to pretend you are forward-thinking.

The Democratisation of AI Skills

What we should be focusing on instead is the democratisation of AI skills across organisations. Rather than creating artificial specialisations, we should be working to ensure everyone has access to and competency with these powerful tools.

This means developing training programs that help all employees understand the basics of working with generative AI. It means creating shared resources and best practices that anyone can access. And it means celebrating when these tools become so intuitive that they require less specialised knowledge, not more.

The organisations that will thrive in the AI era won’t be those with the most prompt engineers – they’ll be those that most effectively integrate AI capabilities into everyone’s workflow. They’ll be the ones where marketing managers, content creators, designers, and strategists all have the skills to leverage these tools effectively within their domains.

A Better Path Forward

Instead of creating dedicated prompt engineering roles, here’s what organisations should be doing:

1. Develop basic AI literacy training for all relevant staff
2. Create centers of excellence that share best practices rather than doing the work
3. Identify and support power users within each department who can help their colleagues
4. Focus on integrating AI tools into existing workflows rather than creating new silos
5. Measure success by how widely AI tools are adopted, not by how specialised your team is

This approach delivers more value, scales better, and creates more resilient organisations than the specialised prompt engineer model. It also prepares companies for the inevitable evolution of these tools toward more intuitive interfaces that require less specialised knowledge.

Conclusion

The prompt engineer role is a solution in search of a problem – a specialised position that addresses a temporary friction point in technology adoption rather than a sustainable career path. As AI interfaces improve and general AI literacy increases, the need for dedicated prompt crafters will diminish, not grow.

Organisations would be far better served by focusing on building broad AI competency across their teams rather than siloing these skills in specialized roles. The future belongs to companies where everyone can effectively leverage AI tools, not just a select few.

So before you post that prompt engineer job listing or update your LinkedIn profile with the title, consider whether you’re solving a real problem or just participating in the latest round of tech industry title inflation. The answer might save you from investing in a role that could be obsolete before you’ve even finished the hiring process.

Right, I’m off for a fresh cup of tea!!!