The fear around AI
When AI really started exploding into mainstream business conversation, there was a huge amount of fear around it, especially within marketing.
People were convinced it was coming for jobs. There was a genuine belief that marketing agencies would disappear, internal marketing teams would shrink and copywriters, designers, developers and strategists would all become irrelevant overnight.
And honestly, I understand why people thought that.
When you first see what some of these tools can do, it’s hard not to have that reaction. Even now, the speed at which AI is evolving is genuinely impressive. The new Claude design and coding capabilities are class. The ability to prototype software concepts, generate code, build drag-and-drop websites, automate workflows and speed up production processes is incredible when used properly.
There’s no question that AI has already changed marketing and business operations significantly, and it will continue to do so.
Personally though, I think a lot of the early conversation around AI was too extreme in both directions. Some people acted like it would destroy entire industries overnight, while others buried their heads in the sand and pretended it was a fad that would disappear in six months.
The reality probably sits somewhere in the middle.
Industries always evolve
AI is a major shift, and businesses that embrace it properly will almost certainly gain an advantage. Equally, businesses that ignore it completely will probably get left behind. History tells us that happens every single time industries evolve.
- CRM & Sales: Spreadsheets and disconnected systems were replaced by platforms like HubSpot and Salesforce centralising sales and marketing.
- Marketing & Advertising: Traditional media buying was disrupted by measurable digital advertising through Google, LinkedIn and Meta.
- Ecommerce: Retailers relying on physical footfall lost ground to ecommerce brands embracing Google Shopping, paid social and conversion-led online journeys.
Then there’s search, which is probably the biggest unavoidable shift happening right now.
AI is already changing how people search online. Google’s AI Overviews are reducing clicks for many searches, while people are increasingly using ChatGPT, Gemini and Claude to research suppliers, products and services directly.
Search behaviour is becoming far more conversational, and visibility is no longer just about ranking number one on Google. Businesses genuinely need to care about this because traditional SEO alone is no longer enough.
Our approach to AI
At Catalyst, we made a conscious decision fairly early on that we were going to embrace it rather than fight it. Not because we think AI replaces expertise, but because we think it massively improves efficiency when used properly.
That last part is important though: when used properly.
One thing we’ve learned very quickly is that large language models are only as good as the information you give them.
If you ask AI to “write a blog about SEO”, you’ll usually get something generic. Technically fine and structurally decent, but ultimately, it will look and sound very similar to the thousands of other blogs being generated on exactly the same subject.
That creates a huge problem.
Everyone starts sounding the same. Opinions disappear, differentiation disappears and the genuinely unique parts of businesses slowly get watered down. Search results become flooded with repetitive content and ironically, as more AI-generated content gets published, genuinely useful human insight becomes even more valuable.
That’s why I don’t believe AI replaces good marketers. I think it changes what good marketing looks like.
Where AI works best
For us, AI works best as an enhancement tool rather than a replacement for thinking. We use it heavily for research, structuring ideas, summarising information, improving workflows, prototyping and speeding up production tasks, all of which creates huge efficiencies internally.
But the human layer – opinions, commercial understanding, experience and context still matters massively.
A business owner talking honestly about what they’ve seen in their industry over 20 years is still infinitely more valuable than generic AI commentary. A strategist understanding buyer psychology or a developer knowing how systems should actually function commercially still matters.
The best results usually come when AI supports expertise rather than tries to replace it entirely.
AI is like HubSpot
A good comparison is actually HubSpot.
HubSpot is an incredibly powerful platform, but simply buying HubSpot doesn’t suddenly make a business commercially efficient. You still need good data, clear processes, strong workflows, internal alignment, reporting structures and strategy. Otherwise, it just becomes an expensive system that people barely use properly.
AI is very similar.
The businesses getting the most value from AI are usually the ones with strong operational foundations already in place. They know what problems they’re trying to solve, where inefficiencies exist and how their customers, systems and internal processes actually work.
How we’re using AI internally
That’s why we’re now using AI internally to help improve how the agency operates itself.
We’ve spent years working in agencies, so we know exactly where common frustrations exist, whether that’s delays, repetitive admin, workflow bottlenecks, reporting inefficiencies, briefing gaps or project management issues.
AI is helping us build tools and processes internally that improve those areas. Not because we want to remove people from the process, but because we want our team spending more time on high-value thinking rather than repetitive operational tasks.
Ultimately, that’s better for clients as well. Less time wasted on admin and manual processes means more time spent on strategy, creativity, problem solving and commercial thinking, the areas that genuinely deliver value rather than simply increasing time spent.
The risks businesses shouldn’t ignore
That said, there are still very real dangers around AI that businesses shouldn’t ignore.
AI still gets things wrong, and sometimes it gets things confidently wrong! It can produce inaccurate information, fake statistics and misleading recommendations surprisingly easily if outputs aren’t reviewed properly.
There are also obvious risks around:
- Generic brand positioning
- Loss of originality
- Weak thought leadership
- Legal and copyright concerns
- Over-automation
- Poor quality customer interactions
In some businesses, teams are already relying on AI outputs without properly understanding the subject matter underneath them, which is a dangerous place to be.
In summary
Personally, I don’t think AI kills marketing agencies. I think it changes what clients value from them.
Anyone can now generate content, ideas and basic assets. But understanding markets, positioning, buyer behaviour, commercial strategy and how to actually drive growth still requires experience.
If anything, AI probably increases the importance of strategic thinking rather than reducing it.
And honestly, I think that’s a positive thing.
If you want an outside perspective on where it could realistically help your business, feel free to contact us.
