Do You Still Need an SEO Agency Now AI Tools Exist?
AI can write your content. It can’t tell you whether that content gets you found, recommended and contacted. Here’s an honest comparison for directors and marketing teams, with the results to back it up.
By Digital Landscope · 8 minute read
The Short Answer
Yes, if you want AI search to bring in customers rather than just more content. AI tools have made it faster and cheaper to write pages, research topics and tidy up a website. What they haven’t done is tell anyone what actually gets a business found, recommended and contacted.
Below, we compare the two honestly, show where AI genuinely helps, and share the results our testing-led approach has delivered for clients.
The choice isn’t AI or people. It’s output or outcomes.
Outcomes in AI search come from testing, judgement and experience that no tool can copy.
AI Tools vs an Expert Agency at a Glance
The quick version for anyone weighing up their options.
| Question | DIY with AI tools | Expert search agency |
|---|---|---|
| What you get | Content, quickly | A strategy built around enquiries and sales |
| Knows whether it worked | No | Measured every month against named competitors |
| Unique to your business | Similar answers to your competitors | Built around your sector, buyers and goals |
| Keeps pace with AI search | Learns from past information | Tested continuously as platforms change |
| Builds authority beyond your site | No | Link building and brand visibility |
| Accountable for results | Nobody | A team reviewing performance with you |
| Best for | Speeding up groundwork | Making search a reliable source of leads |
What AI Tools Are Genuinely Good At
AI tools have their place, and many businesses use them, and so should you. Used well, it speeds up the groundwork.
None of that is the hard part. It’s the part that was already quick to do.
AI is a brilliant assistant. It just isn't a strategy.
- Drafting first versions of pages and posts
- Summarising what competitors are saying
- Suggesting questions your customers might ask
- Tidying up page structure and metadata
Where AI Falls Short
AI tools are built to produce output. They aren’t built to tell you whether it worked.
01
They can't see your results
The tool writing your content has no idea whether last month’s pages brought in a single enquiry.
They give everyone the same answer
Your competitors are asking the same tools the same questions and getting near-identical content back.
They describe search as it was
AI models learn from past information, while the way AI search recommends businesses changes month to month.
They can't explain their own choices
Ask ChatGPT why it recommends one supplier over another and you’ll get a plausible guess, not the real reason.
They can be confidently wrong
AI will invent facts, figures and features. Publish them unchecked and you damage your credibility with buyers and AI engines alike.
They don't know your business
Your margins, the services you want more of, the customers worth chasing and what your sales team hears on calls all shape a good strategy. None of it is given to you by AI tools.
Put simply, AI can write. It can't decide what's worth writing, check whether it worked, or change course when it didn't.
Why Testing Matters More Than Tools
AI search isn’t a fixed results page like traditional search. Intent matters more and certain parameters of a users prompt may move the goalposts for the business worth recommending. Even asking a different AI tool, and the answer can shift again.
So a single check tells you almost nothing. Seeing your name in one ChatGPT answer isn’t visibility, and missing from one isn’t failure. What matters is how often you appear, for which questions, in which tools, and whether that’s moving in the right direction.
The platforms also keep changing. What earned a recommendation six months ago may not today, and traditional Google rankings still drive a large share of enquiries alongside it all. A strategy that only chases one channel, or only checks in once, is guessing.
The only way to know is to test repeatedly, measure properly and change one thing at a time. Tools can collect data. They can’t decide what’s worth testing, spot what the results really mean, or connect them to the enquiries landing in your inbox.
- Competitor B
- Your business
- Competitor E
One check says you're doing brilliantly. Try asking again.
Illustration only, with a made-up example. Real AI answers shift in the same way, which is why one check proves nothing.
How Digital Landscope Approaches It
We started testing how businesses get recommended in ChatGPT and AI Overviews in early 2025. Our own website was the first test subject, and Digital Landscope began appearing in ChatGPT when people searched for answer engine optimisation agencies. Since then, every technique we use has been tested before it reaches a client.
Track.
The questions your buyers actually ask, across the AI tools they use
Measure.
Your share of voice against named competitors over time, not in one-off snapshots
Test.
Controlled changes, watching what shifts the recommendations
Combine.
AI search work with the SEO signals that still drive Google
Refine.
Every result fed back into the next month’s strategy
It’s the difference between publishing and hoping, and publishing because the evidence says it works. What moves the needle differs by sector, which is why every strategy we build is bespoke.
The Evidence: What Testing-Led Search Delivers
These results didn’t come from a tool. They came from deciding what to test, reading the results and acting on them.
more organic enquiries for a Glasgow shredding company
extra enquiries a month from ChatGPT for an Edinburgh caterer
monthly Google clicks for a kilt hire brand, up from 700
monthly visitors for a clinic that started with none
+50 bookings a month
Edinburgh Pizza and Catering Company
A three-restaurant independant Edinburgh pizza chain wanted more table bookings, delivery orders and event catering. We built a bespoke SEO and AEO strategy covering schema, FAQ content, Q&A-style blogs, location pages and website changes. After three months they had around 1000 extra monthly users and a 40% rise in event enquiries. By the end of the strategy, bookings were at 50+ a month.
Glasgow Shredding Company
This client was spending heavily on Google Ads. Our strategy combined competitor analysis, conversion rate optimisation, ranking reviews, phased keyword deployment and technical fixes. Organic email enquiries rose 446%, strong enough that they cancelled their Google Ads entirely. We’re now extending the work into AI search.
Renfrewshire Kilt Hire Brand
We’ve worked with this local kilt company since 2023, building rankings for competitive terms like “kilt hire Glasgow”. Their articles are built from the owners’ own expertise through a review process, and they’ve kept performing since AI Overviews arrived.
Lanarkshire Aesthetics Clinic
This clinic’s website was out of date and receiving no organic traffic at all. After a WordPress rebuild and a long-term SEO strategy, it now attracts more than 530 monthly visitors.
Why Expertise Can't Be Replicated by AI
Every AI tool learns from what has already been published. Expertise is everything that hasn’t.
Judgement built over years
Our consultants average more than 10 years in SEO. They've seen search change many times and know which new tactics are worth testing and which to ignore.
Patterns across many businesses
We run strategies across dozens of clients at once, from engineering firms to letting agents. What we learn testing in one sector sharpens the next, and no single tool sees that picture.
Your knowledge, not the internet's
The content that performs best is built from what you and your team know: your processes, your customers, your results. AI can't interview your experts.
Commercial focus
A ranking that brings no enquiries is worthless. We judge success by the leads and sales it drives, not by how much content gets published.
Accountability
When results stall, someone has to explain why and change course. A tool never will.
There's an irony here, too.
AI engines favour sources that add something new: first-hand experience, original insight, real results. Content that simply repeats what AI already knows gives those engines little reason to recommend you. The more businesses lean on AI-written content, the more genuine expertise stands out.
See Where You Stand in AI Search
Get a straight answer on how your business shows up on Google and in AI answers today, and where the growth is.
- Which buyer questions AI recommends you for today
- Which competitors are winning the answers you’re missing
- Where your website is losing enquiries it should be winning
- The first things we’d change, in plain English
Book a Free AEO Discovery
Free, no obligation and no pitch deck. Tell us a little about your business and we’ll be in touch.
AI, In-House or Agency: Which Is Right for You?
The honest answer depends on where your business is and what you need search to do.
DIY with AI tools can work if…
- You’re a small local business with few competitors online
- You have time to learn, test and check results yourself
- Search is a nice extra rather than a main source of leads
An in-house marketer using AI can work if…
- They have real search experience, not just content skills
- They have dedicated time to measure and test, not just publish
- They aren’t also covering social, events, email and everything else
An expert agency makes sense if…
- You need search to be a reliable, growing source of enquiries
- Competitors are already showing up in AI answers and you aren’t
- You’ve tried AI-written content and seen no change in leads
- Your marketing team is stretched and needs specialist support
A mix often works best: your team handles day-to-day content with AI, and we set the strategy, run the testing and measure what’s working.
So, Do You Still Need an Agency?
You don't need an agency to use AI. You need one to know whether it's working.
If your team is already producing content with AI but enquiries haven’t moved, the problem usually isn’t output. It’s that nobody is testing what actually gets you recommended, measuring it against your competitors, and acting on what the results show.
That’s the work we do every day, and it’s why our clients see results that a tool on its own can’t deliver.
Want to Know Where You Stand?
Book a free AEO discovery and we’ll show you which buyer questions AI recommends you for, and which ones your competitors are winning.