'The biggest risk of AI is over-reliance': How AI is changing the way web agencies deliver value

1. Pro
2. Website Hosting

'The biggest risk of AI is over-reliance': How AI is changing the way web agencies deliver value Features By Owain Williams published 9 June 2026

I caught up with Suhaib Zaheer, SVP, Managed Hosting at DigitalOcean & GM, Cloudways, to get his thoughts on how AI is transforming the web agency space

When you purchase through links on our site, we may earn an affiliate commission. Here’s how it works.

(Image credit: Future)

• Copy link
• Facebook
• X
• Whatsapp
• Reddit
• Pinterest
• Flipboard
• Threads
• Email

Share this article 0 Join the conversation Follow us Add us as a preferred source on Google Newsletter Subscribe to our newsletter

AI is automating the exact tasks that web agencies once monetized, like routine maintenance and web monitoring.

This shift may seem scary to some, but it is also unlocking new opportunities, creating operational efficiencies, and freeing up skilled website professionals to focus on high-value tasks.

I caught up with Suhaib Zaheer, SVP of Managed Hosting at DigitalOcean and GM of Cloudways, to discuss how AI is redefining how agencies deliver value to their clients. During our conversation, he highlights the opportunities and risks of AI adoption, as well as breaking down exactly how agencies can mitigate those risks.

Interview with:Interview with:Suhaib ZaheerSocial Links NavigationSVP, Managed Hosting at DigitalOceanInterview by:Interview by:Owain WilliamsSocial Links NavigationB2B Editor

Historically, agencies would bill clients for ‘maintenance and monitoring’. As AI automates a lot of these tasks, how are agencies' value propositions changing?

What used to take 30–40 minutes of investigation can now be understood in minutes.

For years, agencies have spent significant amounts of time on reactive operational work like investigating outages, reviewing logs, troubleshooting performance issues, and implementing fixes.

AI is changing that dynamic. What used to take 30–40 minutes of investigation can now be understood in minutes. AI can quickly surface root causes, identify patterns across applications and infrastructure, and in some cases even automate the next action.

You may like

• AI is making everyone web app builders - but leaving teams exposed
• Managed Intelligence Providers are the next phase in AI evolution: Here's what SMBs need to know
• AI Agents at your service

Leveraging AI allows agencies to move further up the value chain and focus more of their time on performance optimization, improving SEO visibility, conversion improvements, and helping clients achieve better business outcomes.

This shift matters because the landscape websites operate in is more demanding than it used to be. There’s more automated traffic, more variability, and more pressure on performance and visibility. It’s not enough for a site to just be online anymore. The agencies that will stand out are the ones that use AI to reduce their operational load and reinvest that time into higher-value work that clients are willing to pay for.

Are you a pro? Subscribe to our newsletter

Sign up to the TechRadar Pro newsletter to get all the top news, opinion, features and guidance your business needs to succeed!

Contact me with news and offers from other Future brandsReceive email from us on behalf of our trusted partners or sponsorsBy submitting your information you agree to the Terms & Conditions and Privacy Policy and are aged 16 or over.

Is AI helping smaller, creative-heavy agencies to become competitive with larger technical agencies by lowering the barrier of entry? How are these smaller agencies gaining a competitive edge?

AI is lowering the barrier, but it’s not a complete leveller.

It gives smaller teams access to capabilities that once required deeper technical expertise, like monitoring, diagnostics, and performance optimization. Most of the issues agencies deal with aren’t unique. They follow consistent patterns – inefficient queries, resource limits, or configuration issues that show up across different environments.

AI can identify those patterns quickly and point to the likely cause, which reduces the need for specialist input on every issue.

What to read next

• From experimentation to execution: where agentic AI is delivering real value
• Four key questions insurers must answer to embrace AI effectively
• How AI is reshaping compliance: Why governance still matters

That allows smaller agencies to stay lean while still operating at a much higher technical level than they could have a few years ago.

Agencies often inherit ‘messy’ legacy websites through the onboarding process. How is AI helping agencies audit and stabilise unoptimized portfolios at scale?

Most legacy websites look messy, but the underlying issues are usually not unique. They tend to follow similar patterns – excessive database usage, ineffective plugins, poor caching configurations, or other application-level bottlenecks.

AI helps agencies audit those portfolios by identifying issues quickly and consistently across multiple sites. Instead of manually reviewing each environment, it can surface the likely root cause in real time, which significantly reduces the effort involved in diagnosing problems.

That makes stabilisation much more scalable. Once those common issues are identified, they can be resolved faster and in a more consistent way across the portfolio, bringing sites into a stable state sooner.

For agencies managing large numbers of inherited sites, that’s the shift. Auditing and stabilising portfolios becomes far less manual and much easier to do at scale.

How is AI changing how agencies approach security? For example, can it help tell the difference between legitimate spikes in high-volume traffic and DDoS attacks?

It’s not a replacement for dedicated security tools.

Distinguishing between normal traffic and malicious activity is getting harder. There’s more automation, more bot traffic, and more variability in how sites behave.

AI has the potential to help by analysing patterns in real time and flagging behaviour that doesn’t look normal – whether that’s an unusual spike or activity that suggests something more aggressive. It gives agencies a clearer starting point when something goes wrong.

It’s not a replacement for dedicated security tools, and it shouldn’t be treated as one. Traditional security layers like WAFs, DDoS protection, rate limiting, malware scanning, and access controls remain foundational. Where AI is useful is in speeding up diagnosis and helping teams understand what they’re dealing with more quickly.

What risks could agencies face when relying too heavily on AI-automated hosting? How can they mitigate these risks?

The biggest risk is over-reliance.

AI is very effective on the problems it was designed for, but it doesn’t cover every scenario. More complex or highly customised issues are still much harder to diagnose consistently.

There’s a tendency to assume AI can handle the full workflow, when in reality it’s strongest as a diagnostic layer. It can surface the likely root cause quickly, but that still needs to be validated and acted on in context.

That’s why it needs to stay grounded in a human-in-the-loop model. In our case, Cloudways Copilot operates in a controlled environment, where it guides investigation but doesn’t have unrestricted control.

The balance is using AI to remove the repetitive part of troubleshooting, while keeping decision-making with the team. That’s how you get the benefit without introducing risk.

What emerging AI-powered hosting technologies do you think will prove to be the biggest opportunities for agencies in the coming years?

The opportunity is less time spent reacting to problems and more time spent improving how sites perform across the portfolio.

The biggest opportunity is the move from reactive troubleshooting to more proactive systems.

Today, most of the value comes from faster diagnosis – identifying issues quickly and reducing the time spent figuring out what’s gone wrong. The next step is shifting earlier in that process, where common infrastructure issues are identified and addressed before they impact performance.

In practice, that means systems that continuously analyse signals across servers and applications, and flag patterns that are likely to cause problems before they escalate.

It won’t remove the need for human input, especially in more complex or application-level scenarios. But if agencies can catch and resolve the most common issues earlier, it has a significant impact at scale.

The opportunity is less time spent reacting to problems and more time spent improving how sites perform across the portfolio.

TOPICS AI Owain WilliamsSocial Links NavigationB2B Editor, Website Builders & CRM

Owain has been building websites and online stores for his own and his client's businesses for over 8 years. Having taken on a role at TechRadar Pro in 2023, he now leads on all website builder and CRM content, spending his days researching, testing, and reviewing some of the best website building and CRM platforms on the market. He also has a passion for helping people get a great deal on website builders, delivering the best coupon and promo codes on the market. With an extensive background in business, Owain holds a BA(Hons) in Business and Marketing and has written for several leading publications including MarketingProfs, Website Builder Expert, Digital Doughnut, and NealSchaffer.com.

View More

You must confirm your public display name before commenting

Please logout and then login again, you will then be prompted to enter your display name.

Logout Read more Pro AI is making everyone web app builders - but leaving teams exposed    Pro Managed Intelligence Providers are the next phase in AI evolution: Here's what SMBs need to know    Pro AI Agents at your service    Pro How AI is reshaping compliance: Why governance still matters    Pro From experimentation to execution: where agentic AI is delivering real value    Pro Four key questions insurers must answer to embrace AI effectively    Latest in Website Hosting Website Hosting 'The biggest AI opportunity is not replacing people': How OpenClaw is empowering the next wave of successful SMBs    Pro ICANN opens applications for a rare chance to own a top-level domain    Website Hosting One hour with SiteGround    Website Hosting What’s next for domains in 2026: leveling the field for builders and small businesses    Pro How to turn Minecraft into Animal Crossing, Pokemon and more    Website Hosting Namecheap reveals the most popular TLDs of 2025 - and although an old favorite rules the roost, new AI challengers are coming    Latest in Features Amazon Prime Video Clarkson's Farm season 5 latest episodes are a stark lesson in toxic fan culture    Gaming I spent a month with the ModRetro M64 — and I think there’s no better way to play your classic Nintendo 64 games    AI Platforms & Assistants I tried Siri AI on the iPhone, Mac, and iPad — here's why I'm convinced Apple's long-overdue next-gen assistant will win you over    Televisions Why I wouldn't buy a super-cheap huge-screen TV for the World Cup    Paramount Plus Dutton Ranch fan theory thinks Yellowstone spinoff will have dangerous Landman crossover    AI Platforms & Assistants I tested NotebookLM's impressive new visual storytelling features    LATEST ARTICLES

1. 1'More than any other country in the world' - Ukraine wants to be the biggest drone and robot producer worldwide, outstripping China, Russia, and the US combined —Kyiv claims it could build more than 30 million units annually within years
2. 2NYT Connections hints and answers for Wednesday, June 10 (game #1095)
3. 3Quordle hints and answers for Wednesday, June 10 (game #1598)
4. 4NYT Strands hints and answers for Wednesday, June 10 (game #829)
5. 5'AI tools could lead to nothing less than the death of astrophysics': Researchers predict bleak future for thousands who study black holes, galaxies, and supernovae