AI in the MSP Space
Part One: Helping Clients
A lot of MSPs have AI on the brain – and how could they not? The AI revolution is for real. And MSPs are uniquely positioned to ride the AI wave and to help their clients do the same.
But figuring out your MSP’s path forward is tricky. Things are unfolding in real time, and they’re unfolding at breakneck speed! If you’re looking for some guidance – some clarity amidst the chaos – you’ve come to the right place.
In this two-part blog (stay tuned for part two), we’re breaking down the state of AI in the MSP space. In this first installment, we’re looking at how MSPs can help clients harness AI effectively and responsibly. Part two will deal with the implementation of AI within your MSP business.
What to Expect (and Not to Expect) from These Articles
If you’re looking for a step-by step, foolproof AI implementation plan, sorry. We’re not withholding such a thing; it just doesn’t exist!
If you’ve been talking to other MSPs or perusing online forums, you’ll have noticed that MSPs are approaching AI very differently – and with very different attitudes. On one end of the spectrum, you have the gung-ho early adopters, and on the other end, you’ve got the skeptics and naysayers.
Most MSPs are somewhere in the middle, trying to make sense of things and chart a course forward. So if you’re excited about AI, but not sure what steps you should be taking, you’re not alone!
And while we can’t give your MSP all the answers in regard to AI adoption and implementation, we can point you in some promising directions, clarify the current lay of the land, and help you start formulating a strategy for your MSP business.
What Makes AI Different – and Right Up MSPs’ Alley
To begin, let’s talk about what makes AI special. It’ll be illuminating to compare it to a recent technological advancement: cloud computing. There are interesting similarities between cloud computing and AI, but we’re going to focus on a key difference:
The benefits and power of AI are obvious, self-evident even. The benefits and cloud computing are not.
As someone who works in IT, you know all too well that moving data to the cloud is, all things being equal, a good idea. You understand how cloud computing offers businesses a more scalable, agile, and cost-effective data storage solution. However, getting your clients to understand this and embrace the cloud is a whole different animal (which is why we have a blog post devoted to the topic).
With AI, the challenge is almost the inverse: clients are going to be eager to adopt AI, since the power of AI is so undeniable, so obvious. A lot of us had a similar experience using ChatGPT for the first time, OpenAI’s groundbreaking Large Language Model (LLM): This thing is incredible! The idea that businesses are going to embrace AI much more quickly and easily than cloud computing isn’t just speculation either:
ChatGPT was the first tech application to acquire 100 million distinct users in just two months. And get this: in a recent survey of 1,000 small business owners, an overwhelming 91% reported increased success following the implementation of AI.
Individuals and organizations will use tools that make their lives easier. Your job, then, as an MSP, isn’t going to be convincing clients to use AI, but rather, helping them do so responsibly, effectively, and efficiently. This brings us to…
MSPs as AI Guides
As an MSP you don’t just ‘fix your clients’ computers.’ You act as a trusted advisor, delivering strategic advice to help clients align their technology investments with their business goals.
If there were ever a perfect opportunity to prove your worth as an advisor, it’s in guiding clients through the adoption of AI technologies. In particular, MSPs should prepare to offer guidance in the following areas:
LLM ‘Tuning’ and Customization
LLMs are incredibly powerful and useful tools, but they’re not one-size-fits-all. The true power of an LLM is unleashed through what is known as “fine-tuning” or simply “tuning”: refining a pre-trained LLM’s capabilities and improving its overall performance by training it on smaller, specialized datasets (i.e., data specific to a client business).
Forward-thinking MSPs are going to master the art of tuning, so that they can equip their clients with LLMs that produce accurate, tone-appropriate content at the click of a few buttons.
But do organizations really need help with this?
Big time! A recent survey from Predibase polled executives, data scientists, developers, and other key decision makers and found that only 11% had successfully fine-tuned an LLM and been satisfied with the results. A whopping 78% had not, with 44% citing “complexity” or a “lack of knowledge” as the reason why. Did someone say, “Golden opportunity for MSPs”?
Data Security
AI introduces a host of new vulnerabilities and data privacy concerns. For instance, if a client is using an LLM, they need to make sure their data is properly organized and tagged.
This is where your MSP can come to the rescue – in helping your clients prepare their data for AI intervention. This means, among other things, structuring their data such that an LLM, or some other AI system, won’t reveal sensitive information to the wrong parties.
And you’d better believe clients are looking for help with data security in AI implementation. The aforementioned Predibase survey found that 33% of respondents listed “giving up access to proprietary data” as a top challenge preventing them from using LLMs in production. This was the most commonly listed top challenge, with second place going to “customization and fine-tuning” at 30%.
So if your MSP can help clients leverage and customize LLMs and other AI tools without compromising data security, it will help you stand out and your clients gain a competitive edge. Win-win!
Microsoft Copilot
Last November, Microsoft made its own AI-powered chatbot available to enterprise customers. Copilot is designed to help businesses enhance productivity and streamline workflows within the Microsoft 365 ecosystem.
MSPs stand to generate some serious revenue by guiding clients through the implementation and management of Copilot. Getting the most out of Microsoft’s new AI tool takes real work: creating a robust data strategy at the outset, ensuring security through proper access controls, and implementing safeguards for long-term maintenance.
But it’s the sort of work for which MSPs are perfectly suited. Learn more about Copilot and the unique opportunity it presents MSPs. Also, check out some of the cool things Zapier is doing with AI Copilot integrations.
The Human Element
AI adoption is a tricky business. There’s no way around that. The technology is new, powerful, and bursting with untapped potential. And because no two businesses share the exact same needs in regard to AI adoption, how well your MSP fulfills its role as trusted advisor will depend largely on one thing:
How well you know your clients’ businesses – how they operate, what they value, and what they’re trying to achieve.
The full force of AI as a business tool stems from the technology’s high level of adaptability and customizability. The only way to tailor an AI strategy to fit your clients’ specific needs is to understand those needs in the first place. Translation: talk to your clients! This isn’t something AI can do for you – not yet at least. It’s a human thing, and it’s crucial to your success as a trusted source of AI wisdom.
Conquer AI through Collaboration
The AI industry is exploding in size and complexity, and MSPs trying to figure things out for themselves have an uphill battle. But imagine being able to tap into the expertise and experiences of hundreds of MSPs, a whole community of companies who share knowledge and tips as freely as old friends trading tales around a campfire…
That’s exactly what it’s like to be a member of The 20 MSP Group. We’re stronger together, and we win together. To learn more about the powerful benefits we bring our MSP members, schedule a call today.
Thanks for reading, and stay tuned for part two, where we’ll look at how your MSP can leverage AI to improve operations and elevate service.