AI's Impact on SMBs: Revenue Gains, Cloud Growth, and New Tools from Zoom, MailChimp, Microsoft

[00:00:02] It's Monday, June 17th, 2025, and I'm Dave Solvig for Things to Know Today. SMBs show strong interest in AI but need guidance over hype. Cloud spending surges as AI reshapes infrastructure priorities. An Oxford study warns of real-world AI failure risks. And new AI tools from Zoom, MailChimp, and Microsoft demand sharper oversight. This is the business of tech.

[00:00:29] Small and medium-sized businesses are increasingly recognizing the need to adopt artificial intelligence to remain competitive, with a recent survey by Vita revealing that 97% of small businesses using voice AI agents reported revenue increases. Despite these positive outcomes, many SMBs still view AI as unnecessary or overly complicated, with 63% of organizations expressing concerns about lacking the right foundation to support AI initiatives.

[00:00:59] Barriers to adoption include fears about the complexity and cost of AI, as well as concerns that customers prefer human interaction. Salesforce research indicates 91% of SMBs that have adopted any form of AI have reported revenue gains, suggesting a significant opportunity for those willing to embrace the technology.

[00:01:19] This is echoed in data from a recent survey by Reimagine Mainstreet, in partnership with PayPal, which reveals that a growing majority of small businesses view artificial intelligence as essential for maintaining competitiveness. The survey, which included nearly 1,000 small businesses with annual revenues between $25,000 and $5 million, found that 82% believe AI is crucial for competing in today's market.

[00:01:44] Currently, 25% of respondents have integrated AI into their operations, while over half are exploring its implementation. Among those using AI, marketing and customer engagement were identified as key areas for impact, with 77% highlighting its significance. A recent report highlights a significant shortage of artificial intelligence talent in the IT sector, with over half of IT leaders indicating their companies faced an undersupply of AI skills.

[00:02:12] This figure has risen from just 28% in the previous report, published in 2023, marking AI know-how as a top-scarce technology skill. The demand for AI expertise has surged, with 9 in 10 respondents reporting that their organizations are piloting or investing in AI initiatives, up from 59% two years ago. However, despite this investment, more than two-thirds of leaders have not yet seen a measurable return on investment.

[00:02:41] According to Bev White, CEO of Nash Squared, the lack of established practices in AI implementation has contributed to the skills gap, emphasizing the need for a combination of formal training, reskilling, and on-the-job knowledge sharing to address the issue. Why do we care? Well, SMBs are AI-curious but risk-averse. This is a moment for IT service providers to become the guides, not the pushers.

[00:03:06] The real opportunity lies not in pitching AI as magic, but in helping clients use it meaningfully and measurably, especially where it reinforces, not replaces, human value. This is especially relevant in a context where AI talent is in critically short supply. Even large organizations struggle to convert pilots into measurable returns. Most SMBs won't be hiring in-house AI engineers.

[00:03:31] They'll be looking to trusted partners, likely their MSP or IT consultant, to bridge the gap. That creates a ripe opportunity for ITSPs to position themselves not as sellers of AI tools, but as AI translators, trainers, and integration partners. Providers who develop practical AI onboarding services, provide training to non-technical users, and help SMBs structure data and workflows to enable intelligent automation will differentiate themselves.

[00:04:00] Those who merely resell AI tools without context risk becoming part of the problem. The next evolution of managed services isn't AI powered, it's AI enabled, with the provider as the differentiator. This episode is supported by FlexPoint. Managing cash flow can be tough for managed service providers. FlexPoint's working capital solution is designed to bridge the gap between invoicing and payment, giving providers access to funds when they need them.

[00:04:30] With quick approvals and flexible terms, it helps cover expenses, invested growth, maintain financial stability. Keep your operations running smoothly with FlexPoint. Visit getflexpoint.com slash MSP dash radio to learn more. Select, heard it on the business of tech or MSP radio, you'll get 10% off.

[00:04:54] Global cloud spending has surged to $90.9 billion in the first quarter of 2025, reflecting a 21% increase year on year, according to research from Canalys. This growth is largely attributed to the widespread adoption of artificial intelligence, as enterprises increasingly focus on optimizing their cloud infrastructure to support AI applications.

[00:05:16] Leading cloud providers including Amazon Web Services, Microsoft Azure and Google Cloud collectively hold 65% of the market share, with Microsoft and Google both reporting growth rates exceeding 30%. However, Amazon's growth has slowed to 17%, down from 19% in the prior quarter, partly due to supply side constraints affecting its ability to meet rising demand for AI infrastructure.

[00:05:40] Canalys highlights that while AWS's artificial intelligence businesses experience rapid growth, it remains in the early stages of development, prompting the company to adopt a price cutting strategy to enhance competitiveness. A recent survey conducted by Illuminous reveals a growing reliance on cloud partners among IT decision makers, with over half of respondents indicating dependence on professional services for cloud-related needs.

[00:06:05] Notably, 28% reported being heavily reliant on these services for all aspects of cloud adoption, migration and operation. The survey highlights key priorities in the cloud landscape, with 45% of participants focusing on cost management and 44% emphasizing security and disaster recovery. Additionally, 82% of organizations are utilizing both private and public cloud environments, with security being a major factor for opting for private clouds.

[00:06:33] The findings underscore the increasing complexity of cloud computing and the demand for external expertise in managing the platforms. Why do we care? Cloud isn't the destination anymore, it's the foundation for AI-driven business models. And that foundation is growing more complex, hybrid and performance sensitive.

[00:06:52] IT service providers have a narrow but powerful window to stake their claim as AI infrastructure enablers, designing, optimizing and securing cloud architectures purpose-built for modern workloads. The winners won't be just the ones who resell cloud. It will be the partners who simplify hybrid complexity, tame costs and allowed cloud usage with business strategy, particularly around AI adoption.

[00:07:16] If you're not building service lines around multi-cloud management, AI resource orchestration or cloud security compliance, you're leaving money and relevance on the table. Cloud strategy is no longer a cost of conversation. It's now an AI performance and security imperative.

[00:08:03] The study involved just about 1,300 participants who interacted with LLMs under controlled conditions. Yet they performed worse than a control group that diagnosed themselves without assistance. This raises concerns about the effectiveness of LLMs in real-world medical settings and suggests that traditional benchmarks used to evaluate the technologies may not reflect their actual performance when interacting with users. Why do we care? Why do we care?

[00:08:31] Well, the Oxford study is a flashing warning light for real-world deployment of those large language models, especially in mission-critical industries like healthcare, but with broader implications for IT service providers enabling AI adoption in any vertical. The core takeaway is damage. LLMs perform well in benchmark tests, but fail when humans actually use them. That disconnect matters deeply in sectors where the model is not the product. The human plus AI system is.

[00:09:00] If users misinterpret, misuse, or over-trust AI, the system fails, regardless of model accuracy in lab conditions. Gary Marcus, a professor emeritus at New York University, noted that the Apple paper we talked about last week has resonated widely, attracting over 150,000 readers to his related commentary. Performance metrics there revealed that in tasks requiring multi-term reasoning, the success rate plummets to just 35%. Accuracy isn't utility.

[00:09:29] This study reinforces that human outcomes, not model performance, must be the primary measure of AI success. For providers, this means shifting the focus from selling model access, or AI features, to designing complete human-aware AI systems that deliver verified value in practice. The emerging differentiation is not who can offer AI tools. It's about who can deploy them responsibly, effectively, and with measurable impact on user performance.

[00:09:58] Providers who treat LLMs as just another app risk missing the deeper challenge and opportunity of orchestrating AI as part of a system that includes people, processes, and oversight. Zoom Communications has launched the Zoom Virtual Agent 2.0, the next-generation self-service virtual agent powered by agentic artificial intelligence.

[00:10:21] This new version aims to enhance customer support by allowing the virtual agent to autonomously complete complex tasks such as processing returns and booking appointments without human intervention. According to a Gartner report, 85% of customer service leaders plan to explore or pilot customer-facing conversational generative AI this year, highlighting a significant trend towards intelligent automation in customer service.

[00:10:45] Zoom Virtual Agent is designed to meet the increasing demand for fast and accurate self-service solutions across multiple channels. MailChimp has launched a comprehensive suite of new tools aimed at assisting small and mid-sized businesses and leveraging customer data to enhance growth and marketing effectiveness.

[00:11:02] At the FWD London event, the company revealed that 71% of small businesses surveyed expressed concerns over acquiring new customers in the current economic climate, highlighting the need for solutions that facilitate customer engagement and marketing automation. The new features include expanded integrations with major platforms such as Meta, TikTok, Google, Snapchat, and LinkedIn to streamline lead generation.

[00:11:25] Additionally, MailChimp has introduced over 100 customizable pop-up templates tailored to various industries, as well as Metrics Visualizer that provides more than 40 performance metrics for email and SMS campaigns. The updates are part of a broader strategy by Intuit, MailChimp's parent company, to create a unified business management ecosystem that integrates seamlessly with their accounting software, QuickBooks.

[00:11:49] And Microsoft has officially launched Copilot Vision for Windows, which is a feature that allows the AI assistant to see and interact with users' screens in real time. The tool enables users to share their applications and receive guidance, such as coaching on software features or analyzing content. Copilot Vision is now available for free to all Windows users in the US, marking a significant shift from previous requirements for a Copilot Pro subscription.

[00:12:17] The feature operates within the Copilot app, activated through a simple icon, and is designed as an opt-in experience requiring users to share their applications specifically. Microsoft has indicated that this tool is part of its ongoing experimental initiatives within Copilot Labs, further enhancing its AI capabilities. Why do we care? Or, agentic AI, embedded AI, and integrated SMB stacks are arguably all here, and they're impacting customer expectations.

[00:12:47] For providers, the market does not sell the AI tool and instead own the AI experience. There are key questions to address, such as how Zoom's fully autonomous claims may trigger real-world reliability issues. If a virtual agent misprocesses a return or books an error, who's liable? This is the moment to position around AI enablement and governance, not just tool implementation. The platforms are building fast, but the guide is still missing. If that's your role.

[00:13:19] This is the moment to go. The guide is still missing. This is the moment to position around AI, and there's something to do in the world. The guide is soon to go. We'll see you in the next two minutes. The guide is coming. The guide is to bring up on AI and the next two minutes.

[00:13:47] Huntress's 24x7 Global Sock shuts them down before they're even on anyone else's radar. But they do more than just chase alerts. They lead the charge in industry research and knowledge, bringing expert protection and peace of mind. That's why users on G2 rate their EDR number one for growing businesses. To see how their expert threat hunting team gets the job done, visit Huntress.com slash MSB Radio. Thanks for listening.

[00:14:16] Today is National Fudge Day. It's also Take Your Cat to Work Day. Happy birthday to my amazing wife, Sharon. You may not know this, but she's the producer on the live shows, our field events, much of our intro and outro work. Happy birthday. Join me for a webinar sponsored by Nerdio, Modern Endpoint Management with Intune. What works and what doesn't. Visit bit.ly slash Nerdio webinar or link the show notes. The Business of Tech is written and produced by me.

[00:14:45] Dave Sobel under ethics guidelines posted at businessof.tech. If you've enjoyed the show, make sure you've subscribed or followed on your favorite platform. It's free and helps directly. Give us a review, too. If you want to support the show, visit patreon.com slash MSB Radio and you'll get access to content early. Or buy our Why Do We Care merch at businessof.tech. Have a question you want answered?

[00:15:13] We take listener questions, send them in, ideally as a voice memo or video to question at MSP Radio.com. I answer listener questions live on our Wednesday live show on YouTube and LinkedIn. If you've got a comment or a thought on a story, put it in the comments if you're on YouTube or reach out on LinkedIn if you're listening to the podcast. And if you want to advertise on the show, visit MSP Radio dot com slash engage.

[00:15:38] Once again, thanks for listening and I will talk to you again on our next episode. Part of the MSP Radio Network.