Artificial intelligence has moved beyond the buzzword stage and is now driving real efficiency and growth for small and medium-sized businesses (SMBs). Still, many owners think of AI only in terms of tools like ChatGPT or worry that anything more advanced must come with a sky-high price tag. To cut through the misconceptions, get into the practical dollars and cents, and understand how SMBs need to prepare organizational data and content to benefit from AI, I sat down with my former colleague, Naveen Aggarwal from Explained Consulting, who now helps SMBs design and implement AI solutions built for their needs. We asked Naveen the following questions:

1. What’s the #1 misconception that small and medium-sized businesses have about AI?
2. Can you provide examples of solutions?
3. I know that AI can seem daunting, both in terms of effort and cost. We often hear in the news about companies spending millions of dollars. How can SMBs compete with that?
4. Why should businesses invest even that much?
5. For a lot of businesses, it’s hard to know where to start. What are you seeing as concrete business use cases that can make an impact quickly?
6. How is AI changing the way companies interact with their documents and information?
7. Beyond calls and documents, how does AI fit into workflows and approval processes?
8. What are the critical drivers for being successful with AI, and what are the biggest challenges to getting started?
9. We hear about AI “hallucinations” where systems make things up. How should SMBs think about that risk?
10. Can you provide a concrete example of how AI can be used to improve a manual task or workflow that most SMBs face? 
11. What are some of the interesting projects you are currently working on? 

Q&A with Naveen Aggarwal

Q: What’s the #1 misconception that small and medium-sized businesses have about AI?

The #1 misconception heard from small and medium-sized business owners about AI is that it’s mostly about tools like ChatGPT, or that basic tasks like email drafting are the full extent of what AI can do for their business. Many folks think that if they’re using ChatGPT, they’re doing AI. The reality is that’s just the tip of the iceberg—and it leaves a huge amount of value on the table.

AI today can help small businesses automate and improve processes throughout the company, not just with personal productivity tasks.

Q: Can you provide examples of solutions?

For example, in the front office, AI can respond to customers’ product or billing questions, qualify leads, and score prospects, saving staff time and increasing the likelihood that every customer and lead receives the attention it deserves.

On the back office side, AI can automatically tag invoices as they arrive, extract all relevant data, and route them into the correct workflows, thereby eliminating manual work and reducing errors. It’s about making the business more efficient and cost-effective, both for customer-facing and internal processes.

Q: I know that AI can seem daunting, both in terms of effort and cost. We often hear in the news about companies spending millions of dollars. How can SMBs compete with that?

You’re absolutely right to ask that question. I hear it all the time. When people see headlines about companies spending millions or even billions on AI, it’s natural to think, “How can we possibly compete with that?” But here’s the thing: we’re not competing with that, and we don’t need to. What you’re reading about are the tech giants, such as Google, Microsoft, and OpenAI, spending massive amounts to build foundational models from scratch. For SMBs, we don’t need to build those foundational models. We are using them at a fraction of the cost.

The reality is that implementing AI for most small and medium businesses costs a fraction of what people assume. We’re talking about $10,000 to $50,000 for meaningful implementations.

Q: Why should businesses invest even that much? 

Most SMB AI implementations deliver a 5 to 7-fold ROI within the first year, primarily through cost savings and increased efficiency. In fact, a majority of SMBs that have implemented AI into their normal business processes are looking to expand their AI investments. According to research by Salesforce, 75% of SMBs are already experimenting with AI, and 91% are reporting revenue increases. That to me is the proof point that AI is working and paying off for these organizations.

Remember, AI is a foundational technology. It is changing the way we do business. Eventually, you will be using it in your business. The question is not “if” but “when”. I liken AI now to what websites and e-commerce were in 1999. There was a time when companies questioned why they would ever need a website. Eventually, they either had to play catch-up, at a heavy cost, or they are no longer with us.  Companies that start their AI evolution now will be best positioned to adapt and grow as AI redefines how every industry operates.

Q: For a lot of businesses, it’s hard to know where to start. What are you seeing as concrete business use cases that can make an impact quickly?

First is what I call “never miss a call” AI. Small businesses lose customers every day because they can’t answer the phone during busy periods or after hours. AI voice assistants can handle initial calls, qualify leads, book appointments, and even answer basic questions by pulling information from your inventory or FAQ database in real-time. And because they connect directly into your CRM, every lead and interaction is logged automatically; no sticky notes or missed follow-ups. This isn’t the old school, “dial one for …” systems. These are natural language voice assistants that understand your business and your customers.

And it’s not just phone calls; AI chatbots on your website work 24/7 to capture leads, answer common questions, and route serious inquiries to the right person. 

Q: How is AI changing the way companies interact with their documents and information?

AI-powered searching is a prime example. Instead of employees digging through folders, emails, and systems to find information, they can simply ask questions in plain English. “What did we charge Johnson Industries last quarter?” or “Show me all pending invoices from automotive clients.” It’s like having a super-powered search box that understands context and speaks your language. Imagine having Google, but for your documents. It goes beyond keyword search. It understands the context.  I will go a step further. Imagine having a conversation with your product documentation. “I am looking for a SKU that is XYZ size? What is the cost?”

AI is also being used to create content. For example, sales intelligence agents can automatically create customer dossiers before meetings or calls. The AI agent compiles purchase history, past communications, preferences, and relevant notes to provide your sales team with a comprehensive picture in seconds, rather than hours of manual research.

Lastly, AI document processing is a significant development. AI can automatically read invoices, contracts, and forms, extract all the key data, and route them to the right workflows. Instead of someone manually entering vendor information from dozens of invoices each week, the AI handles it in seconds with higher accuracy.

Q: Beyond calls and documents, how does AI fit into workflows and approval processes?

AI shines when it’s embedded in your workflows, rather than being used as a standalone tool. Many SMBs waste hours chasing signatures, approvals, or status updates. AI can automate those flows end-to-end.

Here are some simple examples that save time and manpower by using automation and AI decision-making :

• Approvals: AI agents can verify if a purchase order matches budget rules, auto-approve it, or escalate it to a manager if something appears to be incorrect.
• Compliance checks: AI can validate documents against predefined rules (dates, signatures, customer IDs) and instantly flag any exceptions.
• Notifications: Instead of staff emailing back and forth, AI pushes status updates automatically to the right person in Slack, Teams, or email.

Humans still make the final calls on exceptions or unusual cases, but the AI handles 80–90% of the repetitive steps. That’s process engineering with automation—streamlining the flow so approvals and tasks move faster, with fewer errors.

Q: What are the critical drivers for being successful with AI, and what are the biggest challenges to getting started?

The critical driver for AI success is simple: your data quality. AI is only as good as the information it can access, so the principle of “garbage in, garbage out” is absolutely applicable. You need your business information organized and accessible to AI, not scattered across different systems. The beauty is that this scales with your business size. A landscaping company might start with a single Google Doc containing rates, hours, and FAQs for their AI call center. Larger organizations need more sophisticated data organization with proper segmentation for different regions or departments. 

This is where good document management makes everything easier. Instead of hunting through emails, file cabinets, and various software systems, centralized document management provides AI with a single location to find accurate, up-to-date information. Whether it’s policies, pricing sheets, or customer data, organized document storage becomes the foundation that makes AI implementations both faster and more reliable. Importantly, implementing document management doesn’t have to be expensive; it just needs to be in a location that is secure and accessible to AI technologies and agents. Solutions like the Veladocs repository, which leverages AWS S3 and Azure Blob Storage, ensure that content and data are securely accessed through cloud-native APIs.

The biggest challenge isn’t technical complexity. It’s just getting started with organizing your data. Businesses think they need perfect systems from day one, but you can begin with your most critical information and build from there. Even complex implementations with multiple data sources don’t require hundreds of thousands of dollars.

Q: We hear about AI “hallucinations” where systems make things up. How should SMBs think about that risk?

Hallucinations are real, and they’re the biggest operational risk in using AI. Large Language Models don’t know your business; they just predict plausible text sequences. Without context, they can sound confident but give wrong answers.

For SMBs, the solution is simple but powerful: ground AI in your actual documents and data. That means:

• Centralize your content: FAQs, price sheets, policies, service lists.
• Use metadata: Tag documents with type, date, customer, product information, etc. so AI knows what to pull.
• Apply Retrieval-Augmented Generation (RAG): This sounds more complex than it actually is. This approach involves sending related documents or information to the AI engine along with the question, allowing the answer to be generated from content specific to your business.
• Context prompting: Frame the model’s role clearly, like “Answer only from these docs. If you don’t know, say you don’t know.”

SMBs already know their business. The AI just needs structured access to that knowledge. When you combine your expertise with AI’s speed, you reduce hallucinations and get answers that are accurate, consistent, and trustworthy.

Q: Can you provide a concrete example of how AI can be used to improve a manual task or workflow that most SMBs face? 

After experimenting with Veladocs, I see a world where SMBs stop chasing email chains and missing paperwork altogether. Documents live in one secure place, AI handles the routing, and managers make approvals in the chat tools they already use. The back-and-forth disappears, and the whole process is faster, cleaner, and more reliable.

For example, manually managing Accounts Receivable is clunky and error-prone. An invoice is generated and manually emailed to the vendor. If the invoice is not paid on time, the vendor shows up on a report to the AR manager, who, in turn, assigns follow-up tasks to an AR employee. Emails go back and forth between multiple people on both sides before payment is received. Tracking email responses, follow-up calls, and vendor notes between different people is time-consuming, and missed communication is common. 

With AI and Veladocs, the flow is different. Invoices are automatically pulled in with Veladocs.sync and emailed to the designated contacts. Monthly aging invoice reminders are also automatically sent to customers. AI could further automate processes by implementing smart follow-up scheduling, which involves automating or assigning email, text, or call reminders based on customer engagement patterns. AI can automatically detect at-risk accounts and escalate follow-up tasks based on the account size. Self-service customer portals and chatbots can also be used to enhance the customer experience and expedite payments. 

With the right tools and platforms, this doesn’t have to be out of reach. The tools exist today, and it really is about taking the initiative to think differently about your processes and workflows.

Q: What are some of the interesting projects you are currently working on? 

Currently, I’m working on several exciting projects involving Voice AI for call centers, CRM automation, and rapid prototyping for a handful of clients. I also share updates and insights on LinkedIn. If you’re curious about how AI can work in your business, a short conversation usually surfaces two or three places where we can drive value quickly.

Laying the Foundation for AI Success

As Naveen points out, AI for SMBs is no longer a question of if, but when. The companies that start experimenting today are the ones best positioned to thrive tomorrow, much like businesses that embraced websites and e-commerce early in the late 1990s. The costs are more accessible than ever, the ROI has been proven, and the competitive advantages are undeniable.

Still, success depends on one essential step: preparing your information and content. Without organized, reliable information, even the most advanced AI tools can’t deliver their full value. That’s why content management platforms like Veladocs are so important. These platforms provide AI with a strong foundation: structured, secure, and ready to drive insights, enabling SMBs to reap the benefits of automation, efficiency, and smarter decision-making.

For SMBs, the opportunity is here. With the right guidance from experts and the right tools to manage their information, businesses can start small, scale quickly, and unlock the real power of AI. Contact us to get started and make sure your data and documents are ready for the power of AI.