Posts

Posted on

Shadow IT and Data Governance

Computers are tools that no business can operate without. From the simplest of organizations to the largest of corporate enterprises, computers are the tools that enable the work. Yet business IT often operates without enough thought or attention paid to data access and governance, where applications or services are installed or implemented by non-technical users that don’t always understand the implications of their actions. IT isn’t just about computers and servers, routers and switches. It’s about the user environment, workflow, data, security, applications, infrastructure and more. The resources which provide the foundation for whatever it is the business does – this is the area of IT. 

When businesses need to implement new applications, IT must install or secure or protect the solution. When a business needs to set up databases or analytics infrastructure, it is in the realm of IT to provide those resources. Very little happens in a business without the support of information technology. 

Too often, small and growing businesses minimize the importance and strategic value of closely managing their information technology when it comes to building longevity and reducing risk in the organization. IT departments and MSPs face a constant struggle to keep up with demand while fighting a battle against ungoverned expansion of applications and services in use.  

According to Wikipedia, “Shadow IT refers to information technology systems deployed by departments other than the central IT department, to bypass limitations and restrictions that have been imposed by central information systems. While it can promote innovation and productivity, shadow IT introduces security risks and compliance concerns, especially when such systems are not aligned with corporate governance.” 

When a business elects to implement a solution outside of the existing IT environment, or in the current environment but without consideration of implementation standards or resource availability, it reduces the time-to-benefit of the solution and new risks are introduced. Additional costs may also accompany this activity due to requirements to buy more infrastructure to adequately support the solution, or through closing gaps exposed in an improperly secured deployment. 

Noobeh helps businesses manage and protect their IT environment more efficiently, providing the change control and governance needed to turn IT into a strategic business advantage.

Mendelson Consulting’s cloud services team powered by Noobeh has the experience of helping businesses establish a solid foundation suitable to support business sustainability, improvement and growth. Leveraging the security, flexibility and massive scalability of the Microsoft Azure cloud platform and Microsoft 365 Fabric and framework, Noobeh helps businesses improve their infrastructure, reduce IT break/fix and administrative costs, and take advantage of the power and interoperability of the Microsoft technology stack. Noobeh helps businesses keep their systems managed, protected, and ready to handle whatever comes next.

Whether it is migrating applications from on-premise to cloud, providing remote support and management of computers and devices, licensing and administration of Microsoft 365 services, or setting up database, data warehouse or data lake infrastructure for analytics and AI, the Noobeh cloud team at Mendelson Consulting deploys and supports strong foundations for growing organizations. 

bunny feetMake Sense?

J

Posted on

Unlocking Insights in QuickBooks Enterprise Data

Businesses of all sizes are under pressure to turn their data into actionable intelligence.

As organizations adopt modern analytics platforms like Microsoft Power BI and Microsoft Fabric, the ability to unify, govern, and analyze data across systems is no longer optional—it’s foundational.

High-quality, connected data enables leaders to move beyond intuition and toward AI-assisted, insight-driven decision-making across the organization.

While enterprise companies have long relied on sophisticated ETL platforms, small and mid-sized businesses are often left behind. Many still depend on manual exports, spreadsheets, and point-to-point integrations that are brittle, time-consuming, and fundamentally incompatible with AI and advanced analytics. These approaches create data silos, limit scalability, and make it difficult to trust the results.

Mendelson Consulting and the Noobeh Cloud Services team help SMBs modernize their data foundations using Microsoft Fabric and Azure.

By deploying and supporting core business systems—such as QuickBooks Enterprise Desktop, Acctivate Inventory, Sage ERP, MISys Manufacturing, and others—within the Microsoft cloud ecosystem, we position application data for seamless ingestion into Fabric’s OneLake, enabling analytics, reporting, and AI workloads to work from a single, governed source of truth.

Modern data platforms like Fabric bring together data integration, engineering, warehousing, real-time analytics, and BI into a unified experience. This matters because growing businesses don’t just have more data; they have more types of data. Financial systems, inventory and manufacturing platforms, operational tools, and external data sources all need to be analyzed together to deliver meaningful insights and support AI models.

Even traditionally desktop-bound systems such as QuickBooks Enterprise can be extracted, structured, and integrated into a Fabric-backed data warehouse or lakehouse. Once centralized, this data can be enriched with operational and external data, exposed through Power BI, and used to power AI-driven insights, forecasting, and anomaly detection.

A successful analytics and AI strategy starts with the right data architecture.

Before businesses can leverage copilots, predictive models, or intelligent automation, they must first collect, organize, and govern their data at scale. Mendelson Consulting and Noobeh provide the expertise to build that foundation, helping businesses move from disconnected reporting to a future-ready, AI-enabled analytics platform.

bunny feetMake Sense?

J

Posted on

AI and Cybersecurity: Don’t Trust, Always Verify

Faster, cheaper and more scalable

The advancements in artificial intelligence are reshaping the landscape of cybersecurity, with AI now the single biggest force in the network. AI can discover vulnerabilities faster, it can execute highly scalable automated attacks, and it can help malware adapt and change to avoid detection or gain new capability. In cybersecurity defense, AI is used for real-time threat detection, and it facilitates automated response and triage capabilities, too. But part of the trouble comes from within, where companies are increasingly deploying AI tools without properly securing them, creating entirely new risks for businesses to consider.

The internet is run by machines

Human users versus hackers is no longer a model that applies when it comes to internet security. AI and bot traffic is growing far faster than human user traffic, and automation has given way to AI-driven fraud, account takeovers, credential stuffing and scraping and more. Large-scale attacks are far easier and cheaper to deploy, allowing a literal explosion of bots and automated traffic – machines running machines – across the internet.

You are the product

Free games aren’t really free. Even what seems to be a harmless activity can become a conduit of valuable data, conducting surveillance and recording information. Individual bits of data may not have great meaning, but in aggregate it might. The telemetry gained from devices and applications provides location information, networking and proximity data and more. The exchange of convenience or enjoyment for security and privacy is a well-known tradeoff that bad actors exploit continuously.

Identity is the new attack surface

It used to be that cybersecurity focused on the devices – the endpoints which represented the way into the network. Endpoint security is essential, yet it is the user identity which is the vulnerable element. It has been said that bad actors aren’t hacking systems any longer, they’re just logging in. This means that stolen credentials drive the majority of system breaches. Breaking into and highjacking active sessions, bypassing MFA challenges, and performing other identity-based attacks is now forcing a shift toward continuous authentication and a completely Zero Trust (never trust, always verify) security model.

Cybersecurity has never been easy, but it is harder than ever now that AI is involved. There’s a market out there for enabling the bad guys, like cybercrime as a business model. It’s organized and scalable and terrifying. More than ever before, cyber risk is tied directly to business risk, making security something far more than just IT.

Make Sense?

bunny feetJ

Posted on

Data Is the Real Competitive Advantage

AI is powerful, but it’s not magic. AI can’t compensate for fragmented systems, poor processes, or unreliable data. When a business succeeds with AI, it isn’t necessarily because they have the most advanced models. More often, AI success is found by those with the strongest data foundations.

High-Quality Data is the Foundation for Successful AI in Business

Artificial intelligence is rapidly becoming a competitive differentiator in business. From forecasting demand and optimizing pricing to automating customer service and detecting fraud, AI promises efficiency, insight, and scale. Yet many AI initiatives fail to deliver meaningful results—not because the algorithms are weak, but because the underlying data is.

In practice, AI is only as good as the data it learns from. Without high-quality, well-organized, and reliable data, even the most advanced AI tools will produce inaccurate insights, reinforce bad decisions, or fail entirely. For businesses looking to use AI responsibly and effectively, data quality is not optional – it is foundational.

Garbage In, Garbage Out: The Reality of AI

AI systems don’t “think” or “reason” in the human sense. They identify patterns based on historical data. If that data is incomplete, inconsistent, outdated, or biased, the AI will replicate and amplify those problems.

Sales forecasts built on inconsistent historical revenue data will be unreliable, customer churn models trained on incomplete customer records will miss key risk signals, and AI copilots trained on poorly documented internal processes will give incorrect guidance to employees.

In short, AI can’t fix broken data. It can only scale its flaws.

AI Strategy Should Follow Data Readiness

Many organizations pursue AI due to competitive pressure or fear of falling behind. This “AI FOMO” often leads to rushed implementations that skip essential groundwork.

A better approach is to ask these four simple questions:

  1. Do we trust our core business data today?
  2. Can we explain where key numbers come from?
  3. Are our processes documented and consistently followed?
  4. Do we have a single, reliable version of the truth?

If the answer to any of these questions is “no,” the priority should be improving data quality—not deploying more AI tools.

High-Quality Data Enables Trust and Adoption

AI systems only create value if people trust and use them. When employees see AI outputs that conflict with known realities or change unpredictably, confidence erodes quickly.

On the other hand, when AI is built on clean, well-governed data, the insights it provides align with business intuition and recommendations are explainable and defensible. This increases adoption of the tool across teams, and allows AI to become a decision-support tool rather than a black box.

Trust starts with data.

High-quality data is the real enabler of AI—turning automation into insight, predictions into action, and experimentation into sustainable advantage. For organizations serious about using AI in business, the path forward is clear: fix the data first. That’s where the competitive advantage will come from.

bunny feetMake Sense?

J

Posted on

Fix the Data, Then Let AI Scale It

For SMB’s Using Solutions like QuickBooks Online, Service Titan or Jobber, High-Quality Data Is Critical for AI

Many small and mid-sized businesses now run on a combination of operational and financial tools. A typical stack might be QuickBooks Online (QBO) for accounting plus Service Titan or Jobber for field operations. Noobeh helps these businesses centralize their data, making it available for analysis and AI.

What we increasingly find is that various AI vendors promise AI-powered forecasting, automation and insights, but AI does not create clarity on its own. When data across these systems is inconsistent or poorly structured, AI simply automates confusion. To get real value from AI, SMBs must first ensure their data is accurate, aligned, and trustworthy.

The Reality of Disconnected SMB Systems

For these small businesses, each system serves a different purpose. QuickBooks tracks financial transactions, revenue, and expenses, where Service Titan or Jobber manages the jobs, customers, technicians and billing. There may be problems lurking in these various systems, and it is often revealed when the data is centralized and made ready for reporting and AI-enabled analytics.

These problems arise when the same business concepts—customers, jobs, revenue, costs—are represented differently in each system. Common examples of this include jobs marked as complete in Service Titan or Jobber but not fully invoiced in QBO, or customers duplicated or named differently across platforms, or any situation where manual spreadsheet adjustments are needed to make the reports work.

Imagine training your AI on this data. It isn’t going to resolve the data issues or repair them, it will repeat them at scale.

A Practical AI-Ready Data Path for SMBs

Before deploying AI features across QBO, Service Titan or Jobber, our consulting teams help our clients focus on making sure the data is ready by cleaning and standardizing QBO financial data and ensuring jobs, customers, and invoices align across systems. Our cloud services team leverages Azure platform services to create automation and eliminate manual spreadsheets and workarounds. Then we centralize the data in Microsoft Fabric, creating a single source of truth allowing reports to be validated prior to laying AI on top. This approach turns AI from a grand experiment into a dependable business tool.

Trust Is the Real Measure of AI Success

AI only delivers value when business owners, finance teams, and operators trust the outputs. That trust comes from seeing numbers that reconcile, reports that make sense, and predictions that align with reality. When this alignment occurs through high-quality data, AI forecasts become credible and insights are explainable. Decision-making improves consistently.

Fix the Data, Then Let AI Scale It

AI can help SMBs compete with much larger organizations—but only when it’s built on a strong data foundation. QuickBooks Online, Service Titan, Jobber, and Microsoft Fabric form a powerful stack, but their value depends on data quality and alignment.

For SMBs, the winning strategy is clear: fix the data first, then let AI scale what’s already working.

jm bunny feetMake Sense?

J

Posted on

AI FOMO and Your Business

“AI FOMO” (Fear of Missing Out) has become a major force behind business adoption of artificial intelligence.

Rather than pursuing AI with a clear strategy, too many organizations are investing because of competitive pressure, media buzz, and fear of falling behind. This reactive approach often leads to rushed, expensive, and poorly executed initiatives that fail to create real value—and can even spark internal friction.

Surveys show that a large share of IT leaders and executives—sometimes more than 60%—acknowledge that FOMO significantly influences their AI adoption decisions. This fear is fueled by rapid technological change, assumptions that competitors are gaining an advantage, and limited understanding of what AI can and cannot actually do.

Implementing AI without thoughtful planning or alignment to business needs often results in wasted investments in tools that don’t address real problems. Projects may stall in the early stages or fail to produce any measurable benefit or return on the investment.

Among the biggest challenges with AI centers on data and trust.

When a business puts speed of development above quality and security, it can lead to data errors, AI “hallucinations” and just plain wrong answers that diminish trust in AI systems. Workers may already feel threatened or undervalued, which creates anxiety and slows tech adoption, so care must be taken to not prematurely introduce AI that may further erode trust in the technology.

I’ve always understood that technology isn’t just a tool, it can be a strategic advantage helping businesses gain in ways not previously available. The key is to move away from fear-based adoption and toward a deliberate, value-driven approach.

Start with identifying the real business problem. With AI, figure out what problems you need the technology to solve for you rather than asking what AI can do. Just because AI can do something doesn’t mean you want it to do it for you, or that it will deliver any real value to your process or operation.

Change for the sake of change makes no sense, so it is essential to understand if there is actually a problem that AI may be able to solve and that the benefits of the solution outweigh the cost to develop and the risk potentially introduced. Start small and have pilot projects in low-risk but high-impact areas of the business where the organization can learn and refine before scaling.

Among the most important aspects of AI in business is the data the AI works with. This is where many businesses fail in their initial attempts with AI development, due largely to the fact that data is siloed or segregated and completely unclassified or categorized.

For AI development to deliver effective business benefit, high-quality, organized data and solid data infrastructure are essential.

AI systems learn directly from the data they are given. If the data is incomplete, inaccurate, inconsistent, or poorly managed, the AI’s performance will reflect those flaws. AI models are only as good as their data because AI systems—especially machine learning and generative AI—identify patterns and make predictions based on training data.

Poor-quality data results in biased, unreliable, or incorrect outputs. High-quality data supports accurate, trustworthy, and consistent results. If an AI is trained on inaccurate or inconsistent information, it will learn (and repeat) those errors.

Shift from a fear of missing out to a fear of missing the advantages of AI.

The focus should be on maximizing AI’s potential to create a competitive advantage, taking strategic risks that are aligned with the business goals. Replace fear-driven decision-making with thoughtful, goal-oriented planning and turn AI into a meaningful source of long-term value and differentiation rather than an anxiety-inducing trend to chase.

Noobeh cloud services works on the Microsoft Azure platform, creating data platforms and delivering services that fuel and support AI development. Let us create the dynamic data infrastructure your business needs to develop the intelligence to propel you forward.

jm bunny feetMake Sense?

J