Yet behind the scenes, leadership teams experience a very different reality. The business is becoming harder to run every year. Margins are difficult to predict. Delivery dates feel less reliable. Engineering changes become the norm rather than the exception, disrupting production. As a result, management meetings increasingly revolve around reconciling numbers and explaining surprises instead of planning for the future.

Very few companies hit a wall because of a single, dramatic failure. Instead, they slowly decline because dozens of small operational failures quietly accumulate. Visibility erodes a little at a time. Manual workarounds become permanent. Firefighting becomes the norm. Individually, each issue appears manageable. Together, they create an organization that is increasingly reactive, increasingly fragile, and increasingly dependent on the institutional knowledge held by a few key people who “know how things really work.”

If these symptoms feel uncomfortably familiar, it is a sign that your business has outgrown the stage for which its systems and processes were designed.

Why machinery manufacturing is fundamentally different

Unlike high-volume, repetitive make-to-stock (MTS) production environments, most machinery businesses operate in a hybrid world that blends engineer-to-order (ETO), make-to-order (MTO), and configure-to-order (CTO) models. Even when standard components can be reused, the finished product is often unique in meaningful ways:

• Bills of material evolve as engineering refines designs.
• Routings change based on capacity, material availability, or customer priorities.
• Engineering change orders or changing customer requirements can appear midstream and must be absorbed without derailing the production schedule.
• Lead times for specific raw materials or subassemblies can extend into months or longer.

Problems emerge when the systems and processes used to manage this variability were designed for a much simpler version of the company. As the company grew, the processes that worked when there were fewer jobs, fewer suppliers, and fewer concurrent projects slowly became a bottleneck.

By the time leadership feels the strain of the complexity brought on by growth, the old processes have become deeply embedded in the way the business operates. The core challenge of machinery manufacturing is not the complexity itself; it is the need to manage that complexity deliberately, systematically, and at scale.

Lack of visibility: You don’t know what you can’t see

Many machinery manufacturers believe they have visibility into their business. They can produce financial statements, run backlog reports, and analyze performance. But in many cases, this information arrives after the decisions that mattered have already been made. This creates a visibility trap. On the surface, the organization appears informed. In practice, it is managing the business by looking in the rearview mirror.

Work-in-progress looks healthy in reports, but key jobs are quietly stalled waiting for parts or engineering input. Material shortages are discovered only when production is ready to start. Margin problems are uncovered at the end of the job, when there is no time to address them. Leadership meetings become exercises in reconciling whose numbers are right rather than discussing what should be done next.

In addition, many manufacturers lack comprehensive visibility into their supply chains, making it difficult to identify vulnerabilities and respond to disruptions effectively. True operational visibility is not about dashboards or reports. It is about having access to real-time information that matters, knowing what is happening while you can still change the outcome. When data is delayed, fragmented, or manually assembled, the organization becomes reactive by definition. Decisions are always one step behind reality.

The deeper issue is not a lack of reporting. It is the lack of a shared, real-time understanding of what is actually happening. Without that visibility, even the best management team is forced to manage by approximation and guesswork.

Job costing in custom manufacturing: The silent profit killer

Job costing in custom and project-based manufacturing is uniquely challenging. Engineering time is often under-allocated or treated as overhead. Change orders are tracked operationally but are not always accurately reflected in financial performance. Overhead is spread across multiple projects using broad averages that mask each project’s actual cost. Rework, expediting, and disruption disappear into general expense buckets.

The result is that pricing decisions, customer strategy, and product investment choices are made based on distorted signals. Some jobs that appear profitable are not, while others that appear unprofitable are actually carrying the business.

Accurate job costing is not an accounting exercise. It requires that material, labor, overhead, and engineering effort be accumulated accurately as the job progresses. It requires that change orders are visible as both schedule events and financial events. And it requires that variances are addressed while there is still time to correct them.

Without this real-time visibility, margin erosion becomes a slow, silent leak. The business may grow, but it becomes progressively less predictable and less resilient. The most dangerous part is that the organization often does not realize this is happening until profitability is already under pressure.

Scheduling chaos: When growth breaks planning

In many young organizations, planning starts as an informal coordination exercise. As the business grows, spreadsheets appear. Then, basic MRP tools are introduced. Eventually, capacity constraints become a problem, and scheduling tradeoffs become unavoidable.

Several symptoms begin to appear when this happens:

• More and more orders must be expedited.
• Promise dates slip frequently.
• Engineering, purchasing, and production optimize for their own priorities rather than for the business as a whole.

At this point, scheduling is no longer just a planning problem. It is a coordination problem across the entire organization. Sales is selling one reality. Engineering is designing another. Purchasing is reacting to shortages. Production is doing its best to keep things moving. No one is wrong, but the system as a whole is out of alignment. The result is late deliveries, lost trust, internal friction, and a constant sense that the business is running faster only to stay in the same place.

Inventory: Your biggest asset…and your biggest cash drain

Because most machinery manufacturers operate in engineer-to-order, make-to-order, or hybrid models, their inventory and supply chain challenges are fundamentally different from those of high-volume, make-to-stock manufacturers.

In theory, inventory exists to protect the business. It ensures production continuity, absorbs supply chain variability, and allows commitments to be met with confidence. In practice, for many machinery manufacturers, inventory becomes one of the most significant sources of financial drag, operational friction, and strategic risk.

Long lead times and the commitment trap

Many of the most critical components in machinery manufacturing have long or unpredictable lead times. Some are machined-to-order. Some are imported, and, in some cases, are sole-sourced from highly specialized suppliers. Often, these specialized parts must be ordered weeks or months before final designs are finalized.

This forces planners into a difficult position. They must commit to purchases far in advance of certainty. When designs evolve—as they almost always do—forecast errors are amplified. A single late or incorrect component can stall an entire job, tie up work-in-progress, and cause late deliveries downstream.

Over time, organizations adapt in predictable ways. They overbuy “just in case.” They expedite constantly. They carry excess safety stock for the wrong parts. These behaviors feel prudent in the moment, but collectively they inflate inventory, tie up capital, and still fail to deliver reliability.

When engineering changes impact inventory

In ETO and MTO environments, demand is not driven by sales alone. It is also driven by engineering decisions. Bills of material evolve. Specifications change. Customers revise requirements. Engineering refines designs after projects are already in motion. This is not a failure of discipline; rather, it is the nature of complex, custom manufacturing.

But the inventory consequences are severe. Parts that were already ordered become obsolete, excess, or usable only for “some future job.” New parts must be sourced urgently and often expedited at a premium cost.

There is poor visibility into which materials are tied to which projects. And there is often no clear distinction between project-specific inventory and shared, reusable stock. None of this shows up in forecasts. Yet all of it shows up in cash flow and delivery performance.

Inventory problems in machinery manufacturing are almost always a result of poor coordination and visibility. And until that coordination problem is addressed at a systemic level, inventory will continue to behave like both the business’s greatest asset and its greatest cash drain.

What happens when the business outgrows its systems

Growth changes the shape of the business. New product lines introduce new parts, routings, and suppliers. New facilities introduce more coordination challenges. More customization increases engineering workload and planning complexity. Mergers and acquisitions multiply data, processes, and reporting requirements overnight.

Systems and processes that worked perfectly well when the company started now become constraints as the business grows. At this stage, many organizations feel busy but not in control. Growth continues, but it becomes harder to conduct business as usual.

Traceability, compliance, and rising expectations

Traceability is no longer confined to heavily regulated industries. Customers expect detailed documentation, configuration history, and the ability to trace components, assemblies, and changes. In many markets, traceability has become a prerequisite for doing business rather than a differentiator.

Manual systems and processes can handle this for a while, but eventually they break under the weight of volume, complexity, and audit pressure. When traceability fails, the consequences are not limited to compliance risk. They include lost customer confidence, expensive investigations, and operational paralysis when questions cannot be answered quickly and confidently.

It’s not just one problem

Often, machinery manufacturers attempt to fix these challenges one at a time. They buy a better scheduling tool. They use a better inventory model. They create a better costing spreadsheet. Or they implement a new reporting package.

But the problems they are trying to address are not independent; they are interconnected: You cannot sustainably fix scheduling without fixing material coordination. Fixing margins also requires fixing visibility. And you cannot fix planning without fixing data integrity.

Point solutions that address a specific issue often improve one area at the expense of another, lack integration with other business systems, and create new data silos. Over time, the organization becomes a patchwork of workarounds that only a few people fully understand. The fundamental constraint is the lack of a unified operational and financial picture of the business.

What scalable manufacturers do differently

Machinery manufacturers who successfully scale their businesses must manage complexity as they grow. When it becomes too difficult to manage growth with existing tools and processes, it’s time to seriously evaluate a new Enterprise Resource Planning (ERP) for machinery manufacturers—not as an accounting system but as an operational coordination platform.

Modern, cloud-based ERP systems are designed specifically for this mid-market reality: Complex, growing, mixed-mode manufacturers who need unified data, real-time visibility, and strong alignment between operations and finance. They allow manufacturers to work from a single, shared version of the truth. Operations and finance are tightly aligned. Decisions are based on current reality, not last month’s results. Processes are designed to scale without relying on heroics. And manual processes and multiple spreadsheets are replaced by automation and real-time data.

Why machinery manufacturers choose Dynamics 365 Business Central

Microsoft Dynamics 365 Business Central is built for exactly the kind of complexity faced by mid-market manufacturing companies. It provides:

• A single, unified system for operations and finance
• Real-time visibility into jobs, inventory, WIP, and margins
• Strong support for engineer-to-order, make-to-order, and hybrid models
• Integrated planning, purchasing, production, and costing
• A modern, cloud-based platform that scales as your business grows
• Seamless integration with the broader Microsoft ecosystem (Power BI, Power Platform, Microsoft 365, and more)
• Implemented on Microsoft’s Azure platform, featuring enterprise-level security, system backup and failover capabilities, and advanced AI capabilities.

What this means for you:

• Sales can be directly linked to customizable production or planned in batches
• Product versioning and material changes can be planned, allowing the full use and phase-out of inventory
• Supplier pricing and lead times are easily visible
• Critical information can be captured from workflows for reporting and optimization decisions

Why ArcherPoint

At ArcherPoint by Cherry Bekaert, we work with machinery manufacturers and project-driven manufacturers who have reached this inflection point.

We don’t start with software. We start by understanding:

• How your business actually works
• Where complexity is creating friction or risk
• Which constraints are limiting growth, visibility, or control
• Whether Dynamics 365 Business Central is the right fit for you
• And what a realistic, staged roadmap should look like

Contact ArcherPoint by Cherry Bekaert today and let us help you move from managing complexity to controlling it so you can grow your business with confidence.

The post Hidden Constraints that Hold Back Growth for Machinery Manufacturers…and How to Overcome Them appeared first on ArcherPoint.

Here are some of our favorite new features in BC28.

Supply chain

Add pictures to item variants

Users can store images showing item variants, such as different sizes and colors, and make them available using the Variants action on the Item Card.

Add the Description 2 field to various pages

Users can personalize various pages using the Description 2 field. This feature can be used to include additional context descriptions to BOMs, routings, production orders, and more.

Approval workflows for requisition worksheets and item journals

This feature extends approval workflows to include item journals and requisition and planning worksheets. Users can send batches from the Item, Physical Inventory, Output, and Consumption Journals.

Reporting

View user and group permissions from all apps and extensions

Administrators can view permissions across all apps and extensions, including filtering data by object, scope, extension, or permission set, providing a central location to view permissions, conduct security audits, or troubleshoot issues.

Development

AL developers can search on data and metadata using semantic search

AL developers can search by similarity to the search term using the new codeunit, “Semantic Data Search”.

Enable Troubleshooting MCP Server for AL

Lets developers use runtime data at breakpoints or errors to troubleshoot MCP Server issues and determine which entries triggered a failure during AL debugging.

Evaluate AL coding agents with BC-Bench

Developers can use the BC-Bench benchmarking framework to help determine which improvements to agent performance are most effective in real-world situations.

Copilot and agents

Get item insights with advanced KPIs and Summary

Copilot provides useful KPI information from the new Item Statistics page. Some of the new KPIs available include inventory valuations, such as current inventory value and expired inventory value, and sales performance, including percentage sales growth rate, net sales, and gross margin. A word of caution: This feature comes with a warning box that states, in part, “AI-generated content might be incorrect.”

Manage tasks from all agents in a dedicated task pane

This feature helps users who rely on AI-driven workflows, such as accountants, salespeople, and managers, display all tasks generated by the agents from a single location.

Review content generated by agents directly on pages

Users can review and approve content generated by agents directly on the pages where they work, keeping the review process within the page context.

Show avatars for record creators and modifiers in lists

This feature facilitates traceability by making it easier to identify which user or agent created or modified a record without opening the record.

Stop all active tasks for a selected agent

This feature lets users stop all active tasks for a selected agent.

Use AI resources for your Copilot extensions

This feature makes it easier to set up and secure AI capabilities. Includes scaling and throttling resources along with tracking and billing for AI resource consumption.

Other notable new features

Other significant changes include the ability to create purchase orders directly from sales orders that are set up for drop-shipment, reverse drop shipments, and match invoice lines to relevant purchase order lines and receipts.

There are several enhancements to Shopify connectivity, including the ability to organize products into logical groups, synchronize product variant images with Shopify, and access to the latest Spotify API.

In addition, the Quality Management extension lets you control when quality checks are performed, specify the parameters of the quality check, define quarantine procedures, generate quality status reports, and provide quality certificates.

Items being deprecated

API v1.0: One of the most important changes in this release is that API v1.0 will be deprecated in BC v28. This API is used to connect Business Central to external systems, such as eCommerce platforms, Power Platform applications (including Power Apps, Power Automate, and Power BI), Azure Data Lake, and CRM integrations. Users, administrators, and developers are urged to migrate to API v2.0.

BC 2025 Wave 2 connector’s API expires June 30, 2026: In 2025 Wave 2, Microsoft updated the Shopify Connector to use a specific Shopify API version, which will only be supported through June 30, 2026.

From Microsoft: “The Shopify Connector released in 2025 release wave 2 (October 2025) relies on API 2025-07, which is supported until June 30, 2026. To continue to use your integration, upgrade to the latest version of Business Central before this date.”

Learn more

2026 Wave 1 adds greater security, auditability, and development features that promise to make working with Business Central much easier for users, developers, and administrators.

Check out Microsoft’s preview page for a complete list of the features included in this release.

Contact ArcherPoint by Cherry Bekaert to learn more about this new release. ArcherPoint also offers help with Business Central upgrades and software updates.

The post Get Ready for Dynamics 365 Business Central 2026 Wave 1 (BC28)! appeared first on ArcherPoint.

A researcher is an agent. An analyst is an agent. Facilitator is an agent. Every sponsor demo featured agents. Every hallway conversation mentioned agentic AI.

Then another attendee said what I had been thinking: there don’t seem to be any common definitions of what “agent” or “agentic” actually means.

Everyone nodded. No one clarified. A week earlier, a technology leader told me some of his employees had saved well-crafted prompts as reusable instructions and genuinely believed they had built an “agent.”

That’s when it clicked. The word agent now covers everything from a smart saved prompt to systems that work while you sleep. If everything is an agent, the word means nothing.

And that confusion is going to cost organizations real money.

Why defining ‘agent’ is genuinely hard

My first instinct was to blame marketing. Vendors labeling everything as agent to ride the hype cycle.

But the problem runs deeper. We struggle to define agent because we cannot see what is happening inside these systems.

I call this the observability problem.

What we can see and control

There are external variables we can observe and manipulate.

Model selection. Choosing between models like Claude, GPT, or Gemini dramatically affects speed, depth, and behavior.

Mode selection. Standard chat versus deep research, extended thinking, or agent mode.

Project configuration. The instructions, documents, and context we provide.

Connectors and tools. MCP servers, plugins, and integrations that extend access to systems.

These are the switches we control. We can test them. We can compare results.

What we cannot see

But internal behavior is largely hidden.

• How many reasoning steps did the model take?
• Did it spawn parallel tasks?
• Which internal tools did it call?
• Did it revise its own work before responding?

The same external inputs can produce very different internal processes. And we often have no visibility into why.

The gap that creates confusion

This gap between what we control and what we can observe creates the definition problem.

Is something an agent because of internal complexity? If so, we often cannot tell.

Or is something an agent because of what we experience—how much autonomy we grant and how independently it operates?

Since internal mechanics are opaque, defining agent by what is happening inside is a dead end. We need to define it by experience. By observable autonomy.

An experience-based framework for agent

After dozens of conversations and experiments, I have settled on a four-level spectrum. The framework focuses on what you experience, not what marketing claims.

Level 1: enhanced chat

Reactive. You ask, it responds.

It may reason deeply, search broadly, or take longer to answer. But nothing happens when you close the window.

Examples include deep research modes, knowledge assistants, and advanced analyst features.

Saved prompts that consistently produce strong results live here. Useful? Absolutely. But still enhanced chat.

Level 2: triggered automation with AI reasoning

A predefined trigger starts a predefined workflow.

AI makes decisions within structured paths.

This is workflow automation with an LLM brain.

Examples include event-triggered assistants, meeting summarizers that generate tasks, and prebuilt business agents.

Sophisticated and valuable. But the path is still largely defined in advance.

Level 3: goal-pursuing agent

You provide a goal. The AI plans and executes steps to reach it.

It decides the path within the guardrails you set. It may adapt based on what it discovers. It may ask clarifying questions or request approvals.

But it is session-bound. When you close the window, it stops.

Examples include multi-step research tasks, computer-use capabilities, and complex tool-driven sessions.

The key question is who decides the next step. At Level 2, rules and triggers decide. At Level 3, the AI does.

Level 4: persistent autonomous agent

This is what most people imagine when they hear the word ‘agent’.

It works across time. It may continue while you are away. It sets sub-goals, adapts when things fail, and maintains state across sessions.

Examples include long-running coding agents or research systems that operate overnight.

This is true persistence and autonomy.

What agent-like behavior actually feels like

Recently, I tested ChatGPT’s Agent Mode with a research task. I gave it a goal and walked away.

Eleven minutes later, it was finished. During that time, it searched, evaluated sources, synthesized findings, and structured results. I was not guiding each step. I was delegating.

That felt different. Not because I could see internal mechanics, but because my experience changed. I gave a destination and waited.

Compare that to a deep research response in standard chat. The output may be equally impressive. The internal complexity may be similar. But experientially, I asked a question and received an answer. That is enhanced chat, not agency.

Platform differences reinforce this confusion. Claude offers strong agentic capabilities through Claude Code for developers, but not in its standard chat interface. ChatGPT brings agent-like behavior into chat through Agent Mode. Capabilities vary by vendor and even by product tier. The word agent does not mean the same thing everywhere.

The diagnostic questions

The next time someone claims something is an agent, ask three questions.

Does it work while I am away?
If no, it is Levels 1 through 3. If yes, it approaches Level 4.

Who decides the next step?
If you direct each interaction, it is Level 1.
If triggers and workflows decide, it is Level 2.
If the AI decides within your guardrails, it is Level 3 or 4.

What happens when it fails?
If it stops and waits for you, it is Levels 1 through 3.
If it autonomously tries alternatives, it is Level 4.

These questions cut through buzzwords and force clarity.

From consumer to informed evaluator

Without a framework, you are a consumer of marketing. You hear ‘agent’ and assume autonomy you cannot verify.

With a framework, you become an informed evaluator. You place capabilities on the spectrum. You match what you are getting to what you actually need.

And here is the overlooked truth: sometimes Level 2 is exactly what you need.

If your problem is routing requests or summarizing meetings into structured actions, a persistent autonomous system may be unnecessary. Triggered automation with AI reasoning might be the right tool.

The goal is not to chase Level 4. The goal is to match capability to need.

Why this matters for your organization

Terminology confusion creates real risk.

Misaligned expectations. Leadership expects autonomous agents. You deploy structured workflows. The pilot is labeled a failure.

Wasted investment. You pay premium prices for agentic AI that is effectively enhanced chat.

Missed opportunities. You dismiss Level 2 solutions as unimpressive when they are exactly what your business needs.

Literacy matters. You cannot delegate to something you do not understand. You cannot evaluate what you cannot define.

Your week 22 challenge: cut through the noise

This week, apply the framework.

Find an agent claim in a vendor pitch, product demo, or internal project. Ask the diagnostic questions. Place it on the spectrum. Identify the gap between implication and reality. Then ask whether that level of agency matches your need.

Share what you discover with a colleague. The more leaders who adopt clear definitions, the less money organizations will waste on ambiguous promises.

The bottom line

I left that conference realizing we do not have shared definitions for the term ‘agent’. That confusion will cost organizations money.

We cannot reliably define an agent by internal mechanics because we cannot see them. We must define it by experience. By autonomy granted. By what happens while we are away.

This framework is not about hype. It is about clarity.

If everything is an agent, nothing is an agent. And you cannot invest intelligently in a word that nobody can define.

This post is part of my “AI Over 40” series. It first appeared on LinkedIn: AI for the Over 40 [Week 22]: When Everything Is an Agent, Nothing Is an Agent

Read more AI and Copilot blogs.

The post AI for the Over 40 – Week 22: When Everything is an Agent, Nothing is an Agent appeared first on ArcherPoint.

In this client testimonial, CKS Packaging shares how a ransomware event accelerated its move from on-premises Business Central to Microsoft Dynamics 365 Business Central SaaS, and how ArcherPoint helped the team restore critical operations and move forward in just days.

Manufacturing operations disrupted by ransomware

CKS Packaging, a manufacturer of plastic containers, experienced a ransomware attack that impacted a large portion of the business, including its on-premises Dynamics 365 Business Central system. The disruption affected several critical functions, including:

• Sales order processing
• Shipping operations
• Accounts payable and receivable
• Customer communication and tracking

While production could continue, the inability to access core systems created immediate business pressure. As the testimonial makes clear, shipping products is only part of the equation—processing transactions, tracking customer information, and getting paid are just as critical.

Accelerating a cloud ERP strategy

Before the ransomware event, CKS Packaging had already planned to migrate to Business Central SaaS over 12 to 18 months. That timeline changed dramatically when the business needed to restore access to critical systems as quickly as possible.

With ArcherPoint’s support, what had been a long-term cloud roadmap became an accelerated recovery effort. The team helped:

• Stand up a temporary SaaS environment
• Configure permissions and access
• Restore the most important business functions first
• Support the transition back into the primary tenant after recovery

That combination of speed and prioritization allowed CKS Packaging to move from a long-range SaaS plan to an urgent deployment in a matter of days.

Why cloud ERP matters for business continuity

One of the clearest lessons from the experience was the value of cloud-based ERP. For CKS Packaging, moving critical systems off-premises created greater separation between the internal IT network and the business systems needed to keep operations running.

The move to SaaS was no longer just about modernization. It became a practical business continuity decision—one that improved resilience and reinforced the importance of protecting the systems that support orders, customer service, and revenue flow.

The value of a responsive manufacturing partner

Throughout the event, ArcherPoint’s responsiveness and familiarity with the business were critical. The team was able to:

• Identify which systems and processes need to be restored first
• Mobilize the right resources quickly
• Provide real-time support during a high-pressure situation

CKS Packaging also emphasized ArcherPoint’s understanding of manufacturing operations. Rather than acting as a generalist partner, the team brought process awareness, business context, and proactive recommendations shaped by years of working together. That combination of technical expertise and manufacturing knowledge helped drive better decisions when timing mattered most.

A partnership built on responsiveness and trust

Over time, the relationship between CKS Packaging and ArcherPoint has grown beyond a typical vendor engagement. Across support, consulting, and strategic guidance, ArcherPoint has remained:

• Responsive across teams
• Collaborative in solving problems
• Invested in the business’s long-term success

That level of partnership was especially important during a disruptive event, but it also reflects something broader: when a partner understands your business, your priorities, and your processes, they are better equipped to help when the stakes are highest.

Watch the client testimonial

If your manufacturing organization is evaluating cloud ERP or preparing for unexpected disruptions, learning from real-world recovery and migration experiences can help you plan more effectively. Reach out to ArcherPoint by Cherry Bekaert to explore how a responsive, experienced partner can support your ERP strategy.

Video transcript

My name is Charles Massey. I’m the Vice President of Information Technology at CKS Packaging.

I’ve been here about five years now. It’ll be five years in March.

CKS Packaging produces plastic containers—everything from dairy bottles to industrial gallons, and anything in between in the plastic packaging space.

For us, I think it’s the fact that you guys have people on your team who have been with you for a long time. All the customizations and different things we’ve developed in partnership with ArcherPoint have really driven that relationship.

I think, for us, it’s the interest in understanding the end result—the end goal of what we’re trying to get to. It’s not just developing something and passing it off, but trying to understand the business process involved and really digging into what the need is before starting down a path to resolution.

We did have a ransomware event in December of 2023. It hit a large portion of the enterprise, including our on-prem version of Business Central. Obviously, it affected every part of the business that runs through there—from sales orders to shipping, AP and AR processes as well.

It was a very stressful event. For our business, we can continue to make plastic bottles, but actually getting paid when we ship those plastic bottles is just as important.

There was a lot of tracking—getting customer information out of the system when the system wasn’t available—and keeping customers happy was critical to us.

Again, back to the relationships that have been built with some of the guys at ArcherPoint internally who knew what we needed and what was critical to the business.

As soon as we called Quentin, we were on the phone with the right people within the team, figuring out what we needed to do and the best path forward to get back into the systems.

It’s an interesting thing. We had a roadmap to do SaaS over 12 to 18 months, and then we decided we’d do it in five days.

ArcherPoint and Quentin helped get the right resources on the phone, stand up the SaaS environment, configure it, set up permissions in a temporary tenant, and then migrate everything back into the primary tenant once ransomware recovery was complete. That was a big help.

The cloud-based approach is critical. It’s the air-gapped piece of it. You want separation between what we call your IT network and those critical business functions.

That was one of our lessons learned from the ransomware event—to take everything we can off-prem and push it into the cloud.

Again, it comes back to having the right partner—understanding what’s critical to the business. It all comes down to having a feel within that partner so they understand what the important parts are.

Do we need to get everything back up? Yes. But did ArcherPoint instantly know what the most important pieces were to restore first? Yes. That partner relationship really helped drive that.

I think it’s really important for a partner to understand what you do. There are a lot of partners out there that are generalists, but it’s very evident from my experience with ArcherPoint that they have resources who have been involved in the manufacturing process before.

I think that’s a critical piece—to find partners and create relationships with people who have done what you do.

Absolutely. Anytime I’ve ever reached out to anybody at ArcherPoint—whether on the sales side, the support side, or whatever it may be—I’ve gotten an instant response. That’s critical in business these days.

I think ArcherPoint is a great partner to work with around Microsoft technologies. They’ve come to the table with things we didn’t even think of, which is important too.

They make suggestions based on the relationship they’ve built with us over the last several years.

The post Ransomware Recovery to Cloud ERP: CKS Packaging’s SaaS Move appeared first on ArcherPoint.

Understanding MRP and MPS

To choose between MRP and MPS, you should understand demand, as it ultimately drives decisions in your manufacturing business. There are two main kinds of demand:

• Independent demand refers to customers’ desire for a product. You can determine independent demand by looking at forecasts, market reports, pre-orders, and orders.
• Dependent demand is the demand for raw materials and parts needed to produce the final product. A bill of materials can help you determine dependent demand.

MPS is driven by independent demand and establishes which products you need to produce, when, and in what quantities. MPS is the “big picture” roadmap that creates better workflows by aligning your manufacturing capacity with demand. It ensures customers get products on time while reducing overstocking or stockouts.

While MPS focuses on the finished product, MRP is a micro-level system that determines the parts, materials, and subassemblies required to produce it. MRP answers to the MPS because it’s driven by dependent demand. MRP improves supply chain transparency and minimizes bottlenecks by making sure parts are available when needed.

For example, let’s say you make winter coats. MPS can help you determine when to produce the most coats for winter sales, how many coats to produce, and which types to produce to meet demand. MRP can help you determine when you need to order zippers, down material, fabric, buttons, and other materials (and in what quantities). 

When to choose MRP or MPS

Both MRP and MPS play an important role, but one may better suit your situation than the other. 

What Are the Benefits of MRP?

MRP may be right if you have a product with many materials, especially perishable materials, or components you need to manage. Owners who are concerned about reducing overstock and like a comprehensive approach to inventory may also choose MRP.

• Improves inventory control: MRP processes make supply chain management easier. They use real-time data to display current supplies, so managers can identify materials that need replenishment before they run out.
• Boosts production efficiency: MRP provides managers with significant insights into the time and materials required to keep production flowing smoothly. They can track and use raw data to minimize delays and miscalculations.
• Reduces purchasing costs: This process automates searches, enabling businesses to access the most cost-effective solutions. By minimizing purchasing costs, companies can offer better product cost-efficiency and reap the bottom-line benefits.
• Implements emergency strategies: MRP helps ensure efficient stock management. It sets strategies to prevent excess inventory by ordering when stock needs replenishing. You can also maintain safety stock in case of emergencies.
• Prevents common production bottlenecks: Businesses can maintain uninterrupted production by identifying potential bottlenecks before they occur.

What Are the Benefits of an MPS?

You might use MPS if you manufacture products driven largely by predictable customer demand. You may also lean more on MPS if you need careful scheduling. The main perks are increased profitability and reduced shortages.

• Simplifies order promises: The MPS provides accurate lead times, ensuring everyone knows how long it takes to create a product. It also prevents utilization-dependent exhaustion that may occur due to work intensity.
• Enhances communication: An MPS serves as a communication tool that helps manufacturing share production plans with other departments.
• Boosts efficiency: An MPS enables analysis and identification of potential bottlenecks, improving customer satisfaction and sales in the long run.
• Streamlines supply chain: An MPS reviews supply chain requirements and prioritizes them, boosting supplier relationships through timely delivery.

When to use both MRP and MPS

While both MPS and MRP are useful in different ways, some manufacturers use both because the two systems complement each other — MPS sheds light on your ultimate production goal, and MRP ensures you can meet it. Integrating them reduces the risk of inventory inaccuracies and missed orders by providing a clearer understanding of the final product’s production goal and what’s needed to meet it. 

If you manufacture bicycles, you may use MPS weekly or monthly to determine how many bicycles to produce and which types, based on seasonal demand, orders, trends, and market research. You might run MRP daily to make sure you have the gears, tires, lights, and bells needed to produce bicycles and meet demand. Using both prevents silos. You can be sure you have everything you need to produce the right amount of bicycles, on time, and in the varieties customers want.

The ERP advantage

Enterprise resource planning (ERP) systems like Microsoft Dynamics 365 Business Central are powerful platforms that help you integrate both MPS and MRP into your business while saving time. An integrated cloud ERP system lets MPS inform MRP using real-time data you can access from anywhere. You get automated processes to improve accuracy, and you can see all your data in one environment to make better decisions about ordering parts and scheduling production of completed products.

With today’s ERP systems, you can track inventory in real time, visualize data, and improve communication between production and other departments.

Learn how ArcherPoint can help

Whether you’re implementing a new ERP system or migrating from a legacy platform, ArcherPoint by Cherry Bekaert has deep industry expertise in Microsoft Dynamics 365 Business Central, Dynamics NAV, LS Retail, Azure, Microsoft 365, and Power Platform. We handle implementation, integrations, training, updates, and IT support, so you can focus on your business. Contact us today for a free consultation.

The post MRP vs. MPS: Choosing the Right Planning Approach for Your Manufacturing Business appeared first on ArcherPoint.

Understanding how manual receiving works in Boltrics WMS for Microsoft Dynamics 365 Business Central gives warehouse teams greater control and visibility into receipt documents, quantities, and pallet-level details before inventory is posted.

In this walkthrough, we break down the manual receiving process step by step — from reviewing receipt headers and lines to posting handling units (pallets) into inventory and completing putaway warehouse activities.

When manual receiving is used in Boltrics WMS

While many 3PL operations rely on handheld scanners for speed and accuracy, manual receiving remains an important workflow for:

• Reviewing inbound documentation before posting
• Adjusting quantities at the pallet level
• Handling exceptions or discrepancies
• Lower-volume warehouse environments
• Administrative control and oversight

Manual receiving provides deeper visibility into the receipt document structure inside Business Central before inventory becomes available.

Reviewing the receipt document structure

The process begins by opening the Receipt List and selecting the appropriate document.

Daniel walks through the key components of the receipt, including:

Receipt header overview

The header provides high-level logistics information such as:

• Arrived date
• Departed date
• Vendor or customer information
• Overall receipt status

These fields provide operational context before any inventory is posted.

Receipt lines overview

The lines section is where item-level detail is managed. Users can review:

• Ordered quantity
• Received quantity
• Outstanding quantity
• Handling units (pallet references)

Understanding the difference between ordered, received, and outstanding quantities is critical for maintaining accurate inventory and avoiding over- or under-posting.

Managing handling units before posting

One of the key advantages of manual receiving is the ability to manage handling units directly within the receipt document.

Before posting to inventory, users can:

• Review pallet-level detail lines
• Adjust quantities as needed
• Confirm what will be posted
• Validate accuracy against physical goods

This level of control is particularly valuable in 3PL environments where billing, traceability, and inventory precision matter.

Created vs. posted: What it means

During the walkthrough, Daniel explains the difference between “created” and “posted” handling units.

• Created handling units exist within the document but are not yet reflected in inventory.
• Posted handling units are officially added to inventory in Business Central.

This distinction helps prevent premature inventory updates and ensures warehouse teams validate quantities before affecting stock levels.

Posting handling units into inventory

Once quantities are verified, users can post the handling units into inventory.

At this stage:

• Inventory levels are updated in Business Central
• Receipt statuses move forward
• The system prepares for warehouse putaway processing

This transition marks the official move from receipt review to inventory availability.

Completing putaway warehouse activities

After posting, Boltrics automatically generates putaway warehouse activities.

Users can then:

• Review the generated warehouse tasks
• Post putaway activities
• Confirm storage locations

This ensures goods are moved from staging into their assigned warehouse locations, completing the inbound process.

Manual receiving vs. scanner receiving

Manual receiving offers more document-level control, while scanner-based receiving emphasizes speed and real-time execution on the warehouse floor.

If you’re looking for a walkthrough of scanner-based workflows, see our complete guide to scanner receiving in Boltrics WMS.

Both processes ultimately support the same goal: accurate inventory and controlled inbound operations inside Microsoft Dynamics 365 Business Central.

Bringing it all together

Manual receiving in Boltrics WMS provides flexibility and control for warehouses that need detailed oversight of inbound documents and pallet quantities before posting inventory.

By understanding receipt headers, quantity fields, handling units, and warehouse activities, organizations can:

• Maintain accurate stock levels
• Prevent posting errors
• Improve inbound documentation control
• Support structured 3PL billing processes

Whether used as a primary workflow or as an exception process alongside scanner receiving, manual receiving remains an important part of a well-structured Boltrics WMS implementation.

The post How to Manually Receive Goods in Boltrics WMS for Business Central appeared first on ArcherPoint.

The Dynamics 365 Business Central community, consisting of developers, project managers, and consultants, collaborates across various platforms to share valuable insights. At ArcherPoint, we greatly value their dedication and expertise. To ensure widespread access to this technical knowledge, we created Developer Digest.

BC Agents and Copilot Credits

Business Central Agents are available through Copilot Credits. Credits are consumed based on the features used and the actions the agent performs. These credits are provided through your BC license, and a certain amount is offered free of charge each month. Additional credits are available from the Power Platform Admin Center.

Stefano Demiliani explains how to monitor your Copilot credit consumption directly in Business Central. You can also control who has access to see this information. To learn more, read his blog,

To learn more, read his blog, Dynamics 365 Business Central: how many Copilot Credits my Agent consumes?

New approval workflows

BC users are accustomed to setting up approval workflows for general journal batches. New in  BC28 is support for batch-level approval workflows for item journals, requisition worksheets, and planning worksheets. These new workflows can also be automated in Power Automate.

Mohana Yadav walks through setting up and using these new workflows in his blog post, Approval Workflows for Item Journals and Requisition Worksheets.

Getting ready for BC28

There are three fundamental questions to ask for any new software release:

• What’s new?
• What’s changed?
• And what’s being removed?

And the upcoming release of Business Central 2026 Wave 1 (BC28) is no exception.

Feature management lets you test new features before releasing them to the new version.

Saurav Dhyani gives a great overview of what to expect in the new release, along with how it will affect developers, consultants, and customers. Most helpful is his advice that users should not rush to activate every new feature, but rather work with their partner to identify those features that will have the greatest impact on their business.

Check out Savrav’s video, Business Central 2026 Wave 1: What’s Changing, Deprecated & What You Must Do, to find out more.

Are you interested in Dynamics NAV and Business Central development? Check out our collection of NAV/BC Development Blogs.

Read “How To” blogs from ArcherPoint by Cherry Bekaert for practical advice on using Microsoft Dynamics NAV and Dynamics 365 Business Central.

The post Dynamics Business Central / NAV Developer Digest – Vol. 550 appeared first on ArcherPoint.

Retailers are now navigating a landscape defined not by recovery but by permanent behavioral shifts, new economic realities, and heightened expectations.

Consumer behavior has fundamentally changed

One of the most enduring impacts on retail operations after COVID is the shift in consumer behavior. During the pandemic, consumers rapidly adopted new shopping habits out of necessity, particularly online purchasing, contactless transactions, and home delivery. What initially appeared to be temporary has largely become permanent.

eCommerce adoption, in particular, experienced a dramatic surge. Global online retail sales grew more than 26% in 2020 alone, a leap that would typically take several years to achieve. Even more telling, a significant majority of consumers who tried online shopping during the pandemic have continued using it, effectively cementing digital commerce as a core part of everyday life.

Today, online shopping has become an expected option from retailers. Roughly 70% of Americans now shop online monthly, reflecting how deeply digital channels have become embedded in consumer behavior.

At the same time, the pandemic reshaped how consumers think about value. Economic uncertainty, inflation, and shifting priorities have made shoppers more price-sensitive and selective. Many consumers are trading down, switching brands, or delaying discretionary purchases. Brand loyalty has weakened, replaced by a more pragmatic, value-driven mindset.

The rise of hybrid retail

While eCommerce has become more ubiquitous, physical retail has not disappeared. Instead, the industry has settled into a hybrid model where digital and in-store experiences are deeply interconnected.

Consumers now expect flexibility. They want the ability to browse online, pick up in-store, return items anywhere, and receive products quickly and conveniently. This has turned omnichannel capabilities from a competitive advantage into a baseline requirement.

Retailers have had to respond by integrating their operations across channels. Inventory, fulfillment, pricing, and customer data must now flow seamlessly between online and physical environments. The distinction between “eCommerce” and “in-store retail” is increasingly irrelevant; what matters is the overall experience.

In the process, this shift has also redefined the role of physical stores. Instead of being purely transactional spaces for sales, stores now serve multiple purposes: fulfillment hubs, brand experience centers, and customer service touchpoints.

Supply chain lessons that still apply in retail

COVID-19 exposed significant vulnerabilities in global supply chains. Delays, shortages, and unpredictable demand forced retailers to rethink how they source, store, and distribute products.

Although supply chains have stabilized in many areas, the lessons remain. Retailers are now placing greater emphasis on visibility, flexibility, and resilience. This includes diversifying suppliers, holding more strategic inventory, and investing in better forecasting tools. As a result, many retailers are better prepared to respond quickly to supply chain disruptions, whether driven by global events, economic shifts, or sudden changes in consumer demand.

Retail workforce and labor dynamics

Another lasting impact of the pandemic is the transformation of the retail workforce. Labor shortages during and after COVID forced retailers to rethink staffing models, compensation, and automation strategies.

Many retailers are now operating with leaner teams while relying more heavily on technology to fill gaps. Self-checkout systems, mobile POS devices, and automated inventory tools have become more common, not just as convenience features but as operational necessities.

At the same time, expectations for store employees have increased. Associates are now expected to handle a wider range of responsibilities, from fulfilling online orders to delivering personalized customer service. This requires better training, better tools, and more structured processes.

The data imperative

Perhaps the most important long-term shift is the growing importance of data. During the pandemic, retailers were forced to make rapid decisions amid uncertainty. Those with better data and analytics capabilities were better positioned to respond.

That reality has not changed. Today’s retail environment remains volatile, and decision-making must be faster and more precise than ever. Retailers are investing heavily in data platforms, analytics, and AI to improve forecasting, pricing, inventory management, and customer engagement.

Data has become a strategic asset, enabling retailers to anticipate trends, personalize experiences, and optimize operations in real time.

A new definition of resilience

The pandemic redefined what it means for a retailer to be resilient. It is no longer enough to have strong products or prime locations. Resilience now depends on the ability to adapt quickly, operate efficiently, and deliver consistent experiences across multiple sales channels.

Retailers that are succeeding in this new environment share several common traits: They are more agile in their operations, more disciplined in their use of data, and more focused on delivering value to increasingly selective consumers. They also view technology not as a support function, but as a core enabler of their business model.

The retail industry has entered a new phase

COVID-19 was not just a temporary disruption; it was a catalyst for long-term transformation. Retailers are now operating in a new environment defined by hybrid commerce, heightened expectations, and constant change.

Contact ArcherPoint by Cherry Bekaert to learn how we can help you become more competitive in this rapidly changing environment.

The post How the Residual Impact of COVID-19 Still Shapes the Retail Industry Today appeared first on ArcherPoint.

Today, Model Context Protocol (MCP) is creating a similar inflection point for AI. Powerful AI capabilities exist across countless services, databases, and applications—but accessing them requires custom integrations, proprietary APIs, and specialized knowledge for each system. MCP servers are poised to become what websites were to the internet: a standardized window that makes the world of AI accessible, navigable, and genuinely useful.

The parallel that matters

The comparison between websites and MCP servers isn’t merely metaphorical—it’s architectural. Just as websites provide a standardized interface layer between users and internet resources, MCP servers provide a standardized interface layer between AI agents and the technological landscape they need to navigate.

Before websites, accessing internet resources meant knowing specific protocols: FTP for file transfers, Gopher for document retrieval, SMTP for email, and so on. Each required different commands, different authentication methods, and different mental models. The web didn’t replace these underlying systems—it abstracted them. HTTP and HTML created a universal language that could represent and access all these resources through a consistent interface.

MCP servers do the same thing for AI. Your organization’s data lives in databases, your business logic runs in applications, your files sit in cloud storage, your communications flow through messaging platforms. Each has its own API, its own authentication scheme, its own idiosyncrasies. MCP servers abstract these complexities, presenting AI agents with a consistent protocol for discovering and interacting with any resource, regardless of what’s happening under the hood.

What MCP servers actually are

At its core, an MCP server is a lightweight service that exposes specific capabilities—data sources, tools, or computational resources—to AI agents through the Model Context Protocol. Think of each MCP server as a specialized website, but instead of serving HTML to web browsers, it serves structured capabilities to AI systems.

The protocol defines three primary types of resources an MCP server can expose:

• Resources are data sources—database records, file contents, API responses, and real-time sensor data. When an AI agent needs information, it can query MCP servers to discover available resources and request access to them via a standardized interface.
• Tools are actions the AI can take—creating calendar events, sending notifications, running calculations, and updating records. Rather than hardcoding every possible action into the AI system itself, MCP servers expose these capabilities dynamically, allowing the AI to discover and use them as needed.
• Prompts are reusable templates or workflows that encode best practices and domain expertise. They enable organizations to capture institutional knowledge about how to approach specific tasks and make it accessible to AI agents.

The genius of this architecture is that the AI agent doesn’t need to know whether it’s talking to Salesforce, SAP, a PostgreSQL database, or a custom internal system. It just needs to speak MCP. The server handles the translation.

The abstraction layer that changes everything

Here’s where MCP servers become truly transformative: they create an abstraction layer that decouples AI capabilities from system specifics.

Consider a business analyst who wants an AI agent to help forecast inventory needs. This requires data from your ERP system, current warehouse levels from your WMS, historical sales patterns from your CRM, and real-time market data from external APIs. Without MCP, you’d need custom integrations for each system, each requiring maintenance as APIs evolve, security protocols change, and systems upgrade.

With MCP servers, you deploy one server per system, each running the standard protocol. Your AI agent doesn’t call SAP APIs or authenticate with Salesforce—it calls MCP servers. When SAP releases a new API version, you update the MCP server implementation, not every AI application that uses SAP data. When your authentication scheme changes, you update it in one place.

This abstraction delivers the same benefits that websites brought to the internet: standardization enables innovation at the edges. Just as web browsers didn’t need to understand every database backend or application server, AI agents don’t need to understand every enterprise system. They just need to understand MCP.

How AI agents navigate your systems with MCP

The navigation metaphor extends deeper than mere access. MCP servers enable AI agents to discover and reason about your technological landscape dynamically, much like search engines and hyperlinks enabled navigation of the web.

When you point an AI agent at a set of MCP servers, it can query them to understand what’s available: What resources can I access? What tools can I use? What workflows are recommended? This discovery mechanism means AI agents can adapt to changing environments without reprogramming. New capabilities become available simply by adding new MCP servers to the environment.

In your Business Central implementations, imagine MCP servers exposing warehouse locations, production orders, inventory levels, and costing data as resources, while providing tools to create pick lists, post production journals, or generate custom reports. Your AI agents could then navigate these capabilities intelligently—understanding that production order data relates to inventory consumption, that warehouse picks connect to shipment posting, that lot tracking requires specific sequencing.

The AI doesn’t need Business Central expertise hardcoded into it. It discovers these relationships through MCP, the same way a user discovers website relationships through navigation and links.

MCP server security: access control and governance

Just as websites introduced both opportunities and security challenges to the internet, MCP servers raise important questions about access control and data security. But the standardization actually helps here.

MCP servers implement authentication and authorization at the server level, allowing organizations to enforce consistent security policies. An MCP server can refuse to expose sensitive financial data unless the requesting AI agent presents proper credentials. It can rate-limit requests to prevent abuse. It can audit all interactions for compliance purposes.

Because the protocol is standardized, security tooling can be built once and applied across all MCP servers in your environment. This is far superior to the current state where each custom AI integration requires bespoke security implementation.

The ecosystem emerges

What made websites truly revolutionary wasn’t just the technology—it was the ecosystem that emerged. Anyone could create a website. Standards bodies ensured interoperability. Search engines made discovery possible. Commerce, education, entertainment, and social connection all found their expression through this common medium.

We’re seeing the beginning of a similar ecosystem around MCP. Major platforms are implementing MCP servers—HubSpot, Microsoft 365, and other connectors. Open-source communities are building MCP servers for popular databases, APIs, and services. Organizations are developing internal MCP servers that expose proprietary systems through the standard protocol.

As this ecosystem grows, the value of any individual AI agent increases exponentially, because it can suddenly access an ever-expanding world of capabilities through a protocol it already speaks. Network effects take hold.

A window opens

The internet transformed from a technologist’s tool to a universal platform not because the underlying technology changed, but because websites provided an accessible window into its capabilities. MCP servers are opening a similar window to the world of AI.

They won’t replace the underlying complexity of enterprise systems, APIs, and data sources—just as websites didn’t replace databases and application servers. But they will abstract that complexity behind a standard protocol that AI agents can navigate intelligently and dynamically.

For organizations, this means AI capabilities that scale across their technological landscape without custom integration for every scenario. For developers, it means building capabilities once that work with any compliant AI system. For users, it means AI agents that can actually accomplish complex tasks across multiple systems without requiring a team of integration specialists.

The window is opening. The question isn’t whether MCP servers will become the standard interface between AI agents and the systems they need to access—the architecture is too elegant and the benefits too compelling. The question is how quickly organizations will recognize this moment and position themselves to take advantage of it.

Just as early web adopters gained tremendous advantages by understanding how to leverage websites effectively, early MCP adopters will gain advantages in deploying AI capabilities across their technology stacks. The world of AI is vast and growing. MCP servers are becoming our window into it.

The post MCP Servers: Your Window to the World of AI appeared first on ArcherPoint.

After 20 weeks of exploring AI as a business leader—not a developer—I finally understood how to think about MCP servers. More importantly, I realized I may need to build one myself.

If you’ve heard the term “MCP server” and quietly nodded without fully understanding it, this article is for you.

What is an MCP server?

Earlier in this series, I described AI as a master improv actor—able to play any role you need.

But there’s a limitation. Your actor only knows what you tell them. Every conversation starts with you providing context, uploading files, or pasting data.

An MCP server changes that.

MCP stands for Model Context Protocol. In practice, it’s a bridge that lets your AI assistant access external systems directly.

Without an MCP server, you export a report, upload it into Claude or ChatGPT, explain the columns, and then ask your question.

With an MCP server, you ask the question, and the AI retrieves the data directly from the source system.

That’s not just convenience. It’s the difference between working from static snapshots and accessing live business data.

My first experience with Microsoft Learn

My first meaningful exposure to MCP servers was through the Microsoft Learn MCP server.

For our developers, consultants, business analysts, project managers, and support team, it has been transformative. Instead of manually searching documentation and piecing together answers, we can bring Microsoft’s published guidance directly into our AI conversations.

We can validate customization strategies. We can confirm whether functionality already exists before building something new. We can align our designs with Microsoft’s principles.

The AI doesn’t just search documents. It synthesizes guidance in the context of our specific client situation.

It became essential to how we work.

But it also shaped my thinking in a limiting way. I began to assume MCP servers were primarily for knowledge bases.

The abstraction gap

My next experiment was the Zapier MCP server. That’s where I ran into friction.

I assumed the MCP server would expose everything available through Zapier’s native connectors. It didn’t.

When I built my Outlook-to-Todoist automation in Week 18, I couldn’t use the MCP server. I had to build the automation directly in Zapier because the MCP layer exposed only a subset of functionality.

That’s when I understood something important: an MCP server does not have full API access. It’s an abstraction layer built on top of an API.

Whoever builds the MCP server decides what to expose. Their priorities may not match yours.

I call this the abstraction gap—the distance between what a system can do and what an MCP server allows AI to access.

This means an MCP server doesn’t guarantee full system access. Native integrations may offer more capabilities. And if a published MCP server doesn’t solve your problem, you’re back to finding alternatives.

It also means you should only use MCP servers from trusted sources. Given the right permissions, they can create, read, update, and delete records.

Where I got stuck

After the success with Microsoft Learn and the disappointment with Zapier, my thinking narrowed.

I viewed MCP servers as advanced connections to specialized knowledge bases. If I couldn’t identify another “knowledge base” to connect, I stopped exploring.

Meanwhile, I was spending hours each month on manual work.

The report problem

My budgeting and forecasting analysis lives in Business Central and other systems. The data exists. But accessing answers required waiting for someone else to build reports—or exporting data and manipulating it myself.

When waiting wasn’t an option, I would export data, open Excel, copy and paste, write formulas, restructure accounts, and rebuild analysis from scratch. Every month, something changed. New accounts. Updated requirements. Different structures.

I had normalized that inefficiency.

The breakthrough

One afternoon, while manipulating yet another spreadsheet, it clicked.

The data in Business Central is already available through APIs and OData feeds. I have been manually extracting and transforming information that is already programmatically accessible.

I don’t need to wait for someone else to publish an MCP server for my specific needs. I could potentially build one.

A custom MCP server that connects directly to the data sources I use would allow me to query live data from within an AI conversation, provide business context, iterate on analysis in real time, and maintain month-over-month continuity rather than starting from scratch.

This is not about novelty. It’s about leverage.

The honest caveat

I haven’t built this yet.

Knowing something is theoretically possible is different from proving a non-technical business leader can accomplish it with AI guidance. This is more ambitious than previous automations I’ve tackled.

But I’m committing to trying. I’ll use AI as my advisory board to help me design and build a custom MCP server in Azure that connects to the data I need.

If I succeed, I’ll document how. If I fail, I’ll share what I learned. Either way, it moves the conversation forward.

The consumer-to-creator shift

The real insight this week isn’t technical. It’s mindset.

For months, I was waiting for someone else to build reports. Waiting for someone else to publish the right MCP server. Filling the gap with manual Excel work because that was familiar.

Now I’m asking a different question: if the data has an API, why can’t I connect AI to it directly?

That shift—from consumer to creator—is where real leverage begins.

A note on platforms

I’ve referenced Claude specifically because it currently makes MCP connections accessible directly within the chat interface.

Other platforms allow similar capabilities but with more friction. Gemini typically requires an integrated development environment or a custom application. ChatGPT requires custom connectors in development mode, which disables memory. Microsoft Copilot requires extension through Copilot Studio.

All can support advanced integrations, but Claude currently lowers the barrier for business leaders who are not developers.

Your week 21 challenge: map your data sources

This week, take a practical step.

Identify the manual reports you rebuild regularly. Trace where that data actually lives. Consider whether an MCP server already exists—and whether it exposes what you truly need. Then ask a different question: could I build one?

Start with a single recurring business question that consumes time each month. That’s your entry point.

The bottom line

MCP servers aren’t magic. They are bridges between AI and your business systems.

The bridges others build may not go where you need to go.

But you don’t have to wait for someone else to build yours.

I don’t yet know if I can pull this off. But I’m done waiting to find out.

This post is part of my “AI Over 40” series. It first appeared on LinkedIn: AI for the Over 40 [Week 21]: What MCP Actually Are and Why I’m Building My Own.

Next Up: When everything is an agent, nothing is an agent.

Read more AI and Copilot blogs.

The post AI Over 40 Series – Week 21: What MCP Servers Are (and Why I’m Building My Own) appeared first on ArcherPoint.