1 min read
7 signs your digital manufacturing system is too much for your shop
Manufacturing software should remove friction from production. It should make it easier to see job status, control work in process, deliver current...
6 min read
Cimx Software : November 18, 2026 at 9:00 AM
The following is a companion article to CIMx's appearance on Holonic's The Dark Arts of Team Optimization. Check out to the full podcast here or listen where ever you get your podcasts.
A production schedule changes. A material arrives late. An experienced operator calls in sick. A customer moves a due date. A quality issue sends a job back for rework. None of these problems becomes simpler because the manufacturer has 40 employees instead of 4,000. The difference is that a smaller company has fewer people, less time, and less tolerance for a software project that takes a year to show value.
That tension is at the center of our conversation with Kyle Johnson and Franz Aliquo on Holonic's The Dark Arts of Team Optimization. CIMx President and Owner Kristin McLane and Marketing Manager Sierra Irvine discuss why we shifted our focus from large enterprise manufacturers to smaller manufacturers, what had to change inside our own company, and why responsible AI begins with reliable production data.
The thread connecting those topics is context. Smaller manufacturers need technology built around the way their teams actually work. Small software companies need operating practices that fit their size. AI needs the right data, boundaries, and human judgment to produce a useful result.
CIMx has worked with manufacturers since 1996. Much of our earlier experience came from complex enterprise environments, where production control, traceability, work instructions, and quality documentation are mission-critical. Over time, we saw that smaller manufacturers faced many of the same operating problems: paper travelers that disappear, spreadsheets that disagree, job status known only by walking the floor, and scheduling decisions made with incomplete information.
What smaller companies often lacked was a realistic path into modern manufacturing software. An enterprise-scale implementation could demand too much budget, time, internal staffing, or process disruption. A lightweight app might be easier to buy but fail to reflect the real dependencies among materials, labor, routings, work instructions, quality, and delivery dates.
Our pivot was therefore larger than choosing a new customer segment. We had to rethink how we packaged, explained, sold, implemented, trained, and supported the software. The goal was to preserve the production discipline learned in complex manufacturing while removing unnecessary weight for a smaller team.
That experience now shapes Quantum, a manufacturing platform designed for small and midsize manufacturers. Its purpose is straightforward: connect production information so people can see jobs, WIP, inventory, schedules, work instructions, quality activity, and performance in the context of the work being done.
The same principle applied to the way we run the company. We did not need to imitate the operating model of a much larger software business. We needed a cadence that made priorities visible, encouraged direct communication, and helped a small team finish meaningful work together.
CIMx now operates on a shared two-week sprint model across the company. Product development is part of that rhythm, but so are marketing, sales, implementation, training, and other business work. For those unfamiliar, the official Scrum Guide defines sprints as fixed-length events of one month or less and emphasizes regular inspection and adaptation. Our two-week cadence is our practical version of that idea.
Using one rhythm does not mean every function does identical work. A software feature, a customer onboarding improvement, and a campaign launch have different dependencies and definitions of done. The shared cadence gives those efforts a common planning horizon. It also creates predictable moments to review what was completed, surface what changed, and decide what deserves attention next.
what changed for us: work became easier to see across functions. A customer question heard in sales or support could reach product and marketing without moving through layers of management. Teams could coordinate around the same near-term priorities, while the review cycle made it harder for lingering work to disappear into an indefinite project list.
what we had to learn: a sprint is not permission to treat every two weeks as a hard release deadline. The podcast conversation also covers how we adjusted release practices to protect quality and avoid burnout. A useful cadence should create focus and learning. It should not turn the whole company into a permanent rush.
Small teams have one structural advantage: the people closest to a problem can often speak directly to one another. That advantage only matters when the company makes priorities and ownership clear. Otherwise, a small team can become a collection of overloaded generalists, each carrying a different version of what matters most.
For us, cohesion comes from a few concrete practices: a visible backlog, a short planning horizon via planned sprints, direct cross-functional discussion, a clear owner for each commitment, and regular review. This reduces handoffs and the time between learning something and acting on it.
There is a manufacturing parallel here. A smaller shop also depends on fast communication, but verbal updates and tribal knowledge stop scaling surprisingly early. When job status, material availability, revision history, or quality information lives with one person, the company is fast only while that person is available. Shared operational context allows a lean team to stay responsive without relying on memory as infrastructure.
AI can help manufacturers search documents, summarize information, identify patterns, support scheduling decisions, draft instructions, and reduce repetitive administrative work. In aerospace and defense, however, the first question cannot simply be, ‘What can this tool do?’ It must also be, ‘What information will it receive, where will that information go, and who remains accountable for the result?’
That is especially important when systems process, store, or transmit controlled unclassified information. Contract obligations, export controls, customer requirements, and the actual system boundary all matter. Manufacturers should verify the rules that apply to their contracts and data with qualified security, legal, and customer contacts.
Responsible use therefore requires more than a generic promise that a model is secure. A manufacturer needs to know what data is allowed, how access is controlled, whether prompts or files are retained, how outputs are reviewed, and whether the use case belongs inside the approved environment. Human judgment should remain explicit wherever an AI output could affect product quality, safety, compliance, delivery, or customer commitments.
For smaller manufacturers, the most important AI investment may be the work that happens before an AI tool is selected. AI cannot reliably answer a production question if job status is stale, material records use inconsistent names, revision context is missing, or the process itself changes from one operator to another without documentation.
In practical terms, an AI-ready manufacturing foundation has six parts:
1. Choose a narrow operational question. Start with a decision that has a clear owner and measurable value, such as identifying late-job risk or finding the current approved instruction.
2. Digitize the relevant workflow. Capture actual job, material, routing, labor, quality, and exception data close to where the work happens.
3. Preserve context and provenance. Keep part, revision, customer, operation, timestamp, and source relationships intact so an answer can be traced back to evidence.
4. Set the information boundary. Decide what data may enter the tool, where it is processed, who may access it, and what contract or regulatory obligations apply.
5. Keep people in the decision loop. Define who reviews an output, what requires approval, and what happens when the system is uncertain or wrong.
6. Measure the result in the workflow. Evaluate accuracy, usefulness, time saved, failure modes, and operational consequences, not simply whether the demo looked impressive.
This sequence is less glamorous than buying an AI feature, but it is how a manufacturer turns automation into a repeatable capability. A connected production system provides both the data and the operational context that future analytics, rules, and AI functions need.
We see CIMx's role as foundational. Quantum brings production activity into a connected system so a manufacturer can establish dependable records for jobs, WIP, inventory, schedules, work instructions, quality, and performance. That shared context helps people run the operation today and creates a more credible base for tomorrow's automation.
This does not mean every smaller manufacturer needs to pursue AI immediately. Many shops can unlock substantial value first by replacing paper travelers, reducing duplicate entry, making job status visible, and standardizing how information reaches the floor. Those improvements are useful on their own. They also make later AI projects less dependent on fragmented data and undocumented assumptions.
The lesson from our own company pivot is similar. Better tools matter, but fit matters just as much. We adapted enterprise manufacturing experience to the resources and pace of smaller manufacturers. We adapted agile practices to the size and shape of our team. We believe AI should be adapted to the context, risk, and information boundaries of the manufacturer using it.
The full episode covers what broke during CIMx's market shift, how we changed onboarding and go-to-market work, how a flat team shares customer insight, and how we are approaching AI inside Quantum. Listen to CIMx on The Dark Arts of Team Optimization.
If your shop is trying to create reliable production visibility before adding more automation, contact the CIMx team for a low-pressure conversation about your current workflow, information gaps, and whether Quantum fits the foundation you need.
You might also like: Manufacturing Scheduling: The Complete Guide to Realistic Production Schedules
1 min read
Manufacturing software should remove friction from production. It should make it easier to see job status, control work in process, deliver current...
Manufacturing has never been static, but the level of uncertainty today feels different. Trade policy shifts, tariff changes, supply chain...
Manufacturing ERP systems can connect accounting, inventory, purchasing, sales, production, and other business functions in one platform. But...