Digital transformation in manufacturing: From Excel to IoT, automation, and AI

Digital transformation is accelerating. But is it delivering results?

According to Gartner, up to 80% of manufacturing CEOs are increasing their investments in new technologies (like AI, IoT, and data analytics) to meet economic challenges and accelerate growth. Despite economic uncertainty, 85% of companies are maintaining or even expanding budgets for Industry 4.0 initiatives.

On the other hand, statistics show that only 8% of digitalisation projects in manufacturing are fully successful. The main reason isn’t a lack of access to advanced digital technologies, but insufficient process preparation and weak organisational support.

A lack of strategy and the misunderstanding that digitalisation is a process – not a sudden revolution – often leads to wrong investments and confusion. A gradual, well-planned introduction of new technologies reduces the risk of failure and allows companies to enter manufacturing digital transformation and Industry 4.0 with greater confidence.

Why move beyond Excel in manufacturing processes?

Many factories still rely on spreadsheets, paper reports, and scattered data sets. While Excel is considered a universal tool for production planning, downtime tracking, or quality monitoring, too many files and manually compiled reports quickly lead to information chaos and create data silos.

According to research, 42% of manufacturing companies have not yet started digital transformation, and 70% of data generated by manufacturing operations is never used.

Only 16% of manufacturing executives monitor production processes in real time. The rest rely on manually gathered data that becomes outdated almost immediately – making decisions difficult and adding unnecessary costs. Without a unified view of what’s happening, many managers fail to see the full potential for improvement.

The “Digital Transformation in Manufacturing: Trends and Challenges” study suggests, that two-thirds of manufacturers admit they’re lagging behind in digitalisation, while market leaders who have embraced digital solutions are gaining a distinct competitive advantage. A Deloitte report notes that enterprises with higher digital maturity achieve stronger financial performance and better profit margins. The lack of a consistent approach to data and processes translates directly into losses.

When handwritten reports, scattered data and legacy software create more problems than they solve, the natural next step is to streamline existing resources and build a solid foundation for further digital transformation in manufacturing.

Stage 1: Systems integration and basic production digitisation

The first step in production digitisation is to create a unified data foundation. While this may sound revolutionary, it’s often more practical to focus on digitising manufacturing processes and integrating what you already have – rather than replacing all your infrastructure at once.

In many cases, you can simply link Excel spreadsheets to a database, integrate your existing ERP system with production, or implement a Manufacturing Execution System (MES) module that automatically collects data from machines and production lines.

Formulating a solid implementation strategy is crucial to ensure efficiency and security as you progress.

At this stage, three main actions matter most:

• Data centralisation – creating a common database where information from ERP (Enterprise Resource Planning), Excel, or warehouse systems converges.
• Electronic documentation – replacing paper sheets with tablets or other devices that feed information directly into the system in real time.
• Basic monitoring – creating dashboards of key indicators (e.g. line efficiency, downtime, defect rates). Even if they are only updated once a day, they provide a consistent view of current operations.

In practice, many companies use supervisory control and data acquisition (MES/SCADA) tools or ‘middleware’ to gather data from multiple sources into one place. This prevents endless copying between spreadsheets and enables a near-real-time view of processes.

For example, a Polish ultra-fresh food producer replaced scattered Excel spreadsheets with a single, integrated production planning application. By automatically pulling data from the company database and providing a clear dashboard, multiple team members can now work simultaneously – without the risk of file conflicts or copy-paste errors.

Even these initial digitisation measures often have a tangible impact: it becomes much easier to identify the causes of downtime and fix them quickly, increasing efficiency and building trust within the team. The tools introduced at this stage provide a strong foundation – both technologically and in terms of employee engagement – for the next phases of manufacturing digital transformation, such as IoT, automation or AI.

If you’re taking your first steps towards digitalising your production, organising your data or making life easier for your team, find out how we help companies to digitise and optimise their processes.

Stage 2: IoT and manufacturing process automation

Once your factory data is consolidated, the next move is to implement Internet of Things solutions that take digital transformation in manufacturing to a higher level. These solutions collect real-time data from machines and automatically respond to specific events.

Manufacturing companies are increasingly installing connected sensors on their machines. This enables faster response to changes, helps reduce downtime, and ultimately increases flexibility across the factory.

Key elements of this stage include:

• Sensor technology & IIoT – fitting machines with sensors (e.g. for temperature, vibration, or load) and connecting them to IoT platforms (e.g. Siemens MindSphere, Azure IoT). Data flows continuously to the cloud or a local server, giving operators a live view of machine status.
• Advanced IoT platforms – for example, dashboards that display real-time performance, production rates, or cycle times (such as the popular Andon boards). Managers and operators can immediately identify deviations from the norm.
• Reactive automation – setting up alerts and automated actions, such as sending notifications to maintenance teams when anomalies occur, switching on a backup machine, or deploying AGV robots to deliver parts. This keeps the factory agile in the face of unexpected events.

Harley-Davidson, for example, used IoT at its York factory to continuously track production, reducing the full manufacturing cycle from 21 days to just 6 hours. A new motorcycle now rolls off the line every 89 seconds – a pace unimaginable without synchronised data flows and automated processes.

At this stage, manufacturers often try out pilot automation solutions on smaller parts of the plant – perhaps predictive maintenance on a single line or the automation of a production cell with a collaborative robot (cobot). This allows you to test the technology on a small scale, refine it and measure ROI before rolling it out across the factory, significantly reducing risk.

The business benefits of IoT and initial automation are often realised relatively quickly. Companies report a reduction in unplanned downtime of up to 15%, with productivity gains of 10-20%. Sensors that track critical process parameters help reduce defects, while real-time data access enables rapid response and dynamic production adjustments.

IoT and automation work best when they’re woven into the daily rhythm of the factory and tested in real-world conditions. If you want more real-time insight into what’s happening on your factory floor, see how our process monitoring solutions can help. And if you’re planning a pilot or wider automation rollout, find out how we can help with automation projects.

Stage 3: Advanced analytics and AI adoption

After collecting historical data (Stage 1) and real-time data (Stage 2), the next step in digital transformation in manufacturing is to apply advanced data analytics and artificial intelligence.

At this level, it’s no longer just about monitoring the production line in real time. Systems begin to predict future events and automatically optimise production processes.

Key elements of this stage include:

• Predictive analytics (Big Data) – analysing large amounts of data (machine parameters, quality data) to detect hidden patterns. Algorithms trained on historical data can predict equipment failures or shifts in demand.
• AI in maintenance – intelligent predictive maintenance systems examine sensor data to predict failures before they occur. According to McKinsey, this can reduce maintenance costs by up to 30% and downtime by up to 50%.
• Automated decision-making – optimisation algorithms (e.g. machine learning) help plan production, manage inventory or adjust machine parameters in real time, making recommendations based on the latest indicators.
• Computer vision and AI-based quality control – cameras using machine learning detect defects on the line, with details of anomalies feeding into root cause analysis.
• Robotics and autonomous vehicles – AGVs, drones, or palletising robots that communicate with the main system and adapt their tasks to changing production schedules.

At this stage, the factory begins to resemble a nervous system, with information flowing seamlessly and a central ‘brain’ (AI algorithms) constantly improving the performance of the entire organisation. From a business perspective, this means increased productivity, streamlined processes and reduced waste. In addition, implementing digital twins allows you to test changes in a virtual environment, reducing the risk of disruption to live production.

Properly implemented, AI takes care of the routine number crunching, allowing operators to focus on problem solving and process optimisation. According to industry surveys, 61% of manufacturers see AI as key to development by 2029 (up from 41% in 2024), while digitising factories can reduce production costs by 15-30%. In some cases, it can even double production efficiency.

If the earlier stages (integration and IoT) are handled superficially, the system’s predictions may be inaccurate. If the data foundation is solid, AI can deliver huge benefits – from reducing downtime to more accurate production planning.

If your business already has a strong data foundation and you’re looking for practical ways to use AI in manufacturing, take a look at how we can help you implement new digital technologies, including advanced analytics and artificial intelligence.

Six challenges of digital transformation in manufacturing and how to overcome them

Each of the above stages presents exciting opportunities, but also specific challenges. Being aware of these obstacles early on allows you to prepare and greatly increases your chances of success.

Simply replacing people with robots or implementing technology without considering existing processes often leads to disruption and expensive rework. Technology alone isn’t enough – you need to consider workforce needs, real business conditions and long-term management buy-in. Successful companies balance technology and people, build internal capabilities incrementally, and engage top-level leadership.

Cultural resistance and changing ways of working

Some employees may fear that automation or AI will inevitably lead to job cuts. Without their acceptance, even the best-planned transformation can stall.

That’s why it’s wise to launch a change management programme right away: engage the team, provide digital training, and communicate clearly that new technology supports people, not replaces them. Early success with pilots will also help build trust by demonstrating real benefits in practice.

Skills gaps and limited resources

Data scientists or IoT engineers are still relatively new roles in many factories. An incremental approach – building internal skills over time and working with external consultants – can work well, provided the management team allocates the right budget and sets clear priorities.

Organisational silos and bridging IT/OT (Operational Technology)

Transformation often spans production, maintenance, IT and finance departments, each with its own objectives. Studies by Zebra technologies show that a lack of collaboration between IT and operations occurs in one-third of companies, seriously hindering progress. A cross-departmental project team with a shared vision makes decision making much easier and increases the likelihood of success.

Cybersecurity

When machines are connected to the network, the risk of hacking or unauthorised access to control systems increases. Properly securing the OT network (e.g. through segmentation and data encryption) and providing regular staff training on data security are essential. Frameworks such as ISA/IEC 62443 provide best practice in this area.

Uncertainty about ROI and pressure for quick results

Investments in production digitisation can be significant and returns are not always immediate. Iterative steps with clearly defined benefits (a solid business case) at each stage allow management to track progress and base further decisions on solid data.

Integrating older devices and legacy systems

Many machines – such as decades-old PLCs – do not have interfaces compatible with modern standards. Companies often use ‘technology bridges’, such as IoT sensors attached to legacy equipment. Older software systems may require data migration to more modern platforms.

How to measure the effectiveness of digital transformation in manufacturing and manage risk

To assess whether digital initiatives are delivering the desired benefits, define key performance indicators (KPIs) from the outset and systematically track your return on investment (ROI) at each stage of the transformation. Measurable indicators make it easier to assess the profitability of projects, adjust operations and motivate teams to seek continuous improvement. Research shows that digital solutions for manufacturing are essential to stay competitive.

Some example KPIs include:

• OEE (Overall Equipment Effectiveness) – a measure of equipment efficiency. The higher the OEE, the fewer stoppages, the faster the cycles and the less waste. Even basic digitisation makes it easier to identify sources of inefficiency.
• MTBF / MTTR – Mean Time Between Failures and Mean Time To Repair. Predictive maintenance extends MTBF and reduces MTTR.
• Production throughput – for instance, units per hour or total cycle time. Automating processes can often double output.
• Inventory levels and turnover – AI-driven demand forecasting can reduce inventory requirements and improve supply chain management.
• Quality and scrap rates – metrics such as FPY (First Pass Yield) or parts per million defects. Digital analytics can identify variances early and reduce the cost of poor quality by double-digit percentages.
• Order lead time – from receipt of order to shipment of finished goods. Eliminating bottlenecks in the flow of information can significantly speed up delivery (Harley-Davidson is an extreme example, but even a 20% reduction in lead time is a significant benefit).
• Team engagement and productivity – measured through employee satisfaction surveys or attrition data. Digitalisation that gives employees access to information (e.g. clear dashboards) boosts motivation and improves performance – highly engaged teams see 41% lower absenteeism and 59% lower turnover rates.

When calculating ROI, it’s best to focus on specific projects, such as rolling out an Andon system, adopting AI for predictive maintenance, or implementing an IoT platform. Estimate the costs and potential benefits before you start, then review these figures once the solution is live.

For example, an investment of £100,000 in predictive maintenance could save £140,000 a year by avoiding unplanned downtime, giving an annual return of 40%. Similarly, a £60,000 IoT platform that reduces scrap by 5% at scale could quickly pay for itself. There are also ‘soft’ benefits such as increased flexibility, a stronger corporate image or even the opening up of new business models.

Comparing before and after (baseline vs. actual) and monitoring trends supports a process of continuous improvement. If certain metrics don’t improve as hoped, investigate the root causes – perhaps the bottlenecks are procedural or due to a lack of employee engagement. Transformation is an iterative cycle, and data provides the insights needed to adapt.

Publicising success – for example, by displaying key KPIs on a screen in the factory – boosts team motivation. When employees see a reduction in defects or downtime, they’re more open to future change, which helps drive further development and improvement.

A manufacturing digitalisation that works and delivers ROI

Digital transformation in manufacturing is best approached in stages – starting with data integration and basic monitoring, moving on to IoT and initial automation, and finally to advanced AI. Case studies from a range of organisations show that this incremental model minimises risk and delivers measurable returns at each stage. Teams have more time to adapt, and production continues smoothly without major disruptions.

Whether you’re starting with a small pilot or taking your current digital transformation further, an iterative approach reduces risk and delivers tangible results.

If you’d like to explore how manufacturing digitalisation could work for your business, book a free consultation. Together, we’ll review your factory’s unique situation and help identify the most practical solutions so you can reap the benefits of data-driven manufacturing, automation and AI

Explore how manufacturing digitalisation could work for your business,

Book a free consultation