Modernise or Improve? A Decision Guide for Operations Leaders

When operations underperform, there are two ways to fix them. You can modernise the technology, or you can improve the process. ITIL guidance names both: information and technology modernisation, and value stream and process improvement. Choosing between process improvement vs technology modernisation is one of the most important early decisions in any improvement plan.

Get it wrong and the cost is high. A new platform will not fix a bad process. It just runs the bad process on more expensive infrastructure. And no amount of process redesign will make an out-of-support database fast, secure or reliable.

This guide helps you find the real constraint, choose the right route and avoid the most common mistake: picking the answer first and looking for the problem afterwards.

The short answer

The rest of this guide explains how to tell which situation you are in.

Why the order matters

Most organisations default to the route they know. IT teams reach for new technology. Operations teams reach for process maps. Consultants reach for whatever they sell.

The better question is: what is actually stopping the work from flowing? Look for the constraint, the single point that limits how fast and how well work gets done. Fix that first. Improving anything else just moves the queue.

Two common failures show why this matters:

• The expensive lift-and-shift. An organisation migrates to a new platform but moves every old workaround with it. The spreadsheets, manual checks and duplicate approvals all survive. Twelve months later, the platform is modern but the work is no faster. This is one reason data migrations fail after go-live.
• The endless process project. A team maps and redesigns a process carefully, then finds the underlying system cannot support the new design. The improvement stalls while everyone waits for a system change that was never planned.

Five signs the technology is the constraint

1. Performance problems are getting worse. Reports time out, overnight jobs overrun and users wait for screens to load. See common SQL Server performance issues.
2. You are running unsupported software. Database or operating system versions are out of Microsoft support, or will be soon. See our SQL Server 2016 end of life guide.
3. Data does not agree between systems. Different teams quote different numbers for the same thing, and nobody is sure which is right.
4. Integration is fragile or impossible. Systems have no API, or integrations break whenever a supplier updates their product.
5. Change is risky. Nobody wants to touch the system because nobody fully understands what depends on what.

If three or more apply, start with information and technology modernisation.

Five signs the process is the constraint

1. Work waits far longer than it takes to do. A task that takes ten minutes sits in a queue for three days.
2. Data is entered more than once. Someone copies information from an email into one system, then from that system into another.
3. Approvals depend on chasing. Requests stall in inboxes, and progress depends on someone sending reminders.
4. People have built workarounds. Teams keep their own spreadsheets because the official process does not meet their needs.
5. Knowledge sits with one person. The process works because one experienced person knows how, and it breaks when they are on holiday.

If three or more apply, start with value stream and process improvement.

A quick test: follow one piece of work

The fastest way to find the constraint is to follow a single real item from start to finish. Pick something common, such as a customer order, a new starter, a supplier invoice or a student enquiry.

For each step, write down:

• Who does it
• Which system they use
• How long the work takes
• How long it waits before the next step

Then look at the totals.

• If most of the time is waiting between steps, the constraint is the process. Handoffs, approvals and queues are where the time goes.
• If most of the time is spent inside a system, for example waiting for reports to run, re-entering data a system cannot import or fixing errors caused by bad data, the constraint is the technology.
• If it is a mix, you need both. Read on for the right sequence.

This exercise usually takes a few hours, not weeks. It is also the first step in value stream mapping.

When you need both

Many organisations find problems in both areas. That is normal. The question is sequence, not choice.

The usual order is process first, then technology.

1. Map and simplify the process. Remove steps that add no value, so you know what the new platform actually needs to support.
2. Automate quick wins on current systems. Many approvals, notifications and reports can be automated with Power Automate before any platform change, delivering value while the bigger work is planned.
3. Modernise the platform around the improved process. Migrate only what the new process needs, with full reconciliation so nothing is lost. See how to de-risk a data migration programme.
4. Automate more deeply. With reliable, connected data, you can automate further, including integrations that were impossible on the old platform.

The exception: fix urgent technology risk first. If a system is out of support, failing or exposed to security risk, stabilise it before redesigning anything. A process improvement project is no use if the database it depends on goes down.

Where automation fits

Automation is not a third route. It is the tool that makes both routes pay off.

• After process improvement, automation removes the repetitive steps that remain.
• After modernisation, automation becomes easier, because data is consolidated and systems have proper connectors.

Microsoft gives you several ways to automate: cloud flows, desktop RPA and AI agents. Our guide to choosing between cloud flows, desktop RPA and AI agents explains which to use when. For practical starting points, see 20 Power Automate workflow examples. And to understand why automating reporting is often the best first step, read the compound return on report automation.

Two examples

A higher education finance team. Month-end reporting takes eight working days. Following the work shows most of that time is spent waiting for figures from departments and re-keying them into a master spreadsheet. The finance system itself is fine. Route: process improvement, with a standard submission form, automated collection and a Power BI report on the existing data.

A transport and logistics operator. Job, fleet and finance data live in separate systems that do not talk to each other, and the main operational database is on an unsupported version. Staff spend hours each day reconciling them. Route: modernise first to stabilise and connect the data, then improve and automate the job-to-invoice process on the new foundation.

These are illustrative examples, not client case studies. See how we work in education and transport and logistics.

Questions to ask before you decide

• What is the single biggest delay in this process, and where does it happen?
• Are any of our systems out of support, or close to it?
• Do we trust the data we report on?
• If we migrated tomorrow, which workarounds would we carry across?
• What could we automate now, on the systems we already have?
• Who owns this process from end to end?

If you cannot answer the last question, start there. Processes without a single owner rarely improve, whatever the technology.

How Cyber Samurai helps

We help organisations find the real constraint and fix it, whether that means modernising data platforms on SQL Server, Azure and Microsoft Fabric, improving and automating processes, or both.

Our assessments are a good place to start:

• Automation Opportunity Assessment: finds where process improvement and automation will save the most time
• SQL Server Data Platform Audit: assesses platform health, performance and upgrade risk
• Data Migration Readiness: plans a modernisation without data loss

Not sure where you stand? Score your data health, or read more about our operational excellence approach.

Book a consultation