Most SMEs do not need custom software for everything.
Modern SaaS tools are powerful, mature, and often the right choice. They can help businesses move quickly, avoid unnecessary development cost, and access proven functionality without building from scratch.
But there comes a point where adding another subscription does not solve the problem.
It can make the business more fragmented.
Off-the-shelf tools are powerful, but they cannot always support unique workflows, integration gaps, or strategic capability.
That is when custom software starts to make sense.
The problem with “just add another tool”
SaaS tools are easy to adopt.
A team has a problem. A tool promises to solve it. The subscription starts. The business gains a new feature, a new dashboard, a new workflow, or a new database.
At first, this feels efficient.
Over time, the stack grows.
The CRM holds one part of the customer record. Finance holds another. Operations uses a separate system. Reporting depends on exports. Staff use spreadsheets to bridge gaps. Management asks for visibility that no single tool can provide. A process crosses several platforms but belongs to none of them.
The business has more software, but not necessarily more clarity.
At this stage, the problem is rarely solved by adding another isolated application.
The business may need a better operating layer.
SaaS works best when the process is standard
Off-the-shelf tools are strongest when the business process is common and well understood.
Examples include:
- accounting
- payroll
- email marketing
- CRM basics
- scheduling
- ticketing
- document storage
- project management
- ecommerce
- common HR functions
In these areas, buying software is usually sensible. The process is common enough that many businesses need similar functionality.
Custom software is usually unnecessary when a mature SaaS platform already solves the problem well.
The issue begins when the business has a workflow, integration requirement, reporting need, or customer experience that does not fit neatly inside one platform.
That is where workarounds start.
Workarounds are a signal
Custom software may be worth considering when workarounds become part of normal operations.
Common signs include:
- staff entering the same data into multiple systems
- spreadsheets acting as operational databases
- reports built manually from several exports
- approval processes managed through email threads
- customer information scattered across platforms
- teams using notes fields to track structured work
- staff switching between systems to complete one task
- no clear source of truth for key information
- managers relying on people for updates rather than systems
- SaaS tools being used in ways they were not designed for
A workaround is not automatically bad. Every business has them.
But when workarounds become critical to daily operations, they create risk, delay, and dependency on individual knowledge.
At that point, the business should ask whether the software stack is supporting the operating model — or forcing the operating model to bend around the software.
Custom software is not always a full application
Many businesses hear “custom software” and imagine a large, expensive platform built from scratch.
That is not always what is needed.
Custom software can be much more focused.
It might be:
- an internal portal
- a workflow layer
- a reporting interface
- a data synchronisation tool
- middleware between systems
- a document processing workflow
- a tailored dashboard application
- a customer or supplier portal
- an AI-assisted internal tool
- a lightweight operational application
- a connector between existing platforms
The best custom software often does not replace the entire software stack. It connects, extends, and simplifies it.
For SMEs, this can be a more practical approach than trying to force every process into one off-the-shelf product.
When integration is the real issue
Sometimes the problem is not that the existing tools are bad.
The problem is that they do not talk to each other.
The business may already have good software for sales, finance, operations, and customer management. But if those systems are disconnected, staff still have to move information manually.
In this case, custom software or integration can create value by:
- syncing records between platforms
- automating workflow handoffs
- creating a shared reporting layer
- building a single operational interface
- reducing duplicate entry
- connecting data to dashboards
- triggering actions across systems
- giving AI tools access to approved context
This is often more valuable than replacing systems.
The goal is not to build everything. The goal is to make the existing environment work better.
When reporting cannot be solved inside one tool
Many SMEs struggle with reporting because the data needed for leadership visibility sits across multiple systems.
A SaaS dashboard may show what happens inside that platform, but not across the whole business.
For example:
- sales activity lives in the CRM
- revenue lives in finance
- delivery progress lives in operations
- customer issues live in support
- staff utilisation lives elsewhere
- manual adjustments live in spreadsheets
No single SaaS tool can always provide the full picture.
Custom reporting layers, data models, dashboards, or internal applications can help bring that information together.
This is especially valuable when leadership needs to understand performance across functions, not just inside one platform.
Custom software makes sense when the reporting question is broader than the software that currently holds the data.
“The aim is not to own more code. The aim is to create a better business system.”
When the workflow is strategically important
Some workflows are not just administrative.
They are central to how the business competes.
This could include:
- quoting
- onboarding
- service delivery
- client reporting
- scheduling
- job management
- compliance handling
- quality control
- knowledge retrieval
- customer communication
- internal decision support
If a workflow directly affects speed, margin, quality, customer experience, or competitive advantage, it may be too important to be shaped entirely by a generic tool.
Custom software can help when the business needs the process to work in a specific way.
Not because custom is always better, but because strategic workflows deserve systems that fit them properly.
When AI needs business context
AI is another reason custom software may become relevant.
Generic AI tools can be useful, but they often sit outside the business systems where the real context lives.
A business may want AI to:
- answer questions using internal documents
- summarise customer history
- assist with quote generation
- interpret reporting data
- classify incoming requests
- extract information from documents
- support staff inside a workflow
- generate commentary from dashboards
- help retrieve operational knowledge
To do this well, AI often needs access to structured information, approved documents, business rules, workflow context, and system data.
That may require custom software, integration, or an internal AI application.
The value is not in “adding AI.” The value is in embedding AI into the way work actually happens.
When another subscription adds complexity
A useful question for SMEs is:
“Will this new SaaS tool simplify the business, or add another place where work lives?”
Sometimes a new tool is exactly right.
But sometimes it creates:
- another login
- another data silo
- another reporting source
- another workflow boundary
- another integration requirement
- another monthly cost
- another system staff need to maintain
- another place where information can drift
The subscription cost may look small, but the operational cost can be much larger.
Custom software may make more sense when the goal is to reduce complexity rather than add another layer to it.
How to decide: buy, build, or integrate
The decision should not be emotional.
It should be practical.
Buy when:
- the process is standard
- mature tools already solve the problem
- speed is more important than uniqueness
- integration requirements are minimal
- the tool will reduce complexity
Integrate when:
- existing systems are useful but disconnected
- the same information is being copied between platforms
- reporting needs data from multiple systems
- workflows cross tool boundaries
- replacing systems would be unnecessary or disruptive
Build when:
- the workflow is unique or strategically important
- existing tools force too many workarounds
- reporting or operational visibility cannot be achieved otherwise
- the business needs a tailored internal tool
- AI needs to operate within specific business context
- the capability could create meaningful advantage
In many cases, the best answer is a combination.
Use SaaS where it works. Integrate where systems need to connect. Build only where custom capability creates clear value.
What good custom software should achieve
Custom software should not exist for its own sake.
It should make the business easier to operate.
Good custom software should:
- reduce manual work
- improve visibility
- connect systems
- support better decisions
- fit the real workflow
- reduce reliance on spreadsheets
- improve data quality
- create consistency
- support scale
- make AI more useful where appropriate
If it does not improve the operating model, it may not be worth building.
The aim is not to own more code. The aim is to create a better business system.
Final thought
SaaS is often the right answer.
Until it is not.
When a business has unique workflows, fragmented systems, manual workarounds, poor reporting visibility, or strategic processes that generic tools cannot support properly, custom software can become the more practical choice.
Not as a replacement for every platform.
As a tailored layer that connects, extends, and strengthens the business.
The best question is not:
“Can we buy another tool?”
It is:
“What capability does the business actually need?”
When that answer is clear, the choice between buying, integrating, or building becomes much easier.