By Darren Enfield, Softkom Solutions
There is no shortage of technology available to businesses today.
There are CRM systems, accounting platforms, project management tools, communication
platforms, automation tools and now an ever-growing number of AI products promising to
improve productivity.
Yet many growing businesses I come across have a surprisingly similar problem.
They have plenty of software, but their businesses are still being run through spreadsheets,
WhatsApp messages, emails and manual processes.
A customer enquiry might arrive through WhatsApp or email. Someone captures the details
into a spreadsheet. The same information may later need to be entered into an accounting
system. Another spreadsheet is used to track progress. Management reports are manually
put together at the end of the week or month.
Each individual system may work perfectly well.
The problem is that the overall process does not.
The problem is often not a lack of technology
When a business starts growing, new systems are normally introduced as new requirements
arise.
Accounting needs a system. Sales needs a CRM. Operations creates spreadsheets. Customer
service uses WhatsApp. Management wants reporting. Employees find other tools that
make their particular jobs easier.
This happens gradually and, for a while, it works.
The difficulty comes when the business grows but the underlying processes do not grow
with it.
Information starts being captured more than once. Employees spend time moving data
between systems. Reports depend on several spreadsheets being updated correctly.
Customer follow-ups are missed because the information sits in somebody’s inbox or
WhatsApp history.
Eventually, the business reaches a point where adding another piece of software may
actually make the problem worse.What it needs is not necessarily another system.
It needs its existing processes and systems to work together.
Start with the business problem, not the technology
AI has made this particularly important.
Businesses are being encouraged to adopt AI quickly, and there are certainly areas where AI
can create enormous value.
But adding AI to an inefficient process does not automatically make it efficient.
Before recommending a particular technology, I believe there are three questions worth
asking:
Can we automate it?
Can we integrate it?
Or do we need to build it?
The distinction is important.
1. Automate it
Many businesses still have people performing repetitive tasks that follow essentially the
same process every day.
A sales enquiry arrives and somebody captures it.
A customer completes a form and somebody sends an acknowledgement.
An invoice is received and somebody forwards it to the correct person.
A manager needs a weekly report and somebody spends several hours collecting
information from different places.
These are often good candidates for automation.
Automation does not have to mean replacing employees. In many cases, its real value is
removing low-value administration so that employees can spend more time dealing with
customers, solving problems and doing work that requires judgement.
If a task is repetitive, predictable and follows clear rules, the first question should be
whether a system can handle some or all of it automatically.
2. Integrate it
This is an area I believe many businesses overlook.
Sometimes the company already owns all the software it needs.
The problem is that the systems do not communicate with one another.
For example, a company may have a perfectly good CRM and a perfectly good accounting
platform, but employees are manually transferring information between them.
Or customer information may exist in several systems, with each department maintaining its
own version.
In these situations, replacing the software may be unnecessary.
Connecting the systems can often solve the bigger problem.
APIs, workflow automation and system integrations can allow information captured in one
place to move automatically to another, trigger actions and update records without
somebody having to re-enter the information.
The objective should be to create a connected flow of information through the business.
3. Build it
There is also a point where automation and integration are not enough.
One of the clearest signs is when a collection of spreadsheets has effectively become the
company’s operational system.
There may be a spreadsheet for customers, another for orders, another for stock, another
for jobs and another for management reporting.
These spreadsheets often contain years of business knowledge. They have been modified
repeatedly as the company has grown.
At that point, the business may have created its own software system without actually
having software.
That is where a purpose-built business system can start making sense.
The aim should not be to develop custom software simply because it sounds more
sophisticated. Custom development requires investment and ongoing responsibility.
It makes sense when the process is important enough, the inefficiency is expensive enough
and standard software cannot adequately support the way the business operates.
AI should be part of the solution, not the starting pointAI adds another layer to this discussion.
There are excellent opportunities to use AI for document processing, customer
communication, knowledge retrieval, data analysis, lead qualification and many other
business processes.
But AI works best when it has access to the right information and is operating within a
properly defined process.
If customer information is spread across spreadsheets, emails, WhatsApp conversations and
disconnected systems, introducing AI without first addressing the underlying information
flow can simply add another layer of complexity.
In many cases, the most valuable AI project therefore starts with work that has very little to
do with AI.
It starts by understanding how information moves through the business.
Look for the warning signs
Business owners do not need to conduct a major technology audit to recognise that
something is wrong.
There are usually fairly obvious warning signs.
Employees are capturing the same information in several places.
Management cannot get accurate information without asking somebody to prepare a
spreadsheet.
Important processes depend heavily on one employee knowing how everything works.
Customer enquiries or follow-ups occasionally fall through the cracks.
Teams spend significant amounts of time copying information between systems.
The company has accumulated multiple software subscriptions, but employees still rely
heavily on spreadsheets and WhatsApp to get the work done.
These are not necessarily technology problems.
They are operational problems that technology may be able to solve.
The objective should be simplicity
For growing African SMEs, technology investment needs to produce practical results.
The question should not be, “What new software should we buy?”It should be, “What is slowing this business down?”
Once that has been established, the technology decision becomes much easier.
If the process works but involves repetitive administration, automate it.
If the systems work but do not communicate, integrate them.
If the business has outgrown the available tools and its spreadsheets have effectively
become its operational platform, consider building something purpose-built.
And sometimes the right decision is to leave the existing system alone.
Good technology strategy is not about introducing as much technology as possible.
It is about using the right amount of technology to make the business simpler to operate,
easier to manage and capable of growing without adding unnecessary complexity.
For many SMEs, the next stage of digital transformation may therefore have less to do with
buying more software and much more to do with making the technology they already have
work together.
