There is a point in every growing business when “we need a tool for that” stops being a solution and starts becoming a problem. The sales team gets a CRM. Finance has its own software. Operations adds a workflow platform. HR brings in another system. Marketing needs a few more. Individually, the decisions make sense. Then someone needs information from three of those systems to finish one task, and the spreadsheet appears. That spreadsheet is often a sign that something is missing. Not another application. A connection.
When people become the connection between systems
Most software purchases are made to solve a specific problem. A finance team needs better financial management. Sales needs a clearer view of opportunities. Operations needs to track work. Leadership needs reporting. The trouble starts when those systems have to work together. Like, a customer order. The order is recorded in one system. The customer details live somewhere else. Finance needs the information for invoicing. Operations needs it for fulfilment. Someone needs to make sure the customer receives what was promised. If those systems are connected, much of that movement can happen as part of the workflow. If they aren’t, people fill the gaps. Someone exports a report. Someone updates a spreadsheet. Someone sends an email. Someone checks whether the previous person has finished their part. None of these tasks looks particularly wasteful on its own. Repeated every day, they become part of the operating model. This is one of the stranger side effects of modern software: companies buy applications to reduce manual work, then create new manual work to keep the applications in sync.
The answer isn’t necessarily fewer systems
It’s easy to turn this into an argument against having multiple tools. That misses the point. Specialist software can be useful. A payroll system does not need to become a CRM, and a CRM does not need to run the finance function. The issue is what happens between those systems. Multiple products can work perfectly well together when information can move between them and workflows don’t stop at the edge of an application. So before buying another tool, it is worth asking a different question:
Where is work getting stuck between the tools we already have?
That question shifts attention from the size of the software stack to the gaps inside it.
Follow the work, not the departments
Businesses are organised into departments. Work rarely is. A new customer might start with marketing, move through sales, involve finance, reach operations and eventually appear in a management report. From the customer’s perspective, that is one relationship. Inside the business, it can become five separate processes. This is where connected workflows make a practical difference. When a deal is closed, for example, the information needed by finance, operations and customer teams should not have to be recreated three times. The commercial details can move with the workflow. The next task can be triggered automatically. The people involved can see the information relevant to their part of the process. The gain isn’t simply fewer clicks. People spend less time explaining what happened and more time deciding what to do next. That distinction matters.
Context is more valuable than a data transfer
There is a difference between moving data and moving context. Suppose a customer contacts a business about an existing order. A basic integration might pass across the customer’s name and order number. A useful connection could make the wider context available too: what was ordered, where it is in the process, what the customer previously asked about and who is handling the issue. That changes the work for the person receiving it. They don’t have to open several systems and reconstruct the story before they can respond. This is one reason disconnected software creates more friction than its users sometimes realise. The problem isn’t always the number of systems. It is the amount of mental work required to piece together information that already exists.
Automation works better when there is somewhere for the work to go
Automation is often discussed as if the goal were simply to eliminate a manual task. But removing one task doesn’t necessarily fix the process around it. Imagine an automated notification that tells an employee a customer has completed a step. The email is automatic, but the employee still has to open another system, find the record, work out what happens next and update something else. One task disappeared. The handoff remained. A connected workflow can take that further. An action in one system can trigger the next step. Relevant information can move with it. A status change can update another part of the process. A person can be brought in only when their judgement is actually needed. The best automation is often the part nobody talks about because nobody has to think about it.
AI makes the connection problem harder to ignore
AI adds another reason to look closely at how systems connect. An AI tool can summarise information, generate content or identify patterns in the data available to it. But useful business decisions rarely depend on one isolated piece of information. A finance decision might depend on sales activity and customer behaviour. An operational decision might depend on demand and inventory. A customer decision might depend on the history of the relationship. If that context is spread across disconnected systems, intelligence has less to work with. This is why adding an AI tool does not automatically make a business more intelligent. The quality of the answer still depends on the quality and relevance of the information behind it.
For businesses, the practical question is therefore bigger than “Where can we use AI?” It is also:
Can the information AI needs actually reach the work where it is useful?
The spreadsheet in the middle tells you where to look
There is a particular kind of spreadsheet that deserves attention. The one everyone calls temporary. It pulls data from the CRM. Someone adds figures from finance. Another person cleans up the formatting. A few formulas turn it into a report. Every Friday, it gets sent around. Eventually, nobody remembers why the spreadsheet was created in the first place. That file is rarely the root problem. It is usually a workaround for a missing connection. The same thing happens with manually maintained trackers, recurring email chains and documents that exist mainly to reconcile information from different systems. These workarounds are useful clues. If people repeatedly build their own bridges between official systems, those bridges are worth examining.
What a connected business actually looks like
A connected business doesn’t necessarily have fewer applications. It has fewer unnecessary handoffs. Information is entered once where possible. Changes can flow to the places that depend on them. Teams don’t have to maintain competing versions of the same record. Reports can draw from connected information instead of being rebuilt manually every week. There is also a practical benefit when the business grows. A new product or capability has somewhere to connect instead of becoming another isolated system that employees have to manage. That makes the technology environment easier to extend without making it harder to operate.
Before buying another tool, look at the gaps
A useful technology review doesn’t have to start with a list of software you might replace. Start with the work. Look for:
- information being entered more than once
- employees manually notifying the next person in a process
- different teams maintaining different versions of the same data
- reports that depend on several exports being combined
- tasks that require constant switching between applications
- spreadsheets that exist mainly to join information from different systems
- changes in one system that someone has to remember to copy elsewhere
These are small operational details, but they show where the technology stack is failing to behave like a connected system. And sometimes the answer will still be another tool. The difference is that you are adding it because it fills a genuine capability gap, rather than because nobody has solved the gap between the tools you already have.
Build around connections, not collections
Most software stacks grow one purchase at a time. A team has a problem, finds a product and moves on to the next problem. After a few years, the business has a collection of capable systems, each solving a different part of the job. The next stage is making those pieces work together. That doesn’t require every system to be replaced. It doesn’t require every process to be automated. And it doesn’t mean forcing the whole business onto one platform. It means being more deliberate about the connections between systems, workflows and the information they produce. Because the value of a tool doesn’t stop at what it can do by itself. It also depends on what happens when it connects to everything around it. And sometimes the next piece of software a business needs is already sitting on its desktop. It just needs to be connected
