Most business owners know the headline: employer National Insurance contributions rose from 13.8% to 15% in April 2025, and the threshold at which you start paying dropped from £9,100 to £5,000 per employee. The numbers have been in the press, in accountants' letters, and in budget planning conversations for months.
What gets less attention is the second-order effect - what that increase does to every hour of admin time in your business that could, in principle, be automated.
This is not a small number. For many businesses, it is bigger than the headline NI cost itself.
The maths most businesses are not doing
Take a straightforward example. You have an admin employee on a £28,000 salary. Under the old regime, you were paying employer NI on £18,900 of that (everything above the £9,100 threshold). At 13.8%, that is £2,608 per year in employer NI.
Under the new regime, you are paying on £23,000 (everything above the new £5,000 threshold). At 15%, that is £3,450 per year.
The difference is £842 per year for that one employee. Not trivial, but manageable.
Now here is the part that does not make it into the budget conversations.
If your admin employee works 37.5 hours a week, their total employment cost per hour - including the new NI rate - is roughly £16.50. If they spend 15 of those hours each week on work that a well-designed system could handle - data entry, scheduling, chasing invoices, generating reports, sending standard correspondence - that is £247.50 per week. £12,870 per year. For one person doing one role.
Multiply that across two or three admin-heavy roles and you are looking at a number that makes the headline NI increase look like a rounding error.
Why this matters now specifically
The NI change did not create inefficiency in your operation. It just made existing inefficiency more expensive. The manual processes that were costing you £12,000 a year before April 2025 are now costing you more - and the gap between what you are paying for that work and what it would cost to automate it has widened.
That gap is where the decision lives. Not "should we automate eventually" but "at what point does the cost of not automating exceed the cost of doing it properly."
For a lot of businesses, that point arrived in April 2025.
How to calculate your own number
You do not need a consultant to do this first pass. You need honest answers to three questions:
1. How many hours per week does each admin-heavy role spend on genuinely repetitive work? Not approximate - actually track it for one week. Scheduling, data entry, copy-paste work, standard emails, report generation, invoice chasing. Write it down.
2. What is the total employment cost per hour for those roles? Take annual salary, add employer NI at the new rate, add pension contributions (typically 3% minimum), divide by 1,950 working hours per year. That is your real cost per hour.
3. What percentage of those repetitive hours genuinely require a human decision? This is the number most people get wrong. They assume the answer is "most of it." In practice, for the majority of admin tasks, the honest answer is close to zero. If the task is: given this input, produce this output - it does not require a human. It requires a system.
That £11,427 is not the saving from automating. It is the ceiling - the maximum you could recover if you eliminated all the repetitive work. In practice, automation does not eliminate roles, it frees them for higher-value work. But it does tell you how much runway you have for an investment in a system that handles it.
A well-built automation that costs £8,000 to implement and eliminates £11,000 of annual manual work pays for itself in under nine months. After that, every year it runs is pure recovery.
The mistake businesses make at this point
The mistake is reaching for a tool. You see the number, you start looking at software subscriptions, automation platforms, AI tools. Most of them promise to fix everything. Most of them add complexity rather than removing it, because they are built for generic workflows, not for the specific way your business actually operates.
The problem is never the tool. The problem is that nobody mapped the work before buying the tool. What actually happens at each step? What decisions get made? What context does the person doing the task need that has never been written down anywhere?
Three hours of honest process mapping will show you more about where your money is going than any software demo. It will also show you which tasks are genuinely automatable and which ones only look automatable until you try.
What to do with this
If you have run the calculation above and the number is uncomfortable, the next step is not to buy anything. It is to understand the landscape properly before committing to a solution.
That is exactly what a Process Audit does. A focused diagnostic of your operation - one conversation, one honest look at how work actually flows through your business - that identifies your three highest-cost manual processes, calculates what each one is costing you annually at current employment rates, and tells you what an automated alternative would look like.
It costs £500. It takes under an hour of your time. And it gives you a written report with real numbers - not estimates, not ranges, actual figures based on your operation - within five working days.
If the numbers do not make a case for change, the report will say so. That is worth knowing too.