• Skip to main content
  • Home
  • About
  • Services
  • Software
  • News
  • Contact

Halo Bookkeeping

Halo Bookkeeping

Giving your business wings

  • Home
  • About
  • Services
  • Software
  • News
  • Contact

Every October I have a version of the same phone call. A client has opened the accounts I sent over, seen a profit figure bigger than last year’s, then logged into the bank to pay a supplier and found less than a month’s wages sitting there. The question that follows is always polite and always slightly accusing. “If I made all that, where is it?”

Nobody has made a mistake. The report is right and so is the bank balance. They measure different things, and most small businesses only ever look at one of them.

Autumn is when the gap starts to hurt. The final quarter is busy, Christmas bends every payment pattern out of shape and January arrives with a tax bill attached. A business that is going to run short will usually do it between now and February.

The profit that went missing

A profit and loss report records a sale on the day the invoice goes out. The bank records it on the day the customer pays, which might be six weeks later. Costs work the same way in reverse. So the report can show a healthy month while the account it describes is heading for its overdraft limit.

Growing businesses get caught by this more often than struggling ones. More sales means more unpaid invoices, more stock on the shelf and more wages paid before the money for the work comes in. Growth eats cash first and pays it back later.

Five hiding places

When I go looking for a client’s missing profit, it is nearly always in one of five spots.

  • Unpaid invoices. The work is done and the sale is counted, but the money is still in the customer’s account.
  • Stock and materials, bought and paid for and waiting to be sold.
  • Tax held for HMRC. VAT, PAYE and the tax due on profits all sit in the business account for a while, making the balance look better than it is.
  • Loan repayments and equipment. Paying down a loan or buying a van drains the bank, yet most of it never appears as a cost on the report.
  • Drawings and dividends. What the owner takes out leaves the bank without touching the profit figure.

None of these is a sign of bad management. Every business has them. Trouble starts when nobody is watching how big each one has become.

Thirteen weeks of foresight

The tool I set up for clients more than any other is a 13-week cash forecast. One quarter, laid out week by week, in a plain spreadsheet.

It starts with today’s bank balance. Each week gets the money expected in, entered on the date that customer normally pays (rarely the due date printed on the invoice). Then the known outgoings come off: wages, rent, suppliers, loan repayments, subscriptions. Last come the large, lumpy ones people forget, such as VAT, PAYE, Corporation Tax, the January Self Assessment payment and annual renewals.

The closing balance for each week tells the story. Somewhere in the thirteen there is usually a week where payroll, a VAT payment and a big supplier bill land together. Spotted in October, that week is a scheduling problem: move a payment date, chase two invoices harder, talk to the bank calmly. Spotted three days before, it is an emergency.

Once the forecast exists, updating it takes about fifteen minutes a week.

Chasing without the cringe

Most customers who pay late aren’t being difficult. They pay whoever makes it easiest and whoever reminds them.

The businesses I see getting paid fastest do boring things consistently. They invoice on the day the job finishes instead of saving it all for month end. Their invoices show the due date as an actual date, because “Due 14 November” gets acted on and “30 days” gets filed. Bank details are on every invoice. Terms were agreed in writing before the work began. And one person has a fixed slot every week for following up anything overdue.

A friendly reminder sent every Tuesday collects more money than a furious email sent in month three.

The account that isn’t yours

For a VAT-registered business charging the standard rate, one sixth of every payment received belongs to HMRC. The tax on profits is the same story on a longer fuse. Both sit in the current account looking like spending money.

A second bank account fixes this. Each time a customer pays, a set percentage moves across and stays there. The right percentage varies from one business to the next, and it is something a bookkeeper can help work out. It is the least glamorous habit in finance and the one that most reliably stops a tax bill becoming a crisis.

Count the weeks

One number is worth knowing before Christmas: how many weeks the business could keep paying its essential costs if no money came in. Add up a month of wages, rent, loan repayments and the suppliers you couldn’t trade without, divide by four, then divide the cash available by that weekly figure.

There is no correct answer, because a business with steady retainer income needs less cover than one living on large, irregular projects. What matters is knowing the figure and knowing which direction it is moving.

Old numbers tell old stories

All of this relies on books that are up to date. A forecast built on records six weeks old is guesswork, and nobody can chase an invoice they don’t know is overdue. That is the part of a bookkeeper’s job that gets talked about least. Tidy records are a by-product. The point is figures recent enough to act on.

The clients who make that October phone call nearly always find their profit in their customers’ bank accounts and their own stockroom. It takes an afternoon to trace. Tracing it now, with thirteen weeks in hand, beats tracing it in January.

This article is general information, not financial advice. What is right for your business depends on your circumstances.

If you’d like help building a cash flow forecast, or want to know where your own profit is hiding, book a call today.

Book a call

TV ads make it sound simple: switch on AI, and your business runs itself.

In reality, AI is brilliant at speeding up tasks. But it’s not a finance function. And if you let it operate without a human in the loop, it doesn’t remove risk — it often hides it.

This matters most once you’ve got a team, more transactions, and more moving parts. Because the cost of being “slightly wrong” compounds.

What AI is genuinely good at in finance

AI and automation can reduce admin when the process is already defined. Used properly, it can help with:

  • Capturing receipts and extracting key details
  • Matching bank transactions to invoices and bills
  • Suggesting categories based on past behaviour
  • Flagging anomalies (duplicate payments, unusual spend patterns)
  • Speeding up first drafts (notes, reports, queries)

This is all valuable. I use automation extensively.

But it only works well when the underlying rules are clear.

The failure mode: confident errors at speed

Here’s the problem with “AI does it all”.

If the inputs are messy, AI doesn’t fix the mess. It scales it.

And it tends to be confidently wrong in the exact ways that are hardest to spot quickly.

Common examples I see in growing businesses:

1) Inconsistent coding creates fake signals

If costs are coded differently month-to-month, your profit report becomes noise.

You’ll see margin “improve” when nothing improved.
Or you’ll think overheads are under control while they’re drifting into the wrong categories.

That leads to bad decisions:

  • pricing stays too low
  • hiring happens at the wrong time
  • costs don’t get challenged early

2) VAT becomes a slow-building risk

VAT is a classic area where consistency matters more than cleverness.

Automation can suggest VAT treatment, but it can’t reliably understand context without rules:

  • what you sell
  • how you invoice
  • what evidence you hold
  • what’s standard vs unusual in your business

If VAT is treated slightly differently each month, the risk builds quietly — and you only notice when you file or get asked questions.

3) Month-end turns into clean-up

When automation is bolted onto a messy process, month-end becomes a scramble:

  • missing paperwork
  • “uncertain” transactions parked in holding accounts
  • reconciliations delayed
  • reports produced late

Late reporting doesn’t just mean inconvenience.

It means decisions happen late — and the same issues repeat for another month.

4) Exceptions get missed (and exceptions are where money leaks)

AI is strongest with patterns. Most profit leaks happen in exceptions:

  • a supplier bill that should have been challenged
  • a customer who is unprofitable because they consume time
  • scope creep that isn’t being billed
  • recurring spend that nobody owns

These are judgement calls. They need a human to notice, question, and decide.

The role of a human in the loop

A good human finance function does four things AI can’t do reliably on its own:

1) Defines the rules

What goes where. What counts as direct cost vs overhead. How you want to see margin. How you tag jobs, clients, departments.

Your reporting should reflect how you run the business — not how the software guesses.

2) Spots what changed (and why)

Not “here are the numbers”.

Here’s what moved, what caused it, and what it means.

3) Converts reporting into actions

Finance is only useful when it leads to decisions:

  • tighten scope control
  • renegotiate a supplier
  • fix debtor chasing
  • adjust pricing
  • pause discretionary spend
  • change hiring plans

4) Creates cadence and accountability

Weekly and monthly rhythms protect the business:

  • weekly cash/debtors/approvals
  • fixed month-end close date
  • clear owners for recurring costs and anomalies

That’s how you stop relying on heroics.

The simple approach that works

Here’s the framework I recommend if you want to use AI properly in finance:

  1. Define categories and rules that match your business
  2. Build a weekly rhythm (cash, debtors, approvals)
  3. Close monthly on a fixed date with a short commentary
  4. Then automate the repetitive parts (capture, matching, reminders, drafting)

In that order.

Because automation on top of a weak process doesn’t create control. It creates faster confusion.

If you want this set up properly

If you’re using Xero and you want automation to save time without creating risk, you need someone to design the system and keep it honest.

Want me to look at your current setup and tell you where AI helps — and where you still need a human in the loop?

Book a call

Cost-cutting is easy to do badly.

If you cut in the wrong places, you save money this month and pay for it later through poor delivery, stalled sales, and higher staff turnover. For a growing limited company, capability is often the real asset.

The goal is cutting costs without cutting capability: removing waste while protecting the functions that keep revenue and margin stable.

Step 1: Separate “overheads” from “cost of delivery”

Start by being clear about what you’re reviewing.

  • Cost of delivery: spend directly tied to producing work (materials, subcontractors, delivery tools, direct labour if treated that way).
  • Overheads: spend that supports the business (rent, admin tools, non-billable salaries, general subscriptions, professional services).

Many businesses blur these, which leads to the wrong decisions. Delivery spend should be optimised through pricing and efficiency. Overheads should be reviewed for value and duplication.

Step 2: Create an overhead map with owners

List overhead categories and assign an owner to each. This is where reviews usually fail: nobody “owns” small recurring costs.

A practical overhead map includes:

  • Premises and utilities
  • Software and subscriptions
  • Non-delivery payroll (admin, leadership, sales support)
  • Marketing and sales tools
  • Insurance and finance costs
  • Professional fees
  • Travel and entertainment
  • Training and recruitment
  • Miscellaneous / “Other”

If a line sits in “Other”, recode it. “Other” is where waste hides.

Step 3: Review overheads using three tests

For each cost, apply these tests:

Test A: Does it protect revenue?

Examples:

  • Customer support tools
  • Account management activity
  • Systems that reduce mistakes and rework

If removing it increases churn or creates delivery errors, it’s not an easy cut.

Test B: Does it protect margin?

Examples:

  • Tools that improve utilisation or throughput
  • Systems that reduce manual work
  • Supplier agreements that reduce variable costs

If a cost prevents leakage, cutting it can be a false economy.

Test C: Is it duplicated or poorly used?

This is where you usually find quick savings:

  • Two tools doing the same job
  • Licences assigned to ex-staff
  • Features you pay for but don’t use
  • “Temporary” services that became permanent

Aim for consolidation, not just reduction.

Step 4: Look for the four common overhead leaks

Most SMEs find savings in the same places.

1) Subscription sprawl

Software stacks grow quietly. A few £30–£80 tools become a meaningful monthly cost.

Actions:

  • Export a list of subscriptions from bank statements.
  • Match each to an owner and a business purpose.
  • Remove anything without a clear use case.
  • Consolidate overlapping tools.

2) Premises that no longer fit how you work

Hybrid changes can leave you paying for space you don’t use.

Actions:

  • Measure actual desk use for 2–3 weeks.
  • Renegotiate where possible or sublet part of the space.
  • Review storage and ancillary costs (parking, utilities, cleaning).

3) Non-delivery payroll creep

Support roles can grow faster than delivery output if you don’t plan capacity.

Actions:

  • Measure output per delivery head and support ratio trends.
  • Clarify responsibilities and remove duplicated work.
  • Where possible, improve processes before adding headcount.

4) Professional services on autopilot

Some services are valuable. Some continue because “we’ve always done it”.

Actions:

  • Review the last 12 months of invoices.
  • Ask: what decision did this service help us make?
  • Replace ongoing retainers with scoped work where appropriate.

Quick wins

  • Cancel or consolidate subscriptions with no clear owner.
  • Recode “Other” costs into meaningful buckets within 30 days.
  • Negotiate supplier terms and remove unused licences.
  • Put travel and small expense claims under a simple policy.
  • Run a quarterly overhead review with one-page outputs and actions.

Step 5: Build a “keep list”, not just a “cut list”

Capability is what keeps your business competitive. Document what you’re protecting:

  • Customer experience essentials
  • Delivery quality controls
  • Sales pipeline activity
  • Systems that prevent rework
  • Leadership time for planning and performance

A keep list stops you cutting muscle by accident.

Conclusion

A structured overhead review creates lasting savings without harming growth. You remove waste, and keep the tools and people that protect revenue and margin.

Done well, it also improves focus: fewer tools, clearer roles, cleaner reporting, better decisions.

If you want help applying this to your numbers, book a call.

Book a call

Most dashboards fail for one reason: they try to measure everything.

As you build a team, performance becomes a system. You need a small set of indicators that tell you what’s changing early, and what to do next.

A KPI dashboard for limited company owners should do three things: show profitability, show cash pressure, and show operational drivers.

The rules for a dashboard that gets used

Keep to these rules and you’ll review it every month:

  • 8 metrics max. If it doesn’t change decisions, remove it.
  • Definitions are fixed. Same formula, same routine, every month.
  • Show trend, not just today. Three months minimum, ideally 6–12.
  • Tie each KPI to an owner action. If it drops, who does what?

Now, the metrics.

The 8 metrics that drive profit (and how to use them)

1) Revenue (and revenue split)

Revenue matters, but only with context.

Track:

  • Total revenue month and year-to-date
  • Split by stream (retainers vs projects, service lines, regions)

Action trigger:

  • Revenue growth in the “wrong” stream can hide margin decline. Use the split to see what’s really driving results.

2) Gross margin %

This is the fastest profit signal.

Gross margin = (Revenue – direct costs) / Revenue

Direct costs might include subcontractors, materials, delivery software tied to a job, or direct labour if you track it that way.

Action trigger:

  • A 2–3% margin drop is worth investigating immediately. It usually points to pricing drift, scope creep, or supplier increases.

3) Net profit (or operating profit) %

You need one “after overhead” measure.

Operating profit = Gross profit – overheads

Action trigger:

  • If gross margin is stable but net profit falls, overheads have drifted or productivity has slipped.

4) Labour as a % of revenue

For growing teams, labour is the main cost lever and the main operational risk.

Labour % = Total payroll cost / Revenue

Action trigger:

  • If labour % rises for more than two months, you either need higher pricing, better utilisation, or a staffing plan change.

5) Utilisation or output per head

Choose one that fits your business:

  • Service businesses: billable utilisation (even estimated)
  • Trade/ops businesses: jobs completed per week, or revenue per delivery head
  • Agencies: billable hours delivered vs planned

Action trigger:

  • A small utilisation fall often precedes margin decline. It also signals process issues, rework, or poor scheduling.

6) Average selling price (ASP) or average job value

This keeps pricing honest.

ASP = Revenue / number of sales (or jobs, or retained clients)

Action trigger:

  • If ASP trends down, you’re discounting, selling smaller work, or failing to pass on cost increases.

7) Debtor days 

Profit without collections becomes a cash problem.

Debtor days = (Trade debtors / revenue) x days (use monthly or rolling basis)

Action trigger:

  • Rising debtor days means cash pressure is coming. It usually needs a collections routine, clearer terms, or staged billing.

8) Cash runway (weeks of cash)

Owners need a simple “how safe are we” measure.

Cash runway = cash at bank / average weekly cash outflows (or payroll + key fixed costs)

Action trigger:

  • If runway drops below your comfort level, stop guessing. Use it to prioritise collections, timing of hires, and optional spend.

How to review the dashboard in 20 minutes

A dashboard only works if it leads to actions.

Monthly review agenda:

  1. What moved, and by how much? (no debate)
  2. Why did it move? (one or two drivers, not ten)
  3. What are we doing next month? (3 actions)

Capture actions and owners. Then revisit them mid-month.

Quick wins

  • Remove any KPI that you can’t explain in one sentence.
  • Add a trend line for 6 months, not just the latest month.
  • Set a threshold for each KPI (green/amber/red).
  • Tie each KPI to one operational meeting (sales, delivery, finance).
  • Make debtor days a weekly check if cash is tight.

Conclusion

A small dashboard gives you speed. It highlights issues early, before they show up as a cash squeeze or a surprise in your profit.

Review it monthly and treat it as a management tool, not a reporting exercise.

If you want help applying this to your numbers, book a call.

Book a call

Most owners know their biggest customers by revenue. Far fewer know their best customers by profit.

But that gap matters. Capacity becomes valuable. If you allocate it to the wrong customers, you can grow turnover while profits stay flat.

Client profitability is about understanding what work really costs to deliver, and where the commercial terms no longer make sense.

Why revenue is a poor way to judge clients

Revenue hides three issues:

  • Delivery effort: two £10k clients can consume very different hours and attention.
  • Commercial friction: late changes, slow approvals, and meetings add cost without adding income.
  • Cash timing: a “good” client on 60-day payment terms can create cash pressure even when margins look fine.

If you’re not ranking customers by profit and cash behaviour, you’re managing blind.

What data you need (and what you can do without)

You can do a useful first pass with what you already have:

  • Sales by customer (invoiced revenue)
  • Direct costs tied to delivery (subcontractors, materials, project-specific tools)
  • Staff cost estimate (even if you don’t track time precisely)
  • Credits, write-offs, and discounting
  • Payment history (days to pay)

If you don’t have timesheets, don’t stop. Use a simple proxy: the delivery team’s monthly cost divided by realistic billable hours to get an internal hourly cost. It won’t be perfect. It will be directionally right.

Build a simple “profit per client” view

Create a table for the last 3–6 months. For each client, include:

  1. Revenue (net of discounts and credits)
  2. Direct costs (subcontractors/materials/other job costs)
  3. Estimated labour cost (hours x internal cost, or a % allocation)
  4. Gross profit (revenue minus direct costs and labour)
  5. Gross margin %
  6. Cash behaviour (average days to pay; any disputes; churn risk)

This gives you a ranking. You’re looking for the extremes: your top 10 and bottom 10.

The patterns that usually show up in “worst clients”

Unprofitable clients rarely look unprofitable on day one. They become that way over time.

Common patterns:

  • Scope creep that wasn’t priced or documented
  • Account management drag: lots of meetings, lots of admin, few paid outcomes
  • High rework: unclear briefs, slow approvals, repeated changes
  • Hidden discounting: “keep them happy” credits, free extras, rushed fixes
  • Bad-fit work: you’re doing work you’re not set up to deliver efficiently

Once you see the pattern, the fix becomes clearer.

What to do with the results

This is where owners often overreact. Don’t.

Split clients into four groups:

Group A: High profit, low friction

Protect these relationships. Make sure you’re not underserving them because you’re busy elsewhere.

Actions:

  • Lock in renewals early
  • Raise service levels where it improves retention
  • Ask for referrals when delivery is going well

Group B: High profit, high friction

These clients can be worth keeping if you change how work is delivered.

Actions:

  • Tighten scope and approvals
  • Move to staged billing or deposits on larger work
  • Increase pricing on change requests, not just base fees

Group C: Low profit, low friction

Often the easiest to fix commercially because the relationship is stable.

Actions:

  • Review pricing and package deliverables
  • Remove non-essential extras
  • Adjust the service level to match the fee

Group D: Low profit, high friction

These are the clients that drain your team.

Actions:

  • Reset terms and scope with clear options
  • Increase price materially, or reduce scope materially
  • If neither is acceptable, plan an orderly exit

You don’t need to “fire” clients dramatically. You do need to stop subsidising them.

Quick wins

  • Rank clients by gross profit, not revenue, for the last 3 months.
  • Identify the top 3 causes of delivery drag (meetings, rework, scope creep).
  • Add a change request rule: price it, defer it, or decline it.
  • Move large projects to staged billing (cash timing improves fast).
  • Review the bottom 10 clients monthly until the list stabilises.

Conclusion

Client profitability analysis gives you control over growth. It helps you decide where to focus sales, where to improve delivery, and where to reset terms.

The aim is a healthier portfolio: profitable work, delivered predictably, with cash arriving on time.

If you want help applying this to your numbers, book a call.

Book a call
  • Page 1
  • Page 2
  • Page 3
  • Interim pages omitted …
  • Page 6
  • Go to Next Page »
HB With Wings

07930 106932

support@halo-bookkeeping.co.uk

Halo Bookkeeping & Accounting Ltd
87 Lullington Road, Overseal, Swadlincote
Derbyshire, DE12 6NG

Xero Silver Partner Badge
Scan to Whatsapp? Halo QR Code

Copyright © 2026 · Halo Bookkeeping · Privacy Policy · Cookie Statement

Manage Consent
To provide the best experiences, we use technologies like cookies to store and/or access device information. Consenting to these technologies will allow us to process data such as browsing behaviour or unique IDs on this site. Not consenting or withdrawing consent, may adversely affect certain features and functions.
Functional Always active
The technical storage or access is strictly necessary for the legitimate purpose of enabling the use of a specific service explicitly requested by the subscriber or user, or for the sole purpose of carrying out the transmission of a communication over an electronic communications network.
Preferences
The technical storage or access is necessary for the legitimate purpose of storing preferences that are not requested by the subscriber or user.
Statistics
The technical storage or access that is used exclusively for statistical purposes. The technical storage or access that is used exclusively for anonymous statistical purposes. Without a subpoena, voluntary compliance on the part of your Internet Service Provider, or additional records from a third party, information stored or retrieved for this purpose alone cannot usually be used to identify you.
Marketing
The technical storage or access is required to create user profiles to send advertising, or to track the user on a website or across several websites for similar marketing purposes.
Manage options Manage services Manage {vendor_count} vendors Read more about these purposes
View preferences
{title} {title} {title}