The question put to the accountant is almost always the same: “how do I pay less?”. The honest answer disappoints: the levers that actually work are rigorous bookkeeping, deducting what is genuinely deductible, and choosing the right regime — and they return more than any structure. What is presented as a trick is usually a risk whose price is paid later.

The essentials in five points
- The first lever is bookkeeping. An unsupported cost is a lost cost, and that is the most frequent loss of all.
- The second is the choice of regime, decided upfront and rarely revisited along the way.
- The third is timing: the date of an investment or a cost changes which year bears it.
- The line is the reality of the transaction: anything whose only purpose is tax is visible and gets argued.
- A tax saving that depends on not being inspected is not a saving, it is a debt with an unknown due date.
1. The levers that genuinely pay
There are few of them, they are well known, and all of them are boring. That is precisely why they are under-used in small businesses, where time is scarcer than ideas.
| Lever | What it returns | Where it happens |
|---|---|---|
| Supporting every cost | What gets deducted instead of lost | Daily, as documents arrive |
| Deducting recoverable VAT | The tax you advance and recover | As each purchase is entered |
| Choosing the right regime | A base and obligations that fit | At formation, then at thresholds |
| Placing a cost in the right year | A legitimate cash-flow shift | Before the close, not after |
| Provisioning what is justified | A cost recognised when the risk arises | At the close, on evidence |
| Keeping to deadlines | The absence of avoidable surcharges | All year round |
The first line is by far the most profitable and the least spectacular. In a small business, what costs most is not the tax calculated on a correct base; it is the real costs that cannot be deducted for want of a compliant document.
2. Where the line runs
The distinction between planning and abuse is not a question of amount but of reality. Four questions are enough to place a transaction.
- Does the transaction have an economic purpose other than the tax advantage?
- Would it have been done, in something close to this form, with no tax in mind?
- Do the documents describe what actually happened?
- Are the prices between related parties those that unrelated parties would use?
- Would you be comfortable explaining it out loud, without rephrasing?
If the answer is no to any of these, the subject is no longer planning. The last is the most useful in practice: a transaction you cannot describe simply is almost always one you will struggle to defend.
3. The false good ideas
Certain practices circulate as obvious truths in conversations between traders. What they share is that they move the problem rather than solve it.
- Putting personal spending through as costs: The gain is immediate and the risk permanent. The cost is added back, and repetition turns an error into a method, which changes the nature of the discussion.
- Invoicing with no substance between related companies: A service that does not exist shows up in the absence of a deliverable, time spent and resources. The structure comes apart at its weakest point.
- Paying cash to stay discreet: Above the thresholds the cost stops being deductible: the price paid for discretion is exactly the advantage sought.
- Delaying invoicing to shift the tax: Attachment follows the chargeable event, not the issue date. Moving the invoice does not move the revenue.
- Not filing for a period with no activity: The gain is nil and the blockage real: a missing nil return blocks a clearance certificate.
What these five practices share is that all of them require, in order to work, that nobody looks. A strategy that depends on a third party’s inattention is not a tax strategy, it is a bet.
A saving that depends on not being inspected is not a saving
This is the simplest test and the least often applied. If the advantage disappears the moment someone examines the file, it was never earned: it was only borrowed, and the due date is unknown. A small business does not have the reserves to absorb an assessment covering several years. Real planning is the kind that stays true when you describe it.
4. Timing, the only genuinely reversible lever
Deciding when to incur a cost or an investment is legitimate, documented and free of risk, provided the decision is real and so are the dates.
- Estimate the result before the close, with interim accounts.
- Identify costs already decided on whose date remains open.
- Check the real effect of an investment, which is depreciated rather than deducted at once.
- Make sure the date used corresponds to a real delivery or performance.
- Have the judgement validated by your accountant before committing, not after.
The third line corrects the most widespread illusion: buying equipment in December does not reduce the result by the amount of the purchase. The asset is depreciated over its useful life, and the effect on the year is limited to the charge, calculated pro rata.
Mistakes to avoid
- Looking for a structure before securing the deduction of real costs.
- Believing a year-end purchase is deducted in full in that year.
- Paying cash above the thresholds to stay discreet.
- Delaying invoicing in the belief it shifts the revenue.
- Skipping a nil return because no amount is due.
- Making a timing decision after the close rather than before.
Frequently asked questions
What is the most profitable lever for a small business?
Rigorously supporting costs. In most small businesses, the amount lost each year for want of a compliant document far exceeds what any more sophisticated judgement would return.
Does buying equipment before the close reduce tax?
Far less than people think. A fixed asset is not deducted at once: it is depreciated over its useful life, and only the year’s charge, calculated pro rata, weighs on the result.
Where is the line between planning and abuse?
At the reality of the transaction. A decision with an economic purpose of its own, whose documents describe what actually happened, remains defensible. One whose only purpose is tax does not.
Can invoicing be delayed to change the year?
No. Attachment follows the chargeable event — delivery or performance — not the invoice issue date. Moving the document does not move the revenue; it only creates an invoice to be issued.
Does BelloPOS help with tax planning?
Indirectly, and through the most profitable lever: split sales and documented spending. BelloPOS records sales from Lite, purchases from Go, and accounting journals and exports with Pro. The tax judgements are made with your accountant.
What to take away
Planning for a small business comes down to three unmysterious moves: support what you deduct, choose a regime that fits, and decide the timing of spending before the close. Anything that requires not being looked at is not on that list, and generally costs more than the tax it claimed to avoid.
Sources
The figures and rules quoted above come from these pages, read on the date given in the article.
- Moroccan Tax Administration, 2026 General Tax Code
- Ministry of Economy and Finance, Taxes in brief, read 1 September 2026
The most profitable lever, kept up daily
BelloPOS records sales from the free Lite licence onward, purchases and their documents from Go, and accounting journals and exports with Pro.
Read next
Other practical guides on the same subject: