Operator Notebook May 2026: NLW Year One, AI Stock Rumour vs Reality, EOFY Prep

The National Living Wage increase has now been live for a full year. AI-powered stock control is everywhere in vendor decks. End-of-financial-year is three months away for most operators. This edition covers all three, without the noise.

NLW Year One: What the Numbers Actually Look Like

The April 2025 National Living Wage increase pushed the adult rate to £12.21 per hour, a 6.7 percent rise on the previous year. For most restaurant and bar operators in the UK, that landed in payroll before most other cost adjustments could be made. Twelve months on, the picture is clearer, and more mixed, than the initial projections suggested.

The operators who absorbed the increase most cleanly were those who had already tightened their food cost and rota controls going into it. The ones still working through it tend to share a common pattern: labour cost went up, food cost stayed roughly where it was on paper, but the margin gap closed faster than expected because input costs also rose independently of wage pressures.

UKHospitality's Q1 2026 operator survey found that 58 percent of independent operators had revised their menu pricing at least once since April 2025, with the average menu price uplift sitting at 4.2 percent across surveyed venues. That covers some of the wage increase, but not all of it, and it does not account for the parallel rise in energy and produce costs over the same period.

The honest takeaway from NLW year one is that the venues managing best are not doing anything structurally different. They are running tighter purchase-to-recipe reconciliation, reducing untracked wastage, and reviewing portion cost against current supplier prices more frequently than they used to. Those are operational habits, not finance department interventions. And they apply whether your wage bill went up by £200 a week or £2,000.

Takeaway: Margin recovery from NLW pressure comes from cost discipline on the controllable side, not from menu price increases alone.

AI Stock Control: What Is Real and What Is a Sales Deck

Every software vendor in the hospitality space has added AI to their marketing in the past 18 months. Some of it is real. A lot of it is a rebrand of features that already existed. Here is a plain-language breakdown of what AI is genuinely doing in stock and inventory management right now, and what you should push back on when you see it in a pitch.

What AI is actually doing well right now: Invoice scanning and data extraction. This is the most mature and useful application. AI can read a supplier PDF invoice, extract line items, match them to your existing product list, flag price variances, and update your cost database automatically. For a kitchen processing 20 to 60 invoices a week, this is meaningful time saving and reduces manual entry errors. This is not experimental. It is in production at scale and works.

Demand forecasting linked to stock ordering. AI models trained on historical sales data, weather, events, and day-of-week patterns can generate suggested order quantities that reduce over-ordering. The accuracy depends entirely on the quality and volume of historical data fed into the model. For a venue with 12 to 18 months of clean POS data, the outputs are useful. For a venue with patchy records or recent menu changes, the suggestions are starting points, not answers.

Waste pattern identification. Systems that log daily waste at ingredient level can surface patterns that a human reviewing a weekly report would miss. Which prep steps generate the most avoidable waste? Which days see unexplained discrepancies between theoretical and actual stock? AI can flag these faster than manual analysis.

What is still largely a sales deck promise: Fully autonomous stock management where the system decides, orders, and reconciles without human review. The underlying logistics, supplier relationship variables, and menu complexity in real hospitality operations make full automation impractical for most venues. The operators who have tried to remove human oversight entirely from ordering have generally added it back within a quarter.

The question to ask any vendor is simple: show me a live operator using this feature and tell me how long they have been using it. The answer tells you whether you are buying a roadmap or a product.

For operators who want to move beyond manual invoice processing and fragmented stock tracking, restaurant stock control software from Stocktake Online includes AI invoice scanning, recipe-linked inventory, and multi-site stock visibility built for venues running on real operational volume.

Takeaway: AI in stock control is most useful where the data pipeline is already clean. Start with the data, then evaluate the AI layer on top of it.

EOFY Prep: Three Things to Do Before July

For UK operators, the end of the financial year falls in April, meaning many are already in their new financial year. For operators in Australia, Singapore, and the UAE, EOFY sits between June and September depending on structure. Regardless of your financial year end, the three months before it are when stock control decisions made all year either show up cleanly or create problems.

Here is what matters most in the run-up:

1. Reconcile your theoretical vs actual usage for the year. If you have been running recipe-linked stock, you should have a theoretical consumption figure for every ingredient across the year. Compare that to what you actually purchased and what your closing stock shows. The gap is your combined waste, unrecorded transfers, and any stock losses. If you have not been tracking this, now is the time to baseline it so the next year has a clean starting point.

2. Run a supplier price audit before renewal season. Most supplier contracts and pricing arrangements are reviewed in Q2 and Q3. If your stock system has been logging invoice prices automatically, you can generate a 12-month price movement report by ingredient category without manual spreadsheet work. This puts you in a stronger position for supplier conversations and helps you identify which categories are driving food cost increases independently of volume changes.

3. Close out your wastage records properly. Wastage logged inconsistently across the year becomes a problem at EOFY when your accountant is reconciling stock values. A closing stock count, matched against your system records, with a reconciliation of any logged variances, makes the audit process significantly cleaner. For multi-site operators, this is even more important because site-level variances need to be attributed correctly before group-level reporting.

The operators who find EOFY straightforward are almost always the ones who have been doing weekly or monthly stock reconciliation throughout the year. The operators who find it painful are catching up on 12 months of unresolved variances in a compressed window. If that sounds familiar, the next six weeks are a reasonable time to reset.

You can also use the free tools for operators available from Stocktake Online, including food cost calculators and stock variance templates, to run a quick pre-EOFY health check on your numbers.

Takeaway: EOFY is not a finance event. It is a data quality event. Clean records throughout the year make it routine. Patchy records make it a reconstruction project.

Before the Next Edition

Three things to consider going into June:

Summer menus change your cost profile. If you are switching to seasonal produce, your theoretical food cost needs to be updated to match current supplier prices, not last summer's. A recipe cost that is 90 days out of date is not a food cost. It is a historical reference.

Ramadan trading data from operators in the UAE and GCC is worth reviewing now if you operate in that market. The demand shifts, purchasing patterns, and staffing profiles from Ramadan 2026 are your baseline for planning 2027. If you did not capture it systematically this year, build the data structure now so you do not repeat the gap.

AI features are landing fast. If you are evaluating new stock or POS systems before EOFY, the AI features in the sales demo are not the right comparison point. Ask about the data model underneath. How does it handle multi-brand recipes? How does it allocate shared stock? How does it handle supplier price changes mid-period? The answers tell you more about operational fit than any dashboard screenshot.