When the second site looked almost there, the numbers told a different story.
The Scenario
You opened your first restaurant in Richmond in 2019, built it through the disruptions that followed, and by 2023 it was producing consistent returns and a strong local reputation. The natural next step seemed clear. You found a unit in Twickenham, negotiated the lease, fitted it out, and opened the second site in January 2024. The concept transferred well enough, the early reviews were encouraging, and the combined revenue figures looked broadly in line with your expectations.
Fourteen months in, the picture is less comfortable. Revenue at both sites is roughly where you expected it to be. The bookkeeping is being done. Monthly accounts arrive, eventually. But the cashflow is tighter than it should be for a business taking combined receipts approaching £50,000 a month, and you cannot quite identify why.
Your accountant mentions, almost in passing, that the way the two sites are being reported makes them difficult to read separately. You had not thought of them separately.
At a Glance
- Client profile: Owner-managed hospitality business, two sites across Richmond and Twickenham, operator with five years’ experience at the original site
- Situation: Second site trading for 14 months with no site-level financial visibility; cashflow tighter than expected across the group
- Core issue: Blended management accounts masking a loss-making second site; two different bookkeeping systems; shared costs not allocated by location
- Cause: No consolidated chart of accounts, shared staff costs attributed to one site, monthly accounts arriving three to four weeks after period end
- Resolution: Cloud accounting unification, site-level P&L with properly allocated costs (all figures net of VAT), weekly cash dashboard, rolling 13-week forecast
- Outcome: Second site identified as loss-making at £2,096 per month; turnaround plan initiated; Twickenham approaching breakeven at month four
Why multi-site visibility is a structural problem, not a bookkeeping one
When a hospitality business opens a second site, the financial structure almost always lags behind the operational one. The first site has its own rhythm, its own supplier relationships, its own settled team. The second site borrows from all of that informally. A kitchen porter covers both. A senior chef splits the week. A shared supplier account means the invoice goes to one location and gets allocated later, or not at all. None of this is deliberate mismanagement. It is simply the practical reality of how growth happens in hospitality.
The result, almost inevitably, is a blended view of the business.
Blended accounts are not entirely without value. They tell you whether the group as a whole is broadly profitable. What they cannot tell you is which site is carrying the other, how margins compare between locations, or whether the second site is on a trajectory that will reach profitability without structural intervention.
The things most commonly hidden in a blended view include:
- Shared labour costs that are not proportionally allocated, making the newer, lower-revenue site look more efficient than it is
- Rent and rates for each site absorbed into a group total, obscuring a higher occupancy cost at a newer or larger-footprint lease
- Food and beverage costs running as a combined percentage, hiding a worse buying position at a site with lower purchasing volumes
- Owner or manager time spread across both sites but rarely costed anywhere
Once those costs are correctly attributed, the picture changes.
The companion articles How Management Reporting Actually Works and Why Management Reporting Often Feels Overwhelming both address the gap between the data owners receive and the decisions they actually need to make. This case illustrates what that gap looks like in a specific operational context.
Why the blended view is particularly risky in hospitality
In most business sectors, a blended view is inconvenient. In hospitality, it can be actively dangerous.
The economics of a restaurant or café are highly sensitive to small movements in margin. For this type of neighbourhood restaurant format, a food cost percentage running at 34 rather than 30 per cent does not sound alarming in isolation, but across a full month of trading it represents a meaningful erosion of the surplus the business needs to cover its fixed costs. Benchmarks vary by concept, service style and menu, so any target should be tested against the specific operation. Similarly, a staffing ratio that looks reasonable at group level can conceal a site where wages are running at 40 per cent of revenue rather than the range that sustainable economics for this format typically require.
There is also a particular feature of hospitality businesses that amplifies this risk. A busy second site generates receipts, pays suppliers, runs the payroll. The bank account moves in recognisable rhythms. It is only when the question shifts from whether cash is arriving to whether anything is actually being made that the underlying picture changes. Cashflow can look reassuring long after profitability has already departed.
“The owner in this case is not unusual. Most multi-site operators we speak to have a rough sense that one site is doing better than the other. What they rarely have is the actual numbers to confirm it, which means they also lack the numbers to decide when patience stops being a strategy. The absence of site-level clarity turns what should be a management decision into a waiting game.”
How the scenario unfolded
The case described here is drawn from composite client experience, not from any single individual. The details reflect patterns we encounter regularly across Richmond-area hospitality businesses.
The owner had run the Richmond restaurant for five years before opening the Twickenham site. The first site was well-established, with a strong repeat customer base, a settled front-of-house team, and a reputation built through consistent neighbourhood service. The second site was a different kind of challenge.
Fourteen months in, the owner knew the Twickenham site was not yet where it needed to be.
What was not known was how far behind it actually was. The bookkeeper was using two different systems, and the combined reporting showed a group that looked, in aggregate, like it was generating a small monthly surplus. A blended P&L arriving three weeks after period end showed combined revenue of around £47,500 per month and an overall position that appeared modestly positive. The cashflow was tight, but this was attributed to the investment phase of a new site and the slower-than-hoped ramp-up in covers.
When the owner came to see our business advisory team, the concern was not crisis but lack of clarity. What was found instead was the visibility that had not been there before.
“The owner came in expecting reassurance that the business was broadly on track and that patience was the right call. We had to have a different conversation. The numbers, once properly separated, told a story that had not been seen before. That conversation was uncomfortable in the short term, and it was also the one that changed the trajectory.”
What the process involved
The first stage was accounting unification. The legacy bookkeeping platform at the original site was retained as a data source but we migrated the consolidated reporting onto cloud accounting software, rebuilding the chart of accounts with a site dimension so that every transaction could be attributed to either Richmond or Twickenham. This took approximately two weeks of data cleaning. All figures in the site-level P&L were prepared net of VAT throughout, to ensure comparability across reporting periods.
The second stage was a weekly cash dashboard.
A simple weekly summary covering seven lines: opening cash balance, card receipts by site, food and beverage cost by site, wages paid, fixed overhead allocations, and a closing net position for each location. Nothing complex. The value was not in the sophistication of the tool but in the regularity of the signal. A problem visible on a Wednesday could be acted on before the following weekend, rather than appearing in an account pack three weeks later.
Alongside the dashboard, we built a rolling 13-week cash forecast covering expected card receipts, supplier payments, payroll, rent, VAT, PAYE and known seasonal pressure points. The forecast was updated weekly so the owner could see not only the current site-level position but also the cash impact of the turnaround plan before it hit the bank account. A 13-week horizon is long enough to anticipate pressure points but short enough to keep the estimates reasonably accurate.
The third stage was cost allocation. Shared staff costs were assigned proportionally based on documented rota records. The senior chef who split the week between sites was allocated 60 per cent to Twickenham and 40 per cent to Richmond, reflecting actual service hours. Two front-of-house workers covering both sites on an ad hoc basis were allocated by the same method. For management reporting purposes, rather than statutory accounts, the owner’s own time was costed at a benchmark management rate and split equally. The bookkeeping entries that had previously blended these costs were restated.
Once all of that was in place, the actual site-level position became visible for the first time.
What the numbers revealed
The site-level P&L, calculated as a monthly average over the preceding three months and presented net of VAT:
Before the allocation work, the blended view had shown Twickenham running at roughly breakeven, close enough that patience felt like a reasonable strategy. After proper allocation, Twickenham was losing £2,096 a month, had been doing so for most of the fourteen months it had been open, and the Richmond site’s profitability had been quietly absorbing the deficit throughout.
The combined group surplus of approximately £2,000 per month was entirely a function of one site subsidising the other. That was not an investment. It was an unplanned transfer.
The three things driving the Twickenham loss
The loss was not traceable to one cause, which is the usual pattern in these situations.
Rent and rates on the Twickenham lease were £900 per month higher than Richmond, reflecting a more prominent position and a lease negotiated in a period of strong commercial landlord confidence in the local market. That fixed cost was immovable in the short term. The food cost percentage was running two points higher than Richmond, partly because of lower purchasing volumes at a site generating £9,000 a month less in revenue, and partly because portioning consistency had not been as tightly managed as at the original site.
The staffing ratio, at 40 per cent of revenue, had been built for a cover count that had not yet materialised. This did not mean the hiring decisions were careless; it meant the rota had been built for the cover count the owner expected, not the cover count the site was actually achieving.
None of these were structural. All three were improvable. But improvement required a plan built around specific numbers, not a general intention to grow into profitability.
What came next
The owner set a six-month turnaround target for Twickenham, with three specific interventions: a menu review targeting a food cost reduction from 32 to 29 per cent, a lunch service trial to increase weekly cover count, and a staffing structure review that removed one front-of-house position and redistributed the hours across the remaining team. Each intervention had a measurable outcome tied to the site-level P&L.
The weekly cash dashboard and rolling 13-week forecast meant those outcomes could be tracked in near real time, rather than being discovered six weeks after the fact. The forward view also allowed the owner to anticipate the cash impact of the turnaround interventions before they appeared in the bank account.
By month four of the turnaround, the Twickenham site had moved from a £2,096 monthly loss to a £340 loss. Not yet profitable, but on a clear and evidenced trajectory. Our financial control team continued to provide the site-level reporting, and the relationship that had begun as a remediation exercise had become a regular advisory one. The article Why 2026 Feels Harder Than It Should for Richmond Businesses explores the wider context in which these operational pressures sit.
Frequently asked questions
What does a site-level P&L include for a restaurant or hospitality business?
A properly allocated site-level P&L separates revenue by location and assigns all directly attributable costs to each site: food and beverage cost, wages, rent and rates, utilities, insurance, and card processing fees. All figures should be prepared net of VAT to ensure comparability. Shared costs, such as centrally employed management or shared supply contracts, are allocated between sites on an agreed basis, typically proportional to revenue, trading hours, or staff headcount. The result is a net profit or loss figure for each location that reflects what each site actually costs to operate.
What is a healthy food cost percentage for a restaurant?
Benchmarks vary by concept, service style and menu, but for most casual dining and neighbourhood restaurant formats the target range is 28 to 32 per cent of food and beverage revenue. A food cost persistently above 34 per cent is usually a signal of portioning inconsistency, supplier pricing drift, waste, or a menu structure that does not adequately reflect ingredient costs. It is worth investigating before attributing it to the type of operation.
What is a typical staffing cost percentage for a restaurant?
Staffing benchmarks vary by format, but for a neighbourhood restaurant of this type the target range we were working with was 32 to 36 per cent of revenue. Sites trading below their projected cover count tend to run higher staffing ratios because the fixed element of the wage bill is spread across fewer covers. A staffing ratio consistently above 38 to 40 per cent is a significant margin risk and usually requires either a cover count improvement or a structural review of the team.
How should shared staff costs be allocated across two restaurant sites?
Shared staff should be allocated based on documented evidence of time spent at each site, typically verified against rota records or timesheets. Where a staff member’s time is not formally recorded by location, a reasonable allocation can be agreed based on operational patterns, but it should be consistently applied and reviewed periodically.
What is a rolling 13-week cash forecast and why does it matter in hospitality?
A rolling 13-week forecast models expected cash receipts and outflows for the next thirteen weeks, updated weekly as actuals come in. It covers card receipts, supplier payments, payroll, rent, VAT, PAYE and known seasonal pressure points. It gives an operator a three-month forward view of cash, which is long enough to anticipate pressure points and short enough to keep the estimates reasonably accurate. In hospitality, where the gap between profit and cashflow can be significant, a forward cash view is a more useful management tool than a historic P&L alone.
Why do blended accounts create problems for multi-site hospitality businesses?
Blended accounts aggregate revenue and costs across all sites, which makes it impossible to identify whether any individual location is profitable or loss-making. A profitable first site can subsidise a loss-making second site for an extended period within a blended view, with the overall group margin declining slowly and no clear signal about where the pressure originates. By the time the overall position deteriorates to a point that feels alarming, the underlying site-level issue is usually many months old.
What does a weekly cash dashboard include?
A weekly cash dashboard for a hospitality business typically covers six to eight lines: opening cash balance, receipts separated by site (net of VAT), payroll paid, supplier payments, fixed cost payments for the period, and a closing balance. Some operators also track a rolling food cost percentage and revenue per cover. The dashboard should be producible in under an hour using data the business already generates, and its value comes from consistent weekly production rather than occasional depth.
At what point should a hospitality operator consider closing a second site?
The decision to close a site should be driven by three questions: whether the loss is structural or operational, whether the trajectory is improving or static, and whether the resource being consumed by the underperforming site could generate better returns elsewhere. A site losing money because of a temporary cover count shortfall and identifiable operational issues is a different situation from one where the fixed costs exceed what can be earned at full capacity. The point at which closure becomes the rational choice is rarely obvious without a genuinely clear site-level P&L.
If this feels familiar
If you operate more than one site and find that your management accounts give you a group picture but not a location-level one, the gap between what you are seeing and what is actually happening may be wider than you realise. A short diagnostic conversation, focused on how your costs are currently allocated, usually gives a much clearer starting position within a few hours.
Our team in Richmond works with owner-managed hospitality and service businesses across the south-west London area. If the numbers feel harder to read than they should, we are happy to help you understand why.
About the author
Donovan Crutchfield, Area Managing Partner, Xeinadin Richmond
LinkedIn: Donovan Crutchfield | LinkedIn
Donovan works with owner-managed businesses and private individuals across Richmond-upon-Thames and the wider South West London catchment. His practice includes a significant number of multi-site operators navigating the management reporting and cashflow challenges that come with growth.
This case study is based on composite client experience and does not represent any single individual. All identifying details have been changed. The information provided is for general guidance only and should not be treated as a substitute for professional financial or business advice specific to your circumstances.
