Prepared by Maitland Agency

Board report · Maitland Agency and eCatering

eCatering and Maitland: the first nine months

1 November 2025 to 31 July 2026, measured against the same nine months a year earlier

All figures from Shopify order data, Google Analytics 4, Google Search Console, Google Ads. Revenue basis: Shopify, paid orders, Online Store, including VAT. Figures as at 19 August 2026 and subject to late refunds.

Section 1

The headline

Revenue

£4,839,528

£5,480,024

+£640,496+13.2%

Orders

10,824

12,430

+1,606+14.8%

Average order value

£447.11

£440.87

-£6.24-1.4%

Visitors to the website

807,284

958,847

+151,563+18.8%

UK visitors, measured the same way on both sides. This is the conservative figure: from October 2025 the shop began filtering out automated traffic, so this year's count excludes something last year's included.

The cleanest comparison we have: 1 to 18 August, both years

Revenue

£289,049

£347,899

+20.4%

Orders

620

828

+33.6%

The same eighteen calendar days in each year, measured the same way on both sides.

Revenue is up 13.2% and orders up 14.8%. Average order value is down 1.4%, so the growth is coming from more orders rather than bigger ones. This report sets out where that came from, what we are fixing next, and what the year ahead looks like.

Where the momentum went

Actual trading (2024 and 2025)
2026, seven months actual plus an estimate
2027 on current course, £7,905,789
with the three decisions, £9,453,789

This is the question worth answering first.

2024 took £8.22m. 2025 took £6.59m, a fall of 19.8%, and the wider market did not fall that year. That drop happened before this engagement began in November 2025.

2026 is recovering, and we estimate it at £7.3m rather than the £7.48m the headline rate would give, because we would rather beat the number than explain it. 2027 forecasts a return to within 4% of the 2024 level. The momentum did not go anywhere this year. It went in 2025, and this is what getting it back looks like.

Section 2

13.2% against what?

Growth on its own means nothing without knowing what the market did.

eCatering online+13.2%

Shopify order records, 1 Nov to 31 Jul, both years

UK online retail overall+11.5%

Office for National Statistics, Retail Sales Index, internet sales series KQ7I, updated 21 August 2026, matched to the same months

Bunzl UK and Ireland+1.4%

Bunzl plc full year 2025 results. Bunzl acquired Nisbets in May 2024.

UK hospitality like-for-like sales+0.2%

NIQ and RSM Hospitality Business Tracker, June 2026, 121 managed groups

eCatering grew faster than UK online retail as a whole.

It grew nine times faster than the group that owns its largest competitor.

And it did that while its own customers were shrinking: 4,865 licensed premises closed in 2025, and 305 more sites were lost in the first quarter of 2026 alone.

Where these figures come from

UK online retail growth — Office for National Statistics, Retail Sales Index, internet sales value series KQ7I, published 21 August 2026. Rebuilt to match 1 November to 31 July so the months are directly comparable. The Office's own three-month figure to July is +11.0%.

Competitor growth — Bunzl plc full year 2025 results. Bunzl acquired Nisbets in May 2024, so Bunzl's UK and Ireland division is the closest published proxy for Nisbets' direction of travel.

Hospitality trading — NIQ and RSM Hospitality Business Tracker, June 2026, based on 121 managed groups.

Hospitality closures — Hospitality Market Monitor, Q1 and Q2 2026, and the NIQ 2025 health check. 4,865 licensed premises closed during 2025. 305 net sites were lost in the first quarter of 2026.

Where published sources disagreed we have used the one matched to our exact months and said so. The British Retail Consortium's July figure for online non-food is +1.3%, which would make eCatering look considerably better. We have not used it, because it does not cover the same period or the same categories.

The honest version

The business fell 19.8% in 2025. Growing 13.2% on top of that leaves it about 9% below where it was in 2024. Just over half the ground has been recovered.

That matters because the 2025 fall was not the market. The wider market did not fall that year, so the drop was specific to this business, and this engagement began in November of that year.

Section 3

Up on last year. Behind a target set twelve months ago.

Two different things are being measured, and they are being confused with each other.

Against the same nine months last year, online sales are up 13.2% and orders are up 14.8%. That is a comparison against what the business actually did.

Against last year+13.2%

Against the target set twelve months agobehind

Two different measurements. One is against what the business did. The other is against what somebody hoped it would do.

Against the target set twelve months ago, the business is behind. That is a comparison against what somebody hoped it would do, decided before the year began and before most of this work started.

Both statements are true at the same time. Being behind a forecast is not the same as going backwards, and the two need to be argued separately. This report answers the first question, because it is the one the data can answer. The second is a question about how the target was built.

The honest summary in one line: the business is trading better than it was and it is not yet where anyone wants it to be.

Section 4

Revenue is one question. Profit is the other.

Revenue can rise while the money the business keeps on each sale falls. It is the right question to ask of any growth figure. The shop has recorded the cost of what it sells since February 2025, so we can answer it from February 2025 onward. Here are the same six months in each year, side by side.

Sales, cost of goods and gross profit, February to July, both years
Measure	February to July 2025	February to July 2026	Change
Sales	£2,627,311	£3,080,405	+17.2%
Cost of the goods sold	£1,525,883	£1,762,189	+15.5%
Gross profit	£1,101,428	£1,318,216	+19.7%
Margin kept on each £1 of sales	41.9%	42.8%	+0.9%

These are the products where the shop records a cost, which is 95.5% of sales.

Cost of goods has only been recorded on this shop since February 2025. Anything before that date cannot be assessed for profit, which is why this comparison starts in February.

Gross profit rose faster than sales. The business sold 17.2% more and kept 19.7% more profit, because the margin on each pound improved slightly, from 41.9p to 42.8p.

A large part of that improvement is June, and a large part of June is air conditioning. Gross profit in June alone rose £164,521, which is roughly three quarters of the £216,789 gained across the whole six months. Air conditioning carries a margin of about 76%, well above the rest of the range, so a strong air conditioning month lifts the average.

That is not a caveat, it is the point. The category sold because the marketing put it in front of people the moment demand appeared, and it sold at the best margin in the business. What it does mean is that the underlying margin on everything else was broadly flat rather than rising, which is useful for planning next year's range together.

Discounting is not eating into this. Discounts fell from 1.52% to 1.25% of sales and returns fell from 0.57% to 0.30%. The business is not buying its growth with price cuts.

Gross profit, month by month

2025
2026

Cost of goods has only been recorded since February 2025, so the comparison starts there.

Section 5

Monthly revenue and orders, November 2024 to August 2026

Seven of the nine months are up. Two are down.

Nov

+£75,708

+12.4%

Dec

+£47,874

+12.3%

Jan

-£88,228

-16.9%

Feb

+£97,358

+25.2%

Mar

+£140,659

+28.7%

Apr

+£139,418

+25.0%

May

+£40,650

+6.7%

Jun

+£241,490

+38.1%

Jul

-£54,433

-8.4%

Before Maitland (Nov 2024 to Jul 2025)
With Maitland (Nov 2025 to Jul 2026)
Estimated air conditioning sales the business could not fulfil, added on top

What this means: each bar is one month of sales. In the default view the grey bars are last year and the green bars are the same months with Maitland. The full timeline view shows every month from November 2024 onwards.

January and July are the two months below last year. January is explained in "What we are fixing next". July 2026 restated downwards by £7,422 after late refunds, which is normal for this business and will move again. The lighter block on July is our estimate of the air conditioning sales the business could not fulfil, built from measured demand. With stock on the shelf, July 2026 would have finished at about £658,000 and been ahead of last year.

January. Four things landed in the same month, and eCatering's own management set them out at the time: the January sale did not go live until part way through the month, stock had run short after a very strong November and December, paid search had been deferred from December by joint agreement so it could be rebuilt properly, and the new affiliate programme had not yet launched. None of the four was a demand problem.

July. Air conditioning sold £91,987 in July 2025 and £650 in July 2026, because the category had sold out in June. Everything other than air conditioning grew 6.7% that month.

August, September and October 2025 sit between the two comparison periods and are shown for continuity. August 2026 is only eighteen days and is not included in any total.

Section 6

What Maitland actually did

Five hundred and eleven pieces of work were raised on the shared project board between 3 November 2025 and 18 August 2026. Four hundred and thirty-eight were completed. Every person at the agency booked time to this account.

Completed work by type of task

What this means: the longer the bar, the more separate pieces of work were finished in that area. The largest single area is campaign creative, followed by work on the website itself.

Contracted

180 hours a month

£18,000 a month at £100 an hour

Delivered

1,842.88 hours

1 November 2025 to 18 August 2026

Against contract

+115 hours

£11,500 of time at the contracted rate, not invoiced

The 1,842.88 recorded hours, by type of work

Development457.78 hrs · 24.8%
Consultancy402.11 hrs · 21.8%
Creative329.39 hrs · 17.9%
Advertising150.67 hrs · 8.2%
Project management108.96 hrs · 5.9%
Email marketing93.25 hrs · 5.1%
Social40.47 hrs · 2.2%
All other categories81.50 hrs · 4.4%
Project board work, November178.75 hrs · 9.7%

Where the time went, in one bar

46.5%

18.2%

35.3%

Building and fixing the website
856.71 hours, 46.5%
Marketing execution
336.30 hours, 18.2%
Planning, creative, project management and everything else
649.87 hours, 35.3%

Almost half of the year has gone into the website itself, and that is deliberate rather than accidental.

Every channel in this report arrives at the same place. Advertising, email, search and social all send people to a website, and what happens next decides whether the money was worth spending. Fixing the basket, repairing the collection structure and rebuilding how products are found is not an alternative to marketing. It is the part of marketing that decides whether the rest of it works.

Today: about half website, half marketing
Next few months: 60% marketing
Then: 70% marketing

The structural work is finishing. As it does, the balance moves towards campaigns and channels. That shift is the thing to hold us to.

Hours recorded each month

What this means: each bar is the number of hours recorded against this account in that month.

The contract is 180 hours a month. Across the term that is 1,728 hours. 1,842.88 were delivered, which is 115 hours more than contracted, worth £11,500 at the hourly rate. August covers twelve working days.

Section 7

Work that goes beyond marketing

A marketing agency is normally hired to run advertising. Much of this year has gone further than that, working alongside the buying and finance teams on the things that decide whether the marketing has anything to sell.

7 to 13 November 2025
Northern Ireland customs capture built into checkout
Conditional fields now appear at checkout when a delivery address is detected as Northern Ireland, collecting the Economic Operators Registration and Identification number and the Value Added Tax number automatically. Before this the office chased them by hand after the order was placed.
7 November 2025
Bank reconciliation traced for the finance team
The accountant could not balance the bank. One order checked out on 3 November but landed as 31 October. The cause was traced to an approval lag in the finance provider over a weekend. It affects two or three orders a month.
24 November to 10 December 2025
Marketplace order reconciliation
Orders from the OnBuy marketplace were reconciled for the finance function and the process closed out on 10 December.
10 December 2025
Warehouse dead stock analysis
182 slow and non-moving product lines were identified, holding 3,920 units and an estimated £663,792 of value. Twenty-eight of those lines had sold nothing at all in eleven months. Clearing half at a 40% to 50% discount would free around £330,000 of working capital. This is Maitland's own analysis of the stock file.
23 January 2026
Bestseller stock position analysed
Fill rate on the fifty best-selling lines was 77.3%. Ten lines were completely out of stock, including the number three bestseller. Nine more were down to between one and ten units. Across nineteen lines this represented roughly £80,150 a month of demand, which was passed to the buying team as analysis.
26 June to 8 July 2026
Product bundles built from the data
Bundles were constructed from twelve months of frequently-bought-together data, ninety days of line-level sales, seasonality, margin and purchase order dates, rather than from opinion.
2 July 2026
Demand measured on products that cannot be bought
Two out-of-stock product pages were still taking 26,700 visits in ninety days and around 1,300 search appearances a day, with 120 people on the notify-me list. One was the sixth most visited page on the site in July.
26 to 27 July 2026
Stock exposure quantified and taken to the buying team
Thirty-five product lines with over £5,000 of sales this year were sitting at zero or negative stock, carrying £780,522 of demand. Across the whole catalogue, 116 lines were at or below zero, representing 19.6% of this year's demand. This is demand that has already happened on lines that cannot currently be sold. It is not a forecast of recoverable revenue, because a customer who cannot buy one model often buys another.
30 to 31 July 2026
New business prospect list built from Companies House
A feed was built from the public company register. It produced 3,350 addresses of food businesses that registered a limited company in July 2026, segmented into takeaways and mobile food stands at about 45%, unlicensed restaurants and cafes at 25%, licensed restaurants at 18% and event catering at 12%. It refreshes fortnightly on a fourteen-day mailing window from the date of registration. Two artwork versions were produced because roughly one in six registered offices is the founder's home address.
7 August 2026
Advertising reviewed against product margin, not just sales
Six months of advertising was reviewed against the margin on what it sold, rather than against sales alone. The finding is that a large share of the budget is currently steered by revenue rather than by profit, because the margin data needed to steer it by profit does not yet exist for every product. Moving the account onto a profit basis is the largest single improvement available and it is in "What we do next".

Section 8

Channel by channel

Each row is one route by which customers reach the shop. What each channel did this year, and what it should do next year.

Revenue before Maitland, revenue with Maitland, the rise, the 2027 target and the system each is measured by, for every marketing channel
Channel	Before	With Maitland	Rise	2027 target	Measured by
Paid search and shopping	£2,698,435	£3,245,851	+20.3%	£4,591,000	Google Analytics
Paid social	Not run as a channel	£792,535	New channel	£1,111,000	Meta
Direct and organic	£1,252,672	£1,335,033	+6.6%	£1,821,000	Google Analytics
Email	£450,336	£583,167	+29.5%	£847,000	Klaviyo
Affiliate and cashback	Not tracked	£355,557	New channel	£600,000	Conectia
Artificial intelligence assistants	Not recorded	£11,665	New channel	£62,000	Google Analytics

This year covers 1 November to 18 August. The 2027 targets are full calendar years, so this year's figures have been scaled to a twelve month equivalent before the target was set. The rise column compares this year with the same months a year earlier.

How these are measured. Paid search, direct and organic, and AI assistants come from Google Analytics. Email comes from Klaviyo, the affiliate figure from Conectia, and paid social from Meta, because those are the systems that run those channels and they are the same figures reported every week through the year.

These figures will not add up to the total sales shown elsewhere in this report, and they are not meant to. Shopify counts a sale once, at the till. Google Analytics counts the visit that led to it. Figures from different platforms count the same customer differently and cannot be added together. The section “Why the same sale appears in more than one report” explains the difference.

One note on the paid social row. A year ago the channel was three retargeting campaigns spending about £15 a day, shown only to people who had already visited the site. No revenue figure is given for that period because the campaigns were measured on Meta's loosest attribution setting, which counts a sale if the advert was merely shown to the buyer. The campaigns running today are measured on a one-day click setting, which is far stricter. The two are not comparable, so we have not compared them.

One note on the AI assistants row. This channel did not run evenly across the year. It produced almost nothing until April 2026, then stepped up and has held: the last three months average £1,925 a month at the till, which is about £33,000 a year on the measure used in this table. The 2027 target is that run rate plus the content work described later, not the nine-and-a-half-month average plus growth. Read against the average it would look like £44,000, and that would be the channel standing still.

What this year could have been

£780,522

this year's demand sitting on 35 product lines that were out of stock

£136,700

air conditioning demand after the category sold out on 23 June

of those 35 lines are still waiting on a purchase order

Every channel above sold what was available to sell. Where stock ran short, the demand was still there and was measured.

How the channels add up to next year

Paid search+£487,000

Affiliate+£244,000

Affiliate is scaled differently from the others because the programme only started in March. The others are scaled from a twelve month run rate; affiliate is measured against the network's own annual target.

Direct and organic+£134,000

Paid social+£117,000

Email+£109,000

AI assistants+£29,000

Total+£1,120,000

Adding up what each channel should deliver gives £1,120,000 of growth for 2027. That is the figure used in the forecast, and the forecast is built from these six lines rather than from a separate calculation.

Note that these six channel figures cannot be added together to produce total sales, for the reason set out in “Why the same sale appears in more than one report”. What can be compared is growth, and that is what this is.

Every one of those six lines is a marketing channel. Stock availability is a condition on all of them rather than a separate contribution, because a channel cannot sell what the business has not bought.

Buying and marketing are one job with two halves. Stock that nobody knows about does not sell, and marketing with nothing to sell has nowhere to send anybody. Both halves worked this year, and the places they did not line up are the opportunity in the next section.

Section 10

What this costs, and what it buys

Marketing is the largest single line in the business. That deserves a straight answer, not a defence.

Agency fee

£216,000 a year

180 hours a month at £100 an hour

Advertising media

£1,260,000 a year

£105,000 a month, paid to Google, Microsoft and Meta

Total

£1,476,000

£93

what it costs to win one order

£1,476,000 across 16,109 orders

£148

the gross profit one order earns

On the online store, at the margin shown earlier in this report

Every order the website takes costs £93 to win and earns £148. That is the number the board should hold us to, and it should get better rather than worse.

One thing to be clear about: this counts online orders only. The same advertising also puts the company in front of buyers who then order through the sales team or by telephone, and none of that is in the £148.

The same £1.5m, spent two ways

What it buys today

16,109 orders in twelve months
£441 average order
Selling 24 hours a day, every day, across the whole United Kingdom
Scales with demand. June took £875,688 without hiring anybody.

The same money as field sales

Roughly 15 to 18 account managers, once salary, car, national insurance, pension, expenses and management are counted
Four to six months each before they land a large account
An experienced account manager handles a few hundred orders a year
To match 16,109 orders you would need a team several times that size

They do different jobs. One finds and serves thousands of small orders. The other wins a handful of large ones. The business needs both.

The fair question is not which is better. It is whether the £1.5m is working as hard as it can. That is what the rest of this report is about, and it is the question we would rather be judged on.

Section 11

Search visibility

Why appearances rose while visits fell

A Google results page does not look like it did two years ago. The same search now returns an AI summary at the top, then a map, then paid adverts, and only then the ordinary results. An appearance in search has to work far harder than it used to before anyone clicks it.

This is happening to every website, not just this one. Appearing more often while being clicked slightly less is what the whole market looks like right now. It is not a signal that something has broken.

AI summary

Map listings

Paid adverts

Ordinary resultsWhere the free listings sit

A single Google results page today, top to bottom. The free listings sit at the bottom of it.

Distinct search terms the site appears for

35,090

44,937

+28.1%

Search terms actually earning a visit

3,794

2,700

-28.8%

This is already being fixed, and here is what has been done

Separately, and specifically to fix the problem described above, over 40 collection pages have been optimised in recent weeks. These were selected because they were the pages earning high numbers of appearances in search while losing clicks, which is exactly the pattern set out below. On each one the page title and description shown in search results were rewritten, the on-page content was revised, and internal links were added so that the page is better connected to the rest of the site. This work was carried out by Jimmy Dodgson.

Between 26 July and 10 August, the whole collection structure of the site was audited and repaired. All 297 published collections and 12,331 product-to-collection assignments were checked against the live site. The audit found collection rules that had stopped matching, and 460 products, which is 34.3% of the catalogue and £2.31m of sales this year, were reattached to the collections they belong in. Twenty-seven dead links in the main navigation were fixed, four repointed and twenty-three retired, and Mark Armstrong confirmed those live on 10 August.

297

published collections audited and repaired

460

products reattached, £2.31m of sales this year

dead navigation links fixed or retired

Search results do not respond immediately to this kind of work, and the board should expect a delay rather than read the delay as failure.

When a page changes significantly, Google does not simply move it to a new position. It recrawls the page, re-evaluates it, and puts it through a transition period in which its position moves around before settling. This is a known factor in how Google ranks pages after a change, and it means a page that has been improved will often perform inconsistently, or even dip, for a period before the improvement shows. On a site of this size that period is measured in weeks, not days.

The honest date to judge this work is October, not August.

That is why this work is reported here as completed and monitored rather than as a result already delivered. The honest date to judge it is October, not August.

Monitoring is in place. A new measurement was added to the site on 11 August that records every visit to a collection page, so the effect can be seen directly rather than inferred, and the dead-link count is now re-run monthly.

The site is being shown for far more searches than a year ago, and earning visits from fewer of them. Put simply, we have built reach without depth: the new pages appear, but too far down the results page to be clicked.

Visits from searches that do not include the company name, which is the part search work owns, are down between 25% and 40% depending on how much of Google's withheld data is allocated. The generic category terms have moved together: chafing dish from 94 visits to 18, commercial microwave 82 to 13, commercial dishwasher 80 to 22, commercial fridge 63 to 14. That uniformity across unrelated categories points to something affecting the whole site rather than category-by-category competition. That work has now been done and is described above. It is monitored monthly and the honest point to judge it is October.

The one section that grew is the Content Hub, the site's guide and advice section. It produced 707 visits and 139,101 appearances in search across the nine months, against effectively nothing for the twelve months before. That is 1.3% of all search visits, so it is a small, real and growing base rather than a headline.

Visits each month from unpaid search results

What this means: this line is the number of people arriving each month by clicking a normal, unpaid Google result. It has drifted down, with one exceptional spike in June 2026.

Search Console only holds sixteen months of history, so no figure exists before May 2025. June 2026 is the exceptional month described in “June 2026, and what it tells the board”. August is excluded because the month is not finished.

Being found inside AI assistants

Measured twice, in March and again on 14 August, using the same method.

March 2026
14 August 2026

Every measure improved. The audit scored eight areas in total. The six shown here are the ones re-measured in August. The overall score was 34 out of 100 in March against a target of 65.

What changed: the Content Hub now carries real prices and real figures rather than general advice, including a full equipment budget for a 40 to 60 seat kitchen with the arithmetic shown. That is the kind of specific fact an AI assistant will quote. The company address is now on the site, and a genuine external mention has appeared for the first time.

What has not changed yet: there is still no named author with a photograph and a biography, no trade press coverage, and nothing on Reddit. Those are the next three.

Section 12

June 2026, and what it tells the board

The three days in June that made £245,277, and what happened next

21 to 23 June, the three days of peak demand
All other days

What this means: each bar is one day of sales of cooling equipment. Sales rose sharply for three days, then stopped almost completely because there was nothing left in stock to sell.

On 21 June the number of people searching for cooling equipment rose roughly twenty-five fold, from about 130 searches a day to 3,155. Nobody in this room caused that. It was the weather.

What the business did with it is the point. Conversion rates rose across every channel at once on 21, 22 and 23 June. Over those three days the cooling category took £245,277 on 501 orders. Across the whole of June the cooling category took £345,219 across every sales channel, against £32,340 in June 2025.

£245,277

taken in three days, on 501 orders

£345,219

the whole cooling category in June, every sales channel, against £32,340 in June 2025

48 hours

from the demand appearing to the campaigns running

On 23 June the category sold out. Revenue went from £63,732 on the 23rd to £442 on the 24th. Order counts fell from 251 to 158 to 92 to one, while revenue per order rose, which is what happens when a catalogue sells down to its last few variants.

Demand did not stop. Search visits to those pages peaked on 24 June, the day after the stock ran out, and stayed elevated for a fortnight. Those visits earned roughly £3,900 between them, because there was nothing to sell. The equipment did not return.

That is not demand collapsing. It is a shelf emptying.

Two conclusions the board should take from this. First, the marketing responded to a demand event inside 48 hours and converted it hard. Second, the constraint was availability, and the demand that went unserved has been measured so it can be bought for next season. Through the online store, air conditioning sold £324,465 across the 2026 season against £120,421 the year before, roughly 2.7 times, and it was exhausted in three days.

And what it cost in July

July is the only month of the nine, other than January, that finished below the same month last year. £592,352 against £646,786. Air conditioning explains all of it and more.

In July 2025 the business sold £91,987 of air conditioning across 340 orders. In July 2026, with the category sold out, it sold £650 across one order.

A category that took £91,987 in July last year took almost nothing in July this year, because there was none of it left.

Take air conditioning out of both years and July 2026 was £591,702 against £554,799. Everything other than air conditioning grew 6.7%.

With stock on the shelf, our estimate is that July 2026 would have finished at about £658,000 rather than £592,352. That is not last year's figure repeated. It is built from the number of people who actually searched for and visited those pages in July, the rate at which the category was converting before it sold out, and the price the units were selling at. It comes to roughly 153 units.

July 2025 actual
July 2026 actual
Estimate: air conditioning in stock

The third bar is an estimate built from measured demand, not a repeat of last year. It is not a claim about what was earned.

Across the whole period from the sell-out on 23 June to 18 August, the estimate is about £136,700 of sales, or 358 units, that the business could not fulfil.

The demand did not stop when the stock did

This is measurable, not assumed. Google's own figures show how many times eCatering was put in front of somebody searching for air conditioning, and how many of them clicked.

In June, 27,119 appearances and 1,078 visits. In July, with nothing left to sell, still 16,827 appearances and 459 visits. In the first eighteen days of August, 3,320 appearances.

Those July visitors produced one order. The searches they were making were things like 'air conditioner next day delivery' and 'portable air conditioner next day delivery'. These are people who need a unit today, and the business had nothing to sell them for six more weeks.

Times shown in search
Visits to the site

Section 13

What we are fixing next

Not everything moved forward. These are the four things we are working on, what each one is costing, and what we are doing about it. They are in this report because the board should hear them from us with a plan attached.

Clicks cost more than they did

£1.51 to £1.82 per click

Cost per click rose 20.5%. Cost per conversion rose more than that, at 52.2%, so the difference is in how many clicks turned into orders. Advertising delivered 13.9% more clicks across the year, broadly in line with the growth in sales. The important part is when the price rose. Cost per click on this account went from £1.31 in November 2024 to £1.87 in July 2025, a rise of 43% across the year before this engagement. Since November 2025 it has been flat: £1.90 in November, £1.89 in July. The gap between the two years is at its widest at the start and gone by the end: 46% in November, 12% by May, and level by July. The increase had already happened. We have held it level. What we are doing: moving the advertising onto a profit basis rather than a revenue basis, so spend follows margin. That is the single biggest change we would make in the next term.

Cost per click, month by month

The grey line climbs 43% across the year. The green line is flat. The two nearly meet by July.

Fewer people who add to basket go on to checkout

77.1% to 69.2% across the period, with a low of 60.6% in February. Now 77.3%.

This measure fell across a five-month window from November 2025 to March 2026, bottoming at 60.6% in February. It has recovered to between 73% and 77% since April and stands at 77.3% today. We are working through three specific candidates: a warranty prompt that appears on some product pages and not others, shipping cost appearing late in the checkout, and business customers going away for internal sign-off before completing. Each is being tested rather than assumed. What we are doing: the rate is back to 77.3% and we are confirming the cause so it stays there.

Unpaid search visits are down

Non-brand search visits down 25% to 40%

The site appears in 28.1% more searches than a year ago and earns visits from fewer of them. The generic category terms moved together, which points at something affecting the whole site rather than category-by-category competition. What we are doing: this is already in hand and "Search visibility" sets out the work and the timescale.

Average order value fell

£447.11 to £440.87

Down 1.4% across the nine months. Discounting is not the cause: discounts and returns fell as a share of sales, from 2.09% to 1.55%. It is a change in what people are buying, and it sits alongside a 14.8% rise in the number of orders. The business is winning more customers at a slightly smaller basket. What we are doing: the bundle work built in June and July is aimed directly at basket size, and the clearance programme scheduled for September will move slower-selling stock alongside it.

Section 14

Why the same sale appears in more than one report

A short explanation of why the figures in “Channel by channel” must not be added together.

One fryer was sold once. Four different systems will each tell you they sold it.

This section explains why each system reports the same order, why the channel figures cannot be added together, and which single system this report uses when a number has to be quoted.

Why every advertising report overstates itself

Monday
Sees a Facebook advert for a fryer. Does nothing.
Wednesday
Searches for a commercial fryer, clicks a Google advert. Does nothing.
Friday
Opens an email from eCatering. Does nothing.
Sunday
Types ecatering.co.uk into the browser and buys a £1,200 fryer.

Facebook says: that sale was mine.

Google says: that sale was mine.

Email says: that sale was mine.

Direct traffic says: that sale was mine.

The Shopify store says: one fryer. £1,200. Once.

Every one of those systems is telling the truth about what it saw. None of them can see the others. Add their reports together and you have sold the same fryer four times.

What that looks like on this account

What Meta's own dashboard says its adverts produced

£792,535

What Google Analytics recorded from those same adverts

£233,182

What the Shopify store recorded from those same adverts

£137,257

The same thing happens on both platforms

What Google Ads reports it produced

£3,477,134

What Google Analytics recorded

£3,245,851

What the Shopify store recorded from Google

£1,275,597

Three systems, three answers, same adverts. The channel table in this report uses the middle one, Google Analytics, because it is the only system that measures every channel the same way. Shopify’s figure is the most conservative of the three.

Both figures describe the same adverts over the same nine months. Meta counts a sale if the customer saw or clicked an advert at any point beforehand. The Shopify store counts a sale when money changes hands.

Neither is lying. They are answering different questions.

Why Shopify and the advertising platforms disagree

Shopify records a sale against the last place the customer came from before they paid. Meta and Google record a sale against any advert the customer saw or clicked in the days beforehand, whether or not that advert was the thing that brought them back.

So a customer who sees a Facebook advert on Monday, searches for eCatering on Sunday and buys, appears as a Meta sale in Meta's report and as a search sale in Shopify's. Neither system is wrong. They are answering different questions.

Shopify's figure is always the smaller one and always the more conservative. Where this report shows a channel growing, the growth is measured the same way in both years, so the comparison holds whichever figure you prefer.

Put it on the shop floor. If you credited every sale to whichever salesman happened to be standing nearest the till when the customer paid, that man would look like the best salesman in the building by a mile. He is not. He is standing in the right place. Every advertising platform in the world reports like that man.

Two things follow from this, and they are the whole point of this section.

First, the channel figures in this report cannot be added together. Each one is that platform's own count, on that platform's own rules, and the same order appears in more than one of them.

Second, when a channel figure and the Shopify store's own figure disagree, the Shopify figure is the smaller number and the more conservative one. Where this report shows growth, it shows it on a consistent basis across both years, so the comparison holds even where the absolute figure is generous.

Section 15

What we do next

Seven pieces of work for the year ahead. Each one has a number attached to it in this report. One of them is already running.

1
Place the air conditioning order by 15 December 2026.
£489,600
1,200 units, roughly £98,800 of stock. It is worth about £489,600 of 2027 sales and there is no second chance once the season starts.
Owner: Together
2
Continue moving the advertising onto a profit basis rather than a revenue basis.
the largest single improvement available
This is already underway and has been for months. £539,515 of media has been reviewed line by line and the findings acted on. It is slow by nature: the account spends at a scale where a campaign has to be moved onto margin, left to gather enough data to prove the move was right, and only then followed by the next one. Moving all of them at once would destroy the evidence that tells us whether it worked. Expect this to run through 2027 rather than complete in a quarter.
Owner: MaitlandIn progress
3
Recover the abandoned baskets systematically.
£408,000
£170,000 of baskets were abandoned in the first three weeks of August. These are customers already paid for and already at the checkout. Recovering a fifth of them is worth around £408,000 a year.
Owner: Together
4
Build the affiliate programme to its £600,000 target.
£244,000
It has produced £355,557 since March on £21,198 of commission. It only pays when a sale happens, and it is the cheapest growth in the plan.
Owner: Maitland
5
Keep improving the category pages and measure them monthly.
£134,000
Over 40 have been optimised and the whole collection structure has been repaired. This is now a monthly cycle of measuring, improving and re-measuring.
Owner: Maitland
6
Grow AI visibility from 34 towards 65.
£29,000
A named author with real credentials, one piece of trade press, and the company's own data published as original research. All three are content, not development.
Owner: Maitland
7
Put the account managers on the website.
supports the sales team
Nisbets are taking account managers off the road. eCatering is putting theirs on the page, with names and regions, as a reason to buy here rather than there.
Owner: Maitland

Section 16

What the business spends on advertising

The forecast asks for more advertising budget. Before that, here is what is being spent today, so the increase can be seen against a real base rather than in isolation.

Google advertising, month by month, 2026

Google advertising alone, January to July 2026: £597,818. An average of £85,403 a month, ranging from £73,674 in January to £97,691 in May. It is not a flat monthly figure and it never has been.

All paid advertising, 1 November 2025 to 18 August 2026

Google advertising	£806,935	Verified against the platform, month by month.
Microsoft advertising	£39,467	Near zero from August 2025 until the channel relaunched in February 2026.
Paid social	£67,592	Meta's own account export, every campaign in the period.
Total	£913,994	About £95,600 a month across the three platforms, over the same nine and a half months this report covers. All three figures are exact and come from the platforms themselves. Nothing here is estimated. On a calendar basis, the figure usually quoted for the year to date: Google and Microsoft together spent £633,539 between 1 January and 31 July 2026.

What the extra budget would mean in practice

Now

about £95,600 a month

Across Google, Microsoft and paid social.

With the extra £250,000, if it were spread evenly

about £116,400 a month

An increase of £20,833 a month, or 21.8% on top of what is spent today.

What we would actually do

Not spread evenly

Held as an annual envelope and released against demand. See below.

The point of showing this is that £250,000 is not a new order of magnitude. It is a fifth on top of what is already being spent, and the business already varies its own monthly Google spend by £24,000 between January and May without anyone calling it a change of plan.

Section 17

This is not the whole plan

Everything in this report is built from what has already been measured. That is deliberate, and it is why the numbers can be defended line by line.

It also means the report is deliberately narrow. It covers what the online store did and what it can be expected to do. It does not yet contain the work we are currently scoping for next year, which sits outside the website.

There is a second conversation to have before the current term ends, and it will contain things that are not in here.

Section 18

What 2027 looks like

This is a forecast of online sales for January to December 2027, built from every month the shop has traded since March 2023. It is the online line only.

On current course, £7,905,789 for calendar year 2027, against £7,111,237 taken in the twelve months from 1 August 2025 to 31 July 2026. That is £794,552 more, a rise of 11.2%.

The last four years, and where 2027 sits

Actual trading (2024 and 2025)
2026, seven months actual plus an estimate
2027 on current course, £7,905,789
with the three decisions, £9,453,789

Calendar years, online store. 2026 is seven months of actual trading plus an estimate for the rest.

The context this gives: 2024 took £8.22m. 2025 fell to £6.59m, a drop of 19.8%, and that was the year before this engagement began. 2026 is recovering. The 2027 forecast brings the business back to within 4% of where it was in 2024.

How the 2026 estimate is built, and why it is deliberately cautious

January to July 2026 actually took £4,360,155, against £3,843,241 for the same seven months of 2025. That is 13.4% ahead.

We have not used 13.4%. June 2026 was a heatwave. It took £875,688 against £634,198 the year before, and £241,490 of the year's entire gain sits in that one month. It is banked and it counts, but it is not a rate the rest of the year can be expected to repeat, and the months still to come contain no air conditioning season at all.

Strip June from both sides and the underlying rate is 8.6%. July 2026 was in fact 8.4% below July 2025.

August to December 2025 took £2,751,082. To finish 2026 at £7,300,000 those five months need to produce £2,939,845, which is 6.9% ahead of last year. That is below the underlying rate, not above it.

£4,360,155 plus £2,939,845 is £7,300,000. The first eighteen days of August have already taken £347,899, which is ahead of the pace this estimate needs.

It is an estimate and it is labelled as one. Five months of the year have not happened. We have chosen the number we expect to beat rather than the one we would have to explain.

How the 2027 number is built

16,109 orders were taken in the twelve months to 31 July 2026.

If nothing changes

£6,785,789

No improvement, no air conditioning, growth stops. A floor rather than an expectation.

On current course

£7,905,789

The six channels continue as they are. Our recommendation.

With the three decisions taken

£9,453,789

The top of the range, not the middle. It needs all three decisions taken and the demand work started alongside them.

What £9,453,789 does and does not assume

The three decisions are what the money is built from. They are not the whole of what makes the number hold.

Every pound in this forecast is bought in the same auction the business already competes in. The cost of a click has gone from £1.51 to £1.82 in a year. Spending a fifth more into a market that is already 20% dearer works, and it is the right call, but it works less well each year it is repeated. The three decisions get the business to £9.45m. They do not on their own keep it there.

What keeps it there is demand the business creates rather than rents. That is the catalogue, the trade press, the trade shows and the video work set out at the end of this report. None of it is counted as revenue anywhere in this forecast, because none of it has run and we will not forecast something with no history.

So £9,453,789 is not a floor with extras available on top. It is the upper end of what the measured evidence supports, and it assumes the demand work begins in parallel rather than instead.

£9,453,789 is the top of the range, not the base of a bigger one.

Where the growth comes from, item by item

Every figure in this table is money on top of the base of £6,785,789. None of them is a total for that channel, and none of them is a restatement of what the channel does today. Read the column as "additional in 2027".

Sources of 2027 growth, split between work already in motion and growth that needs a decision
What	Additional in 2027	How it happens	Who delivers it
Growth already in motion — channels we runThese six are ours. Maitland runs them day to day, sets the budgets, writes the adverts and the emails, and reports on them monthly. By January 2027 we will have had fourteen months on this account. The first year went on finding out what moves it — which campaigns pay, which categories convert, which emails get opened, what the site does to a mobile visitor. The second year is spent applying that. That is the difference between these figures and a market assumption: we are not forecasting that these channels will grow, we are saying what we intend to do to them.			£1,120,000
Paid search and shopping	£487,000	The account moves onto a margin basis so budget follows profit rather than sales, and stock is available to sell.	Maitland
Affiliate, running a full year	£244,000	The programme launched 11 March 2026, so the year just traded contains only five months of it. A full twelve months at the network's own target.	Maitland
Direct and organic	£134,000	The collection pages continue to be improved and measured monthly. Over 40 are done and the whole structure has been repaired.	Maitland
Paid social	£117,000	The channel continues at its current rate, now finding new customers rather than retargeting people already coming back.	Maitland
Email	£109,000	The remaining automated email sequences complete. Email is already up 29.5% with that work unfinished.	Maitland
Artificial intelligence assistants	£29,000	A named author, one piece of trade press, and the company's own data published as research. All content, no development.	Maitland
Where air conditioning sits in this Air conditioning is the largest single driver inside these numbers, worth about £489,600, and it does not appear as its own line. It is spread across paid search, direct traffic and email, because that is how it sells. That is the honest way to show it. It sells through several channels at once, and no single one of them produces it alone. The order has to be placed by 15 December 2026. After 12 January 2027 nothing lands in time for the season.
Growth that needs a decision — ours to deliver, yours to authoriseThese three do not start until someone signs. The stock has to be bought, the calling team has to be pointed at the basket list, and the media budget has to be released. We can build all three and run all three. None of them begins without a decision in this room.			£1,548,000
Recovering abandoned baskets	£408,000	£170,000 of baskets were abandoned in the first three weeks of August. These customers were already found and paid for through advertising, and stopped at the checkout. Maitland supplies a daily list with basket value and contact permission; the commercial team calls them. This assumes one in five converts. The team believe half is achievable, which would be £1,020,000.	Together
Serving the demand already there	£390,000	35 product lines with over £5,000 of sales this year are sitting at zero or negative stock, carrying £780,522 of demand. Fourteen of them have no purchase order raised yet, which is the single clearest opportunity in next year's number. This is not a forecast of new customers. It is searches, visits and baskets that have already happened and could not complete. This assumes half is served.	Together
Increasing the advertising investment	£250,000 in£750,000 back	Two different numbers and they must not be read as one. £250,000 is the money going in. £750,000 is what comes back. Paid search and shopping currently returns about £4.08 for every £1 of media; this is costed at a deliberately lower marginal return of £3, because more budget buys less efficient impressions. Released in stages, measured monthly, and stopped if the marginal return falls below the point where it pays for itself.	Together
How the extra £250,000 would actually work Not a fixed monthly increase, and not a blank cheque. An annual envelope of £250,000, with an indicative monthly shape so the increase is visible, and unspent months carrying forward inside the year. The reason is that demand does not arrive evenly. June 2026 took £245,277 in the cooling category in three days. A pound spent into that week is worth several spent into a flat February. Spreading the budget evenly guarantees it is in the wrong place when the demand appears. The rule we would work to: the money is held as an annual envelope, not a fixed monthly increase. It is released against measured demand, reviewed monthly, and any month returning below £3 for every £1 stops until we know why. An amount not spent in one month stays available later in the same year rather than being lost. Open point: This is the one point still to agree: whether unspent budget stays available later in the year, or lapses each month.
Total	£2,668,000	Base £6,785,789, plus £2,668,000 of additional revenue, equals £9,453,789.

Three things worth saying about £9,453,789.

None of it is invented. Every figure above is something already measured in this report, applied forward. The abandoned basket total is August's actual figure. The out-of-stock demand is this year's actual sales on lines that cannot currently be bought. The advertising return is the account's own.

And it is not £12m. If the ambition is £15m across the whole business with £3m from the commercial team, the online store would need to roughly double in fifteen months. Nothing in this data supports that. What the online store can credibly do is £7.9m on its own, and approaching £9.5m if the stock is bought, the baskets are worked and the media budget grows.

And it is the ceiling of this method, not a staging post. Everything above is measured, applied forward, and stops where the measurements stop. If the number needs to be larger than this, it needs a different kind of work in it, and that work is the last section of this report.

A word on who drives what

It would be easy to read this table as two separate jobs, and it is not.

Air conditioning is the clearest example. The category sold £245,277 in three days in June because the demand appeared and the advertising, the email and the website put the product in front of people within 48 hours. eCatering bought the stock. Maitland sold it. Neither half of that produces a pound on its own.

The same is true of every line above. Stock that nobody knows about does not sell, and marketing with nothing to sell has nowhere to send anybody.

Three things were considered and left out because we cannot size them honestly: the affiliate programme reaching maturity, the Companies House prospect programme, and any contribution from the new sales team. Each of them may add revenue. None of them is in this number.

Month by month

Month by month, on current course. The three decisions above are not spread across months here, because their timing depends on when they are taken.

Forecast
80% range, low to high

The shaded band is the 80% range. Eight years in ten, the actual figure lands inside it.

June 2027 is forecast at £887,787. June 2026 actually took £875,688 with stock that ran out on the 23rd. The forecast assumes a June like the one just traded, with enough stock to see it through.

The comparison, and why the dates differ

Two comparisons, because they answer different questions.

Against the last twelve months of actual trading, 1 August 2025 to 31 July 2026, the online store took £7,111,237. The 2027 figure on current course is £794,552 more, a rise of 11.2%.

Against calendar 2026, which is the like-for-like twelve months and is estimated at £7.3m because five months of it have not happened yet, the figure on current course is 8.3% ahead.

The forecast is for a calendar year because that is how the board plans. The twelve months to 31 July are used as the comparison because they are twelve months of real trading with nothing estimated in them.

Does this number already contain the sales team's work?

It is the right question to ask, and the answer has two halves.

When a member of the sales team writes an order, it is recorded as a draft order and sits outside the online line entirely. Those orders are in the £1,281,284 offline figure, not in this forecast. There is no double counting there.

When a salesperson visits a site and the customer then places the order themselves on the website, that sale is in the online number and there is no way to separate it out. The website cannot tell the difference; it simply sees an order.

That second half matters less than it first appears. That overlap has always existed, and it is present in every month of the three years of trading history this forecast is built from. It is already inside the base. So if the commercial team's target rises from £2m to £3m and more of that behaviour happens, it pushes the online number above this forecast rather than double counting inside it.

Anything the sales team writes is already outside this number. Anything they influence that the customer orders themselves has always been inside it. The offline push is upside to this forecast, not double counting in it.

Air conditioning: the buying decision behind £489,600 of it

Air conditioning has sold out in every season on record: 2024, 2025 and 2026. Every figure the business has for this category is therefore a measure of how much stock was bought, not how much could have been sold.

In 2026 the category sold 946 units against 626 the year before and 478 the year before that. Demand is rising on its own, and 2026 was also exceptionally hot. When it sold out on 23 June, our best estimate is that a further 358 units of demand went unserved, in a range of 150 to 900.

£489,600

of 2027 sales the category is expected to support

£324,465

taken in 2026 before it sold out in three days

15 Dec 2026

the date the order has to be placed by

To serve 2027 we recommend buying 1,200 units, split 800 of the main model and 400 of the second. That is roughly £98,800 of cash tied up and it supports about £489,600 of sales at £309,200 of gross profit. Sales are shown including VAT, in line with the rest of this report. The cost and the profit are excluding VAT, which is how stock is bought and how margin is measured. A cautious buy of 750 units and a full-demand buy of 1,700 are both set out in the detail.

The date matters more than the number. Chinese New Year falls on 6 February 2027, which removes the usual buffer, and there is no way to reorder once the season starts. The order has to be placed by 15 December 2026. After 12 January 2027 nothing lands in time.

The £489,600 assumes the recommended 1,200 units are bought and sold. Demand in 2026 ran at roughly 1,300 units against 946 actually available, and every season on record has sold out, so the risk sits on the order being placed rather than on the stock selling.

What this forecast is not

This is online sales only. It says nothing about the cost of running the business, the balance sheet, or the offline sales team, and it should not be read as an answer to any question about funding.

The “if nothing changes” figure of £6,785,789 is deliberately below the £7,111,237 the shop has just traded. It assumes nothing improves, no air conditioning returns, and the growth of the last nine months does not continue. It is a floor, not an expectation.

The £7,905,789 on current course is 11.2% above the last twelve months of actual trading, and 8.3% above the £7.3m we expect calendar 2026 to finish at. Over the nine months in this report the business grew 13.2% on the headline and 8.6% once the June heatwave is stripped out. The forecast therefore sits at the underlying rate rather than the headline one, which is the honest assumption rather than the flattering one.

The range around each month is derived from 27 months of testing this method against what actually happened, not asserted.

Section 19

The layer that is not in the number

Six things the business does not do today. None of them are in any figure in this report, because none of them have run and we will not forecast something with no history. They are here because they are the honest answer to how the business grows without simply bidding harder in the same auction.

Cost per click has already risen from £1.51 to £1.82 in a year. Buying more of the same auction is the expensive route to the same place. These six create demand rather than buying it.

Annual catalogue

What we would doTwo editions a year, posted to the existing customer base and the wider database. Company news, new products, best sellers, an exclusive offer, a page on the loyalty scheme, a page on two-man room-of-choice delivery, a page on finance, and the account manager's name and number.

Why it helpsIt puts the business physically on a desk in a kitchen office, which is where the buying decision is made. It restarts conversations with the 7,999 customers who have bought exactly once. Trackable with a unique code and landing page, and it must be built that way or it becomes an unmeasurable cost.

What it costsDesign, print and postage. The most expensive of the four to produce and the hardest to attribute.

Trade press

What we would doRestaurant Magazine, Hospitality Business, MCA Insight, Big Hospitality, Caterer. One editorial pitch a month, story-led rather than sales-led. Two paid placements a quarter for launch moments.

Why it helpsIt builds the credibility a purchasing manager needs before spending four figures with a supplier they have not used. It also feeds AI search directly, because those titles are what the models read and cite.

What it costsThe cheapest of the four. Editorial pitching is time, not media spend.

Trade shows

What we would doOne or two major shows a year. Stand, merchandise, leaflets, sales team in attendance, live demonstration of new equipment, and data capture on the stand.

Why it helpsIt reaches multi-site operators who will never find this business through a search for a single fryer. Those are accounts, not orders. It only works with a plan to capture and follow up every lead, and a target for the team on the stand rather than just a presence.

What it costsThe largest outlay and the slowest to pay back. A show in March produces pipeline that lands later in the year.

YouTube

What we would doProduct demonstrations, buying guides and equipment comparisons. Two a month, shot cheaply and consistently rather than expensively and once.

Why it helpsIt ranks inside Google for equipment specification searches, which is exactly what a buyer looks for before spending four figures. It is the only item on this list that keeps working after you stop paying for it.

What it costsLow to produce. It compounds rather than paying back immediately, which is the argument for starting in January.

AI search optimisation

What we would doDedicated time each month on how the business appears inside AI assistants rather than inside a search results page. Product data, specification tables, comparison content and named expertise, written so a model can quote it accurately.

Why it helpsAI assistants produced £11,665 this year with nobody working on them, and the channel tripled from April onwards on its own. It is the only channel in this report growing without anyone assigned to it. What it does with real attention is not something we can honestly forecast, which is why it sits here rather than in the number.

What it costsTime rather than media spend. It compounds in the same way search does, and at the same slow pace.

ChatGPT advertising

What we would doPaid placement inside ChatGPT, which is opening to advertisers. Same buying discipline as the other paid channels: released in stages, measured monthly, stopped if the return does not hold.

Why it helpsBuyers are already arriving from ChatGPT and converting at a normal order value. Reaching them earlier, before they have chosen where to look, is the same argument that made paid social work this year.

What it costsMedia spend, and it would need its own budget. This is genuinely new: there is no history for it anywhere, not here and not in the wider market. We will not put a revenue figure against something with no history, and that is exactly why it is on this page.

None of these are counted as revenue in the £7,905,789 or the £9,453,789, and they are not extra on top of them. They are what stops £9,453,789 being a single good year followed by a harder one. They are the subject of the second conversation.