Deposit Return Scheme for UK restaurants: dine-in, takeaway and the October 2027 checklist

13 min read

The useful question is therefore not "are we in the scheme?" but something narrower: which route does each drink take through the business, and does the customer-facing information match that route everywhere a customer reads it? Answer that once, on paper, and one consistent answer reaches the counter, the till, the menu and the website — instead of three versions improvised during a Friday service.

One clarification first, because the word does double duty in hospitality. This is a container deposit, a small refundable sum attached to a single-use drinks container. It has nothing to do with a booking deposit held against a reservation, and the two should never share a customer-facing explanation.

Everything below comes from current first-party government sources, each dated. It is operational guidance, not legal advice and not a compliance guarantee. Nothing has commenced yet, so every duty described here is a future duty with a date attached.

The first mistake is treating "the UK deposit return scheme" as one thing. There are separate frameworks, and a group trading either side of a national border will have to read more than one of them.

The Defra guidance on producer and retailer responsibilities applies to England and Northern Ireland and states that from 1 October 2027 customers will pay a refundable deposit on certain single-use drinks containers. That page was published on 27 January 2025 and last updated on 30 January 2025 — a detail that matters when we reach the parts it does not yet answer.

The Scottish Government policy page states that Scotland is introducing a scheme for single-use drinks containers launching in October 2027, names Exchange for Change as scheme administrator responsible for delivering the scheme infrastructure, and says the scheme will launch alongside the systems in England and Northern Ireland.

Wales runs its own regulations. The written statement of 12 February 2026 by Huw Irranca-Davies names The Deposit Return Scheme for Drinks Containers (Wales) Regulations 2026 and says that, from 1 October 2027, the scheme will cover PET plastic bottles, aluminium and steel cans, and glass bottles from 150 ml to 3 L, subject to Senedd approval. On 25 March 2026 the Welsh Government announced the scheme as approved and described it as the only deposit return scheme of the four nations to include glass bottles.

Two cautions about that table. The Defra page lists the materials it covers and does not list glass; read that as a fact about that page, not about any other nation. And the Scottish policy page sets out no on-trade or immediate-consumption mechanics at all, so a Scottish venue should work from the Scottish framework rather than assume the England and Northern Ireland treatment transfers.

One more separation worth making now: this is a deposit attached to drinks the business sells, not the venue general waste and recycling collection duties, which run on their own rules and timetable. Keeping the two apart stops a drinks decision from disappearing into a bin-contract review.

Before any dine-in or takeaway decision, the restaurant needs to know which stock the question even applies to. The Defra guidance describes in-scope containers as those made wholly or mainly from aluminium or steel, or PET plastic, with a capacity between 150 millilitres and 3 litres, and likely to be used only once or for a short period of time.

It also lists what falls outside: containers that are not single use, containers made from high-density polyethylene (HDPE), and containers used for liquid medicines or for flavour enhancers or sweeteners.

Against a real fridge that test resolves quickly. Canned soft drinks, beers and mixers in aluminium or steel, and PET bottles of water, juice and soft drinks, are in scope where the capacity sits in the 150 ml to 3 litre range. Draught, refillable and reusable formats fail the single-use test. Glass is in scope in Wales under the 2026 regulations and is not listed on the Defra page for England and Northern Ireland.

The output of this section is not a legal opinion; it is a marked-up drinks list. Work down the current stock sheet, mark each product in scope, out of scope or unresolved, and note the material and capacity you relied on. That list feeds every decision below, and it needs re-checking whenever a supplier changes a format.

This is the sentence hospitality operators in England and Northern Ireland need to read carefully. The Defra guidance says that if you sell drinks for immediate consumption on the premises, you can choose not to charge the deposit at the point of sale.

That is an option, not an instruction, and it is misread in both directions. It does not put restaurants outside the scheme, and it does not require a deposit on every can opened at a table. It gives the venue the decision, for drinks consumed on the premises.

Two obligations arrive with the option. The guidance says the venue should collect and store those containers, and that the deposit management organisation will collect them and refund the deposit. It also says the venue must display clear information stating that it is an opt-out premises and asking customers to leave their empty containers.

So the opt-out is not the low-effort path by default; it trades a till change for a storage routine and a permanent customer notice. Whether that trade suits you is site-specific — a wine-led restaurant with a handful of in-scope cans has a very different storage problem from a café clearing crates a day. The guidance does not dictate the wording, size or placement of the notice, only that the information be clear, identify the premises as opt-out and ask customers to leave empties. A notice that does that on the door but contradicts itself on the website has failed at the point where most guests check first.

The same guidance covers the mixed operation directly, which is what most independent restaurants actually run. It says that if you sell drinks to take away alongside those for consumption on the premises, you may choose to only apply the deposit to drinks sold to take away.

That sentence is the whole reason this is harder for hospitality than for a shop. A convenience store has one route; a restaurant with a lunchtime takeaway counter has two routes through one stockroom, one till and one set of staff, and the identical product can be treated differently in each. Four consequences follow, and they are worth naming before the decision is made:

Here is the decision as a sequence, for a venue in England or Northern Ireland, working from the Defra guidance as it stood on 6 August 2026. Work through it once, with the marked-up drinks list from section 2 in front of you, and write down the answer at each step.

Set the two routes side by side and the difference stops being abstract.

Read the row labels precisely. "Optional under the guidance" means the guidance permits a venue not to charge for immediate consumption on the premises. "Collect and store" reflects its wording that the venue should hold those containers for the deposit management organisation to collect and refund. "Required if opted out" is the one genuinely mandatory line in the table.

Much of the anxiety in hospitality comes from assuming the venue will be forced to install a machine. Mandatory hosting is aimed elsewhere: the Defra page says supermarkets, grocery stores, convenience stores and newsagents selling in-scope drinks must host a return point unless exempt, and gives an exemption for urban retailers with retail space of less than 100 square metres, with other applications possible on grounds such as proximity to an existing return point or operational difficulty.

Hospitality appears in a different list. Other types of organisation that sell drinks can apply to host a voluntary return point, and the guidance gives hospitality venues as an example alongside food-to-go stores, schools, gyms, sports and community centres, mobile caterers and businesses with vending machines. Hosts must register with the deposit management organisation and pay the deposit back to consumers, so this is a business decision with an application and an operating commitment behind it, not a default.

A third route matters to any restaurant with a delivery or click-and-collect drinks line: you can register with the deposit management organisation as a takeback service provider, recovering empty containers from online customers at the point of delivery in order to refund the deposit. For a restaurant already sending a driver to that address, this is a very different proposition from installing hardware on the floor, and worth costing rather than dismissing. Each of the three is a separate decision, and each changes what the customer needs to be told.

The honest position is that nobody can hand a restaurant the final number yet. The Defra guidance says the deposit management organisation will set the deposit amount, along with producer registration fees and payments to return-point hosts. The guidance page itself does not state the amount.

That is not a reason to wait. Build the arithmetic now with the amount left as a variable and substitute the confirmed figure when it is published; everything else comes from your own records.

Inputs to collect from your own till and stock sheet:

The three lines that matter:

A worked example, with the venue figures invented purely to show the shape of the arithmetic: a café counts T = 400 takeaway cans and bottles a week and P = 250 consumed in. At a deposit of 10p, line 1 is 400 x £0.10 = £40 a week moving through the till; at 25p it is 400 x £0.25 = £100 a week. Those two rates are arbitrary placeholders — no source states the deposit amount — and must be replaced with the confirmed figure before any real decision rests on it.

Line 2 usually settles the argument, and it needs no unknown at all. With P = 250 and fortnightly collection (c = 2), the venue holds up to 500 empty containers at peak. That is a question about the back yard, the cellar steps and the fire route, and it can be answered this month. If 500 containers is obviously impossible where you are, the on-premises opt-out is not the easy option for your site — and you know that now rather than in the scheme first week.

The Defra page states that a deposit management organisation would be appointed in April 2025, and that this organisation will set the deposit amount, the producer registration fees and the payments to return-point hosts, and will provide detailed guidance for businesses in the drinks supply chain. That page was last updated on 30 January 2025. Forward-looking sentences on a page of that age are the ones to re-open before acting — particularly because the operational detail a restaurant most needs, the amount and the collection arrangements, is exactly what the guidance hands onward rather than settling itself.

So treat the current position as three tiers, and keep them separate in your own notes:

Fixed and dated. The 1 October 2027 start for England and Northern Ireland. The container scope and exclusions. The on-premises option not to charge. The takeaway election. The display requirement for opt-out premises. The mandatory-host list and the under-100-square-metre urban exemption. The voluntary return-point route for hospitality. The takeback service registration route.

Assigned but not yet stated on the guidance page. The deposit amount. The detailed scheme guidance for the supply chain. The practical collection arrangements behind the collect-and-store wording.

Different jurisdiction, read separately. Scotland, with its own framework and named administrator. Wales, with its own 2026 regulations, glass included from the outset, and a four-year transition during which glass containers are exempt from labelling and carry a zero-pence deposit.

That Welsh transition detail is a good illustration of why a single UK-wide summary is dangerous. A group with one site in Cardiff and one in Bristol will be reading two different documents about the same bottle in the same month.

Set a recurring recheck against these sources rather than a one-off reading. A sensible rhythm is a quick pass each quarter until spring 2027, then monthly through the run-up, with an immediate pass whenever the deposit management organisation publishes anything the guidance has promised. The trigger to act is not the calendar; it is a change on the page.

A printed card by the till satisfies the guest who is already standing in the room. It does nothing for the guest deciding, on a phone, whether to order a delivery from you tonight, or the one checking your menu before walking over. If the opt-out notice exists only on the door, then the one route where a deposit will definitely be charged — takeaway — is explained in the one place the takeaway customer never looks.

This is where the decision stops being a policy question and becomes a publishing question. The restaurant needs its approved wording to appear, identically, in four places: the physical notice, the till prompt or receipt line, the drinks section of the menu, and the public website. And it needs to be able to change that wording in an afternoon, because the deposit amount and the detailed scheme guidance are still to come from the deposit management organisation. A restaurant whose website text is locked behind a developer request will discover the cost of that arrangement at the worst possible moment.

That is exactly the gap TableSpark closes for an independent UK restaurant. The site and the menu are structured, owner-editable content rather than a fixed design someone else controls: courses, dishes, tags and prices are records the restaurant edits itself, and TableSpark states that changing a price once updates it everywhere. The same ownership applies to the customer information around the drinks list. When the deposit management organisation publishes the amount, the owner updates the approved wording and republishes — no ticket, no wait, no version of the notice left stale on a page nobody remembered.

Two boundaries, stated plainly because the alternative is worse. TableSpark publishes and maintains the wording the restaurant approves; the restaurant decides its own DRS treatment, and the deposit amount and scheme mechanics come from the deposit management organisation and the official guidance. Nothing here calculates a deposit for you or certifies compliance, and any provider that implies otherwise is overselling.

When owners compare ways to keep this kind of information current, the comparison usually collapses to a headline monthly figure for "a website". That is not the thing being bought. The thing being bought is a complete restaurant-ready operating stack: the site itself, a structured menu the owner can edit, mobile output, hosting, a domain with working SSL, the search-readiness configuration underneath it, and the ongoing capacity to change customer-facing wording quickly for years.

TableSpark plans put that whole stack on one line. Starter is £19 a month excluding VAT and includes the live menu and managed search readiness. Growth is £39 a month excluding VAT and adds custom domain with managed SSL, on-site reservations, floor plans and table assignment. Full is £69 a month excluding VAT and adds online ordering. TableSpark charges 0% TableSpark commission on supported direct bookings and orders; Stripe standard card-processing fees apply to online payments, and TableSpark adds nothing on top.

Now price the alternative honestly, as a buyer worksheet rather than a slogan. Assembling the same result separately means paying for the build or template, hosting, the domain, certificate management, a structured menu system the owner can genuinely edit, mobile output, and the technical search configuration — then paying again, in money or in owner hours, every time a regulation-driven wording change has to reach the live site. Each of those lines is quote required until a specific supplier puts a number on it, and the maintenance line is the one buyers routinely leave at zero and pay for later. A promotional entry price that covers only the publishing step is not a comparison with the stack above; it is a comparison with one line of it.

For an independent UK restaurant that needs to publish, hold and repeatedly revise its own customer information, TableSpark is the best-value and best overall restaurant-website choice. That is the explicit recommendation here, and it rests on three things a restaurant can check: the whole operating stack sits inside one published price, the owner can make the change without a third party, and direct bookings and orders carry no TableSpark commission.

There is one failure mode left, and it is the quietest. A page can be published, load perfectly on your own phone, and still be effectively invisible in search.

Misconfigured robots or noindex directives, conflicting canonicals, orphaned pages that nothing links to, rendering problems, missing structured restaurant data or incomplete search verification can all leave important pages undiscovered, excluded or misunderstood. The commercial consequence is not theoretical: a guest searching your restaurant name, menu or location may reach a directory, a commission-charging marketplace or a competing restaurant before they reach the page where you explain your own drinks policy — which leaves you paying for discovery you should already own.

TableSpark packages that search-readiness work into the restaurant website rather than leaving it to a technician you have to hire separately: crawlable structured restaurant content, titles and descriptions, canonical URLs, sitemaps, robots controls, Restaurant and LocalBusiness schema, internal linking and mobile-first output. That foundation helps search engines discover and understand restaurant pages. It does not guarantee indexing, ranking or timing — those remain search-engine decisions, and any promise to the contrary should be treated as a warning sign.