Peach Gum Dessert Manufacturer: Raw Material to Finished Bowl

The three peach gum desserts we make ourselves all began as an amber lump that nobody could eat. Turning that resin into a bowl you can open and spoon is a development problem before it is a production one. The work follows a fixed order: settle what carries the gum, set the solids level the bowl is specified at, then prove the sealed pack holds through a retort at 121 °C. Each of the three went through that route on the same line, and the line can run the result under a customer's own name.

The part of this work that is easy to underestimate is how little of it is culinary. What the ingredient can carry on a label honestly gets settled long before the artwork does. What the process behind the finished pack can evidence afterwards gets settled by the records the line keeps, because a retailer or a customs officer will eventually ask to see them. Both answers come out of the plant, and both are expensive to change once a launch date is fixed.

A peach tree that has been cut or stressed bleeds a gum to close the wound. It dries on the bark into an amber lump, and that lump is what the trade calls peach gum. It is not a fruit, not a seed and not an animal material, and that last point matters for what may legally go on a label.

The material also varies. A gum lot is an agricultural input, not a synthesised one, so colour and behaviour differ between harvests and suppliers. That is why the incoming inspection in stage one is a real gate rather than a formality.

Peach gum is a polysaccharide, the plain word for a long-chain carbohydrate. It belongs to the same broad family as the plant gums already used as thickeners and stabilisers. Two things follow, and both are commercial rather than academic.

The first is that it is a carbohydrate, so it is not a protein and cannot be presented as one. Ding et al. reported in the International Journal of Biological Macromolecules in 2019 that the purified gum carries no detectable protein, which is why a collagen claim on a peach gum dessert has nothing to stand on. The second is that it binds water. A dry piece dropped into liquid swells and thickens whatever surrounds it, and that behaviour is the gum's whole contribution to a bowl: body, plus a texture between soft and springy.

The four lines below are the ones that decide what you can print on a pack.

Read that table as constraints rather than selling points. The first two lines tell you the gum is a plant carbohydrate with nothing animal-derived to declare, which keeps it straightforward to label. The third line is where the product's value actually sits: the gum thickens the liquid around it, so the body of the bowl comes from the ingredient itself rather than from a starch or a gelling agent added later. The fourth line is the one that constrains your marketing, because no amount of copywriting turns a carbohydrate into a protein ingredient. Everything the gum does for the product it does through texture, and texture is legitimate to sell on.

Turning the resin into a bowl a consumer will buy comes down to three decisions, and all three are made before anyone chooses equipment. Peach gum supplies texture rather than flavour, so it needs a liquid base and a set of companion ingredients before it becomes a dessert at all. Solids then separate a thin sweet soup from a bowl you eat with a spoon. Shelf life decides whether the finished pack travels by sea or by air, and it drives the packaging decision more than the recipe does.

Nutrition is where peach gum gets oversold, so it is worth being exact about the four points that survive scrutiny.

• The gum is a carbohydrate. The protein and fat in a finished bowl come from the other ingredients in the recipe, not from the gum.
• The gum contains no collagen and cannot supply any. It is plant material, and the purified fraction shows no detectable protein at all, which is the finding Ding et al. published in 2019.
• The energy in the finished bowl comes mainly from the added sugar. A bowl sweetened with rock sugar is a dessert, and the calorie count should be read as the recipe's, not the gum's.
• Per-serving figures belong on the pack, and must come from analysis of the finished product in the form you actually ship.

A buyer rarely arrives with one product in mind. These three were formulated on this line, and each was built against a different brief. Two are 168 g bowls specified at 35% solids or above and holding 12 months at ambient temperature below 25 °C, opened and finished within 24 hours of refrigeration. The third is 228 g and heavier. What separates them is not the gum, which is the same material in all three, but what sits around it.

A range rather than a single bowl is the usual starting position: the three answer different markets. A floral bowl suits gifting and seasonal retail, and the rose bowl suits the same shelf with a different flavour story. The 228 g bowl suits catering and volume retail, where portion size matters more than presentation. A buyer who wants one of these reformulated rather than ordered to a fixed specification is asking for ODM work, where the formula is ours to change.

Purified water, peach gum, fried chestnuts, red osmanthus, white fungus and rock sugar, declared in that order. The chestnut does the structural work. At the same 35% solids, a floral bowl with no solid companion reads as a thin sweet soup; osmanthus on its own is a top note dosed for aroma rather than for body. The chestnut gives the spoon something to find and the mouth something to chew.

Water, rock sugar, peach gum, organic white fungus at a stated 5 g addition, red dates and double red roses. The white fungus figure is the one number this product puts on an ingredient. A stated addition weight is worth more to a buyer than a flavour description, because it is a figure you can write into a specification and hold a supplier to. Rose works the way the osmanthus does, as a top note rather than a base.

A 228 g bowl, six to a box, with a 100 g serving size. Tapioca is what makes it different: the pearls add bulk and a chew the gum alone cannot give, which is why this bowl carries a higher solid load than either 168 g product. Its storage instruction is also the loosest of the three, a cool dry place rather than a fixed ceiling near 25 °C. This bowl's ingredient line is published in general terms rather than as a full declaration, so ask for the specification before briefing artwork against it.

Buyers usually arrive with a flavour brief. The cost is decided somewhere else, and it is decided earlier. Solids are the first lever. A bowl specified at 35% solids carries more than a third of its weight as ingredients other than water, and ingredient cost scales with that number, not with how the flavour reads on the label. Two briefs that sound identical can land at very different costs because one buyer wants 30% solids and the other wants 45%.

Format is the second. A sealed bowl needs a container, a closure and a retort cycle validated for that exact pack. A pouch needs a different validation. Neither is cheaper in the abstract; the cost difference sits in the packaging line and in the validation work, both of which are fixed per SKU rather than per unit.

Shelf life is the third, and it is usually the one buyers underprice. An ambient 12-month shelf life is what allows container shipping, distributor warehousing and retail listing without a cold chain. A shorter shelf life quietly removes all three and replaces them with air freight and short-order production. None of this argues against a flavour-led brief. It argues for settling solids, format and shelf life before the flavour conversation starts, because those three set the floor under your unit cost.

One more fixed cost sits above all three, and buyers rarely price it: validation. A retort cycle is validated per SKU, so every format or solids level you add is a separate validation before it can ship. Tooling works the same way, since the bowl, the closure and the carton each carry a one-off setup that does not scale down with a small first order. A brand planning four flavours in two formats is planning eight validations rather than two, and a low launch volume is rarely as cheap per unit as the same product ordered later at volume.

The line runs in three stages: material in and ready, then a sealed pack, then heat, testing and dispatch.

The order matters more than the count. Nothing is retorted before it is sealed, because the retort only reaches commercial sterility inside a closed pack. A gap at that join is where a batch stops being traceable.

1.  Raw material selection

2.  Washing

• Every incoming lot is inspected against its own specification before release, and the record is filed under the batch number.
• The gum itself is graded on arrival, because the extraction route changes how the material behaves later in the process.

• Which specification does an incoming gum lot have to meet before you release it, and can I see one filled-in record?
• Where does the gum come from, does the origin change between orders, and what happens to a lot that fails inspection?

1.  Proportioning and feeding

2.  Filling

3.  Sealing

• Ingredients are weighed to the formula on a fully automatic feeding line rather than by hand, which is what makes the 35% solids specification repeatable.
• Fill weight is checked on a statistical sampling plan, and net-content compliance is verified against the destination market's rules rather than a single house figure.
• Under the US rule for thermally processed low-acid foods in hermetically sealed containers, four values are entered on the record at the time they are observed. They are the closing machine vacuum on a vacuum-packed product, the maximum fill-in or drained weight, the code number and the date. That requirement sits in 21 CFR 113.100(a), and it is written as a timing rule rather than a content rule.

• What tolerance does the feeder hold, and what is the sampling plan for fill weight, stated as a frequency rather than as "we check"?
• How is the seal checked, and is that check recorded per pack or per batch?

1.  Retort

2.  Cooling

3.  Coding

4.  Testing

5.  Cartoning

6.  Dispatch

• The pack is retorted at 121 °C, and the cycle is validated per SKU and logged per batch rather than run on a shared setting.
• A scheduled process is not a setting copied from another product: it is established by a qualified person for that product and that container, as 21 CFR 113.83 requires.
• Cooling brings the pack down on a defined schedule, and the end point is recorded in the retort record rather than left to the operator's judgement. Coding then puts the batch number, the production date and the best-before date on every pack.
• Every batch is checked against its product specification before release, and the result is filed in the batch file.

• What is the retort cycle for my SKU, and who signed the scheduled process?
• How long does a batch sit between retort and release, and can you show me one full batch file from incoming lot through to release?

The table below sets the eleven steps against the control point that governs each one and the record that should exist because of it.

Read the fourth column first. It is the only part of the eleven steps a buyer can check from outside the factory, and it is where a supplier's account of the process either holds up or does not. A supplier who can walk you through the third column but cannot produce the fourth is describing a process, not running one. When you visit, ask for one complete batch file. Ask for it by batch number rather than by date, so the supplier has to find a specific run instead of describing how records are kept in general. The gaps show up at the joins, between the incoming lot record and the retort log. That is where a recall investigation would start.

Two things decide whether a peach gum project works, and they are not the same. The first is the retort. A bowl that holds at ambient temperature for 12 months is a thermally processed product, and the validation work behind that shelf life is what a brand is really buying. This line runs 30+ high-speed lines, and retort capacity decides whether your launch order and your reorder can both be scheduled rather than queued.

The second is what we will not put on a label. Peach gum invites claims the ingredient cannot support, and collagen is the most common of them. We will not print around a claim the material cannot carry, because that cost lands on the brand rather than on the factory.

ODM, OEM and private label are three different contracts, and the first question in any brief should be which one you are asking for. What separates them is who owns the formula, and that single fact decides everything downstream: how many sampling rounds you get, who signs off the specification, and whose document governs release.

The distinction is not administrative housekeeping. It decides who carries the cost of a sampling round that goes nowhere, and who owns the recipe if the relationship ends. Buyers who leave the question open tend to discover the answer during a dispute rather than at the start, when it is still free to ask.

Under ODM we own the formula. You bring a product direction, a target market and a price position, and we develop the recipe, the process and the specification around it. Samples come back as a finished bowl with a draft specification attached, so you are tasting the product and reading the numbers at the same time.

Under OEM you own the formula and we run it. You send the recipe and the specification, we confirm the line can hold the tolerances you have set, and then we produce against your document. If your formula calls for a solids level or a pack format this line cannot hold, we say so before sampling rather than after.

Neither route is subcontracted. Whether the brief is ODM, OEM or private label, the product runs on the same line, through the same 121 °C retort cycle, under the same QC checks, and lands in the same batch file. Someone buying a private label programme is buying exactly the same process as someone buying an OEM run, and the batch file looks the same either way. That matters at audit time: a brand that has been told its OEM work is outsourced has no way to verify the process, and no way to answer a retailer asking who actually made the product.

We do not make pharmaceuticals, and we do not take orders that require a health claim on the pack. We do not formulate around a claim the ingredient cannot support, and we do not accept a brief where the label has to say something the analysis contradicts. If any of that describes your project, we are the wrong supplier and it is better to know now.

Two mistakes show up in almost every first conversation, and both are cheap to avoid. Neither is about competence. Both are about what a buyer is willing to accept as evidence.

The pattern is the same in each case. A supplier answers a question with a description, the buyer accepts the description as proof, and the gap only surfaces later, when a retailer or a customs officer asks for the document standing behind it. Fixing this costs nothing at the briefing stage and a great deal afterwards.

A supplier who can describe an eleven-step process fluently may still hold no records from it. Process narration is the easiest thing in this industry to produce and the hardest thing to falsify, because the description is written once while the records are made daily. A supplier can describe an eleven-step line accurately and still have nothing to show from a recent run. The test is not whether the story holds together, but whether a document exists for a specific date and a specific batch.

• Signs: the answer arrives as a paragraph with no document behind it, and no batch number is mentioned anywhere.
• Ask instead: how many batches of this product did you run last quarter, and can I see one batch file.
• Verify: the records should join up. An incoming lot number should appear again in the retort log for the same day.
• Watch for: a certificate of analysis offered as a substitute for a batch file. They are different documents.

Certificates describe a management system. Records describe your product. A certificate tells you a system was audited at some point, against a scope that is written on the certificate itself, but it does not tell you what happened on your batch. Buyers ask for certificates because they are easy to request and easy to supply, which is exactly what makes them a weak screening test. Treat the certificate as the entry ticket and the batch file as the evidence.

• Signs: a wall of certificates is offered before any question about the specific SKU is answered.
• Ask instead: which certificate covers this product, who issued it, and what is the scope written on it.
• Verify: ask to see the scope statement, not the logo. A certificate that covers a different product category tells you nothing about your bowl.
• Watch for: a claim that a facility registration means a product is approved. It does not.

If you are scoping a peach gum dessert, the useful next step is a specification conversation rather than a price request. Tell us the pack format, the solids level you are targeting and the market you are shipping into, and we will come back with what this line can hold and what it cannot. That exchange is faster than a price request as well, because a price quoted without a specification is not a quotation.

Ask for one sample bowl, one draft specification and one redacted batch file. Those three documents tell you more about a supplier than any number of certificates, and they cost the supplier nothing to produce. If a supplier cannot send them, that answers your question too. Send the same request to two suppliers and compare what comes back; the difference between the two replies usually tells you more than either price would.

Ready to scope a peach gum dessert? Ask for a sample, a specification and a batch file.

Peach gum is a polysaccharide, a long-chain carbohydrate, produced as resin by the peach tree. It is not a protein, and the purified gum shows no detectable protein at all. Nutritionally it behaves as a carbohydrate that binds water and thickens the liquid around it, which is why formulators use it for texture rather than for flavour. On a specification it appears as one ingredient line alongside the base liquid, the sweetener and the companion ingredients.

No, and this is the most common label mistake on peach gum products. The purified gum shows no detectable protein, so a collagen claim has nothing to support it. It matters commercially because peach gum is frequently marketed alongside collagen products, and a claim the analysis contradicts is a compliance problem in the destination market rather than a copywriting preference. Ask to see the analysis before you approve artwork.

The 168 g bowls on this line are specified at 35% solids or above, and that is a specification value rather than an average. Solids are the main lever on both texture and ingredient cost, so the figure belongs in the specification before sampling starts. Adjusting it after tasting is how a project ends up with a bowl that reads well as a sample and costs more than the target once it runs at volume.

Twelve months at ambient temperature with no refrigeration, on this line. That shelf life comes from the retort cycle rather than from a preservative, which is why the process is validated per SKU instead of run on a shared setting. A shorter shelf life removes container shipping and distributor warehousing from your options and replaces them with air freight and short-order production.

At minimum a certificate of analysis for the lot and the full set of shipping documents. For the US market the batch record should also carry the closing machine vacuum, the fill weight or drained weight, and the code number and date, because 21 CFR 113.100(a) requires those to be entered at the time they are observed. Ask for one complete batch file rather than a sample of each form.

Ask for one redacted batch file and read it backwards, from the shipping documents down to the incoming lot record. A batch file that joins up across the whole sequence tells you more than a factory tour, because it shows what the line recorded rather than what it is capable of. Do that before you discuss price and before you approve artwork, since it is the one piece of evidence a supplier cannot assemble retroactively.

Yes. A private label programme runs on the same line and the same retort cycle as ODM and OEM work