Ask ten brand owners in the adult wellness category to explain their manufacturing model and nine will describe a purchasing decision: how much customisation they can afford, how fast they can get stock, how low the first order can go. That framing is comfortable because it turns a strategic question into a spreadsheet exercise with one output column.

It is also the reason so many well-designed products stall at the second order. OEM and ODM are not tiers of a service menu that you graduate through as revenue grows. They are different answers to one question: who owns the thing that makes the product defensible? Answer that wrong and you can spend three years and a six-figure development budget arriving exactly where a competitor started.

Quick answer. OEM means you buy an existing device and own only your brand and packaging. ODM means you buy a factory-designed product and own the industrial design, not the mould or the firmware. Co-innovation means the mould, PCB source, haptic logic and test specification sit in your name. The model is defined by what you own, not by how much you customise — which is why the first contract questions should be about tooling title and firmware escrow, not about price.

Why "OEM or ODM?" is the wrong opening question

The two acronyms describe who designed the product. They do not describe who owns it, who can reproduce it, or what happens to your brand when the relationship ends. In consumer electronics those questions were historically settled by patents and by the cost of tooling. In adult wellness, patents are rare, tooling is cheap enough to duplicate, and the actual competitive asset — the pattern of vibration over time — lives in firmware that most buyers never see.

So the useful question is not "which model do I choose?" It is "which assets do I intend to own in thirty-six months, and does this contract transfer them?" [Teece, 1986] made the point decades ago in the innovation literature: the firm that captures value from an innovation is rarely the firm that invented it, but the firm that controls the complementary assets — manufacturing capability, distribution, and the intellectual property that makes copying expensive. A brand that owns neither the tooling nor the source code controls neither.

The OEM trap: the commodity death spiral

OEM is the default entry point, and the appeal is real. Lower upfront cost, faster time to market, minimal technical risk. You select a chassis from a catalogue, apply your colour and logo, approve a sample, and take delivery in four to six weeks.

The compounding problem is what happens at the second and third season. When five brands sell the same chassis with different silicone colours, the market cannot distinguish them on function, so it distinguishes them on price and on reach. Gross margin that a differentiated launch can hold at 60 to 70 percent compresses toward 35 to 45 percent as the category fills. Marketing spend rises because you now need paid acquisition to be seen at all. The product has not got worse; it has become interchangeable, and interchangeability is priced by the market, not by the brand.

There is a second, quieter cost. An OEM catalogue product carries the factory's cost structure, the factory's component choices and the factory's quality ceiling. If the motor in that chassis is a coin ERM that measures 62 dB at thirty centimetres, you cannot advertise silence. Every positioning statement your marketing team wants to make has to be checked against a specification you did not write. Your brand promise is a function of somebody else's bill of materials.

None of this makes OEM wrong. It makes OEM a deliberate choice about where you compete, and it works when you treat it as such: short product lifecycles, trend-led designs, test-and-learn channel experiments, or a deliberate cash-flow play while a differentiated flagship is in development.

Figure 1 is the whole argument in one image: the labels on the left are tiers of service, while the right-hand column is the only one that changes what your brand is worth at exit. The move from OEM to ODM transfers industrial design. It does not transfer the mould, the PCB layout or the firmware — those typically remain with the factory unless a contract says otherwise. Buyers who assume that a "custom" ODM project means ownership routinely discover the gap only when they attempt to move the product to a second source, which is the exact scenario covered in protecting brand IP with a Chinese manufacturer and in mould ownership contracts.

The ODM illusion: the intellectual property deadlock

As revenue grows, most brands "upgrade" to ODM and believe the differentiation problem is solved. The product looks unique. The ID is yours. The packaging is yours. What is not yours is the behaviour: the pattern library, the acceleration curves, the way the two motors phase against each other, and the Bluetooth protocol that the app talks to.

This is the IP gap, and it is unusually dangerous in this category because the value has migrated from the silicone to the software. A decade ago a vibrator was a motor, a battery and a switch. Today, a mid-to-premium device is a motor pair driven by PWM from a microcontroller running a pattern library, optionally paired to an app over BLE. The silicone is the part your customer touches; the firmware is the part your competitor cannot copy in a weekend.

When that firmware belongs to the factory, three things follow. First, your product roadmap is bounded by the factory's roadmap — if they do not implement the pattern you want, you ship without it. Second, the factory can license the same behaviour to another brand, in a different shell, at a different price point. Third, if the relationship ends, you cannot reproduce your own bestseller. You are, in the precise sense, renting your innovation. [Merges, 2011] frames intellectual property as a mechanism for coordinating investment in exactly this kind of intangible asset; a brand with no title to the intangible has nothing to coordinate and nothing to sell.

The practical test is short. Ask your ODM partner for the source code of the pattern library and the PCB Gerber files, in escrow if not outright. A partner who will not discuss it is telling you what you own.

Where the money actually sits inside a vibrator

The reason the IP conversation matters commercially is that the value stack in a premium device is not where buyers assume. Working from the retail price backwards, a typical EUR 89 device that costs EUR 22 to land breaks down roughly as follows. The silicone and ABS external parts represent a small share of unit cost. The motor pair, battery and USB charging assembly add more. The PCB with its microcontroller, motor drivers and optional BLE module adds more again. And the development effort that produced the pattern library, the thermal budget and the waterproofing seal design — the part that is written once and amortised across every unit — is invisible in the unit cost entirely.

That last line is the point. Development is a fixed cost. Ownership of development is therefore leverage: the brand that owns it amortises it across its own SKUs and pays nothing when it re-uses it; the brand that rents it pays a margin to the party that owns it on every unit, forever, and has no say in when the next version arrives.

Why the standard OEM/ODM framework breaks in this category

Textbook OEM and ODM models were designed for products that do not change after they leave the factory. A moulded enclosure, a cable, a kitchen tool. In those categories, "design" and "manufacture" are cleanly separable and the buyer's differentiation genuinely lives in the visible shell.

Adult wellness is a convergence product: mechanical, electrical, embedded software and human-factors design in one sealed, waterproof, battery-powered object. Four consequences follow, and each one weakens the textbook model.

• The specification is behavioural, not visual. Two devices can look identical and feel completely different because the acceleration curve and the phase relationship between motors differ. Visual approval cannot catch this, which is why AQL inspection alone never certifies a good product.
• Failure modes are silent. A marginal seal fails in month four, not on the line. A marginal motor fails after ninety hours, not in the sample. Reliability is a design property, and the party who designed it controls it.
• Regulatory responsibility attaches to the brand. The CE mark, the UKCA mark, the FCC ID and the declaration of conformity are issued in the brand's name, not the factory's. That asymmetry — legal liability with no design control — is the single most common structural weakness in ODM-first brands.
• Iteration is the product. Category leaders ship pattern updates and firmware revisions. A brand that cannot touch firmware has no iteration loop, and no iteration loop means the product decays against competitors who have one.

A strategic audit: three questions that decide the model

Before the next quotation, answer these three. They are deliberately not about money, because money follows the answers rather than preceding them.

• The equity question. Will this model let me own the assets that make the product defensible — or am I renting them? Name the assets concretely: mould, PCB layout, firmware source, pattern library, test specification.
• The substitution question. If a competitor matched my colour, my logo and my retail price tomorrow, would my product still be better in a way a customer can feel in the first sixty seconds? If the answer is no, the model has not produced differentiation; it has produced decoration.
• The exit question. Does this model build something an acquirer would pay a multiple for, or does it build a reseller with a mailing list? Acquirers buy defensible assets and repeat purchase behaviour. They do not buy a catalogue relationship.

The table is not a ladder to climb for its own sake. It is a menu of trade-offs, and the right answer differs by product line: many healthy brands run OEM bullets for cash flow, ODM wands for scale, and co-innovated flagships for authority. What kills brands is not choosing OEM — it is choosing OEM while believing they bought ODM. For a side-by-side breakdown of the adjacent vocabulary, see OEM vs ODM for adult wellness brands and the white label, private label, OEM and ODM comparison.

Five contract tests that reveal which model you are actually in

Paperwork is where the mirage collapses. Five clauses settle the question faster than any conversation with a sales manager.

1. Tooling title. Who is named as owner of the mould, and where is it stored? If the answer is "the factory," you are in ODM regardless of what the proposal says. Ownership must be written, and it should include a right to remove the tool.
2. Firmware escrow. Is the source code for the pattern library deposited with a third party, released on defined triggers such as insolvency or termination? Without escrow, a supply dispute ends your product line.
3. Component specification register. Is the motor fixed by model number, or only by a functional description? "N20 coreless, 180 rpm plus or minus 10 percent at 3 V" is a specification; "quiet motor" is a wish. Factories default to the cheapest available motor at production time unless the part number is locked.
4. Second-source rights. May you take the tool and the specification to another factory? A clause that forbids it converts your supplier into a single point of failure with pricing power.
5. Change notification. Must the factory notify you before substituting a component, and must it supply a new test report before shipment? This one clause prevents the most common quality decay in repeat orders.

If four of the five are absent, you have bought a service, not an asset. That is a legitimate purchase — it is just not the one most brand decks describe.

What co-innovation actually looks like as an operating model

Co-innovation is a specific division of labour, not a marketing word. The brand supplies the market-facing intelligence: the use case, the price architecture, the identity, the channel constraints, and the outcome the user should feel. The manufacturer supplies the engineering architecture: motor selection, thermal budget, seal design, PCB layout, certification pathway and the test specification that proves the outcome was achieved.

In practice that means six steps, and the sequence matters more than the speed.

1. Outcome definition. Write the sensory and commercial target before any hardware exists: target retail price, target noise ceiling at thirty centimetres, target run time, target price premium over the nearest competitor.
2. Architecture selection. Choose the motor topology against the outcome, not against the catalogue. A single ERM, a single LRA and a dual-motor system are three different products with three different cost and failure profiles, as set out in linear versus ERM motor technology.
3. Specification and DFM review. Freeze a written specification with tolerances, then let the factory challenge it. A partner who accepts every requirement without comment is not engineering your product; they are recording your wishes.
4. Prototype and behaviour sign-off. Approve on measured behaviour — dB at thirty centimetres, run time at medium power, pattern timing — not on photographs.
5. Tooling and validation. Build the tool, then validate against the same specification you approved, including the seal and drop tests that the sample never experiences.
6. Asset transfer. On final payment, receive the mould documentation, the PCB files, the firmware build, the test specification and the component register. If this step is not in the payment terms, it does not happen.

Two of those steps are routinely skipped and both are expensive. Skipping the outcome definition produces a technically fine device nobody wanted. Skipping the asset transfer produces a brand that cannot reproduce its own product, which is the mirage arriving a year late.

What co-innovation costs, and when it is the wrong answer

Total development cost for a co-innovated dual-motor device typically lands in the range of USD 12,000 to 40,000 before certification, split across industrial design, mechanical and electronics engineering, tooling, prototype rounds and validation. Certification for the EU and US adds several thousand more. Timeline runs twenty to thirty weeks from brief to first shipment, against four to six weeks for a catalogue OEM order.

That arithmetic is a poor fit for three situations: a first product with no validated demand, a trend-led SKU with a twelve-month commercial life, and a market entry where the regulatory pathway is not yet proven. In each case, buying an existing platform and spending the development budget on distribution is the better allocation. The mistake is not choosing OEM. The mistake is choosing OEM and then expecting OEM economics to fund a premium brand.

The verdict: build for the exit, not just the launch

The winner in this category will not be the company with the most SKUs. It will be the company whose products are hardest to copy, whose supply chain is not a single point of failure, and whose specification library compounds with every launch. Those are all ownership questions wearing a manufacturing costume.

If the goal is turnover, OEM is efficient. If the goal is a niche position with a distinctive look, ODM gets you there faster than building from scratch. But if the goal is a business that a larger consumer group would acquire, the assets have to be yours: the mould, the firmware, the pattern library and the test specification. The cheapest way to discover you do not own them is to try to move production.

For the full procurement pathway — brief, sample, tooling, audit, first shipment — start with what OEM/ODM manufacturing actually involves, then compare it against the first-shipment sourcing guide.

---

Evokomoribi builds dual-motor and single-motor platforms under written tooling-ownership and component-register terms, so the specification you approve is the specification you receive on the third order. If you are auditing whether your current model gives you the assets your brand needs, contact our engineering team for a co-innovation assessment, or review our OEM/ODM capability first.

---

References

1. Wikipedia. "Original equipment manufacturer." Wikipedia, The Free Encyclopedia.
2. Wikipedia. "Original design manufacturer." Wikipedia, The Free Encyclopedia.
3. Teece, D. J. (1986). Profiting from technological innovation: Implications for integration, collaboration, licensing and public policy. Research Policy, 15(6), 285–305.
4. Merges, R. P. (2011). Justifying Intellectual Property. Harvard University Press.
5. Porter, M. E. (1985). Competitive Advantage: Creating and Sustaining Superior Performance. Free Press.