A second empty AIO hardware SKU should earn its place by solving a distinct customer or operating problem that the first SKU cannot solve well enough. Before expanding the lineup, count the extra testing, filling, packaging, inventory, training, support, and reorder work the second item creates.[1]
For a small cannabis brand, “more choice” is not a strategy by itself. One clearly defined SKU may produce better learning at an early stage because every sale, fill record, support question, and reorder belongs to the same product. Several SKUs may be justified later, but only when the evidence explains why each one exists.
What counts as a separate hardware SKU?
A SKU is the internal item your team uses to identify, order, receive, fill, package, sell, support, and replenish a specific product configuration.[1] If staff must distinguish the item in inventory or in a work instruction, it probably needs its own SKU and record.
A model, capacity, material, activation method, or physical dimension change may require a separate hardware SKU. The same applies to an output configuration, closure, identification color, or another orderable variant. Packaging-only variants may also need distinct finished-goods SKUs even when the empty hardware is the same.
Do not rely on a short display name such as “white AIO.” Use a supplier, model, variant, capacity, and version reference that receiving and production can verify.
The purpose is traceability, not bureaucracy. When an issue or reorder question appears, the team should be able to identify which exact item was ordered and which process was used.
Give every SKU one clear job
The first SKU should have a simple reason to exist: it serves the brand’s initial tested product and operating path. A proposed second SKU should solve a problem the first SKU cannot solve under the brand’s actual conditions.
Write the problem before naming a product. Examples of a testable job include:
- a different fill capacity requested by a defined customer or channel;
- a control method needed for a documented user workflow;
- a package size or physical format required by a confirmed retail setup;
- a production constraint that the current model cannot meet after testing;
- a separately validated product tier with evidence of demand.
These are questions to investigate, not claims that a second SKU will succeed. “Competitors have many devices” is not a job. “Customers may want options” is too vague. “Five current wholesale accounts requested a smaller fill capacity and agreed to sample it” is a clearer demand signal because the team can verify and follow it.
If the proposed second item cannot be tied to a named problem, keep it in the research list rather than the live lineup.
Compare demand evidence with the learning you already have
Before adding a SKU, review what the first item has taught the business:
- samples requested and completed;
- accepted production runs;
- sell-through by channel;
- repeat-order behavior;
- questions from buyers or operators;
- support cases tied to the product;
- inventory held, used, or written off;
- changes requested but not solved by the current configuration.
Do not interpret a few casual comments as proven demand. Label the source: customer request, sales record, production record, support log, test observation, or internal hypothesis. A hypothesis can justify a sample; it should not automatically justify packaging inventory and a full launch.
Also ask what you will learn from the second SKU that you cannot learn by changing a setting, instruction, package, or selling message around the first. If the two products target the same customer, same occasion, same oil, and same operating need, the new SKU may split the data without answering a new question.
Count the operating burden before approving the lineup
Every additional hardware SKU creates work in several systems.[1]
Product identification and testing
Create a separate product file with the product name, model number, variant, source, version, instructions, and approved sample. Run the relevant oil, filling, capping, storage, charging, and use tests under stated conditions. Do not transfer approval from the first SKU because the products look similar.
Filling and finished-goods control
Write separate work instructions where the models differ. Staff must know which hardware, tools, fill amount, capping method, rest period, inspection step, and finished-goods identifier belong together. Mixing parts or records makes both products harder to review.
Packaging
A second device may need another dieline, insert, instruction, label, pouch, tray, photo, or barcode. Even when the same outer design is reused, verify dimensions and product-specific controls. Button, draw, charging, or setting instructions must match the exact item in the package.
Inventory
Track usable empty hardware, work in progress, finished goods, held units, and in-transit orders separately for each SKU.[1] A combined total can hide a shortage of one item and excess of another.
Training and support
Production, sales, receiving, and support staff need to identify the products and explain the differences accurately. If the second SKU introduces a new control or fill process, the training is not complete until staff can follow the correct record without guessing.
Replenishment
Each SKU may have a different current order route, quantity, packaging condition, ship-from point, or replenishment window. Build a separate reorder point from that item’s real use and current supplier confirmation. Do not assume that one combined purchase plan protects both products.
Decide: one hero SKU, one justified variant, or hold
Use three possible outcomes instead of forcing expansion.
Keep one hero hardware SKU
Choose one SKU when it already serves the defined customer and process. Keep that focus while the team is still learning, demand evidence for another item is weak, or staff cannot maintain separate workflows. “Hero SKU” means the current focus of the lineup. It does not mean the product fits every oil, customer, or future use.[1]
The decision should state what the team will continue measuring and what evidence would reopen the lineup question.
Add one justified variant
Add a second SKU when it has a distinct job and a named customer or operating need. It also needs an approved sample plan, an owner, and capacity for the extra operating work.[1] Define success and stopping evidence before ordering launch inventory.
For example, the team may require a certain number of qualified customer sample requests, a successful process test, approved packaging, and a current replenishment confirmation. The thresholds belong to the brand; they should be documented, not copied from another company.
Hold the candidate
Hold when the need is real but the evidence is incomplete. The candidate can remain in a research queue with the next question, source, sample cost, and decision date. Holding is better than creating a live SKU that no one owns.
Do settings inside one product replace another SKU?
Sometimes a single product offers settings the team can compare before considering another model.[2] Briar, for example, is currently presented as one named 1.0 mL AIO with three output modes and a 5-pack sample path.
Those modes are product settings, not separate hardware SKUs, and they are not proof that one device fits every oil or need. A brand should test its own oil and filling process, record each setting used, and decide what the sample shows under those conditions.
If an adjustable setting solves the documented question, the team may decide that a second hardware item is not yet necessary. If the unmet need involves capacity, architecture, packaging, production, or another factor a setting cannot change, a separately tested SKU may still be justified.
Set the evidence required before the next expansion
End the lineup review with a short gate:
- What problem remains unsolved?
- Who has requested or documented it?
- Why can the current SKU or process not address it?
- What named candidate will be sampled?
- List the filling, packaging, inventory, training, and support work to create.[1]
- What evidence allows a limited launch?
- What evidence stops or postpones it?
If these questions do not have owners and dates, keep the lineup unchanged. Review the same gate after each completed sample, production cycle, or material demand update so the evidence stays current.
There is no universal ideal number of empty AIO SKUs. One is not automatically cheaper, simpler, or better. Several do not automatically increase sales or reduce sourcing risk. The useful lineup is the smallest set of separately tested products that each performs a clear job your team can operate and support with current evidence.
References
- GreenDeagle editorial operating guidance, reviewed 2026-08-01. It supports the decision process described here, not a product-performance, compatibility, inventory, delivery, or supplier-performance claim. Current commercial context: Briar Glass AIO.
- Briar Glass AIO, current product and sample path reviewed 2026-07-31.