Terpene Batch Consistency: Why Flavor Drifts Between Runs

Your last three production runs used the same blend, the same loading and the same hardware, and the third one tastes different. Before anyone blames the fill line, understand where drift usually starts: in a field, months before the material reached anybody's lab.

Botanical raw materials vary. That is not a supplier failing, it is agriculture. What separates a supplier you can build a brand on from one you cannot is whether they formulate around that variation or pass it through to you.

This post is about the mechanism and the controls. For what the resulting paperwork should look like and how to interrogate it, our guide to terpene testing and reading a COA is the companion piece.

The plant is the variable

Botanical terpenes are isolated from plant feedstocks, and plants are not manufacturing equipment. The same species, grown in a different place or cut in a different month, gives you different chemistry.

The scale of that variation is easy to underestimate. Zhang and colleagues sampled 974 individual camphor trees across 35 populations in 13 Chinese provinces and published the results in Molecules in 2023. Essential oil yield ranged from 0.01% to 3.46% depending on where the tree grew. Individual compounds tracked geography in different directions: eucalyptol and linalool correlated positively with latitude, camphor correlated negatively with longitude.

One species. Same genus, same common name, same supply chain. Yields spanning more than two orders of magnitude and a compound profile that shifts systematically with where the material came from.

Now stack that across the dozen or more feedstocks that go into a single formulated blend, each with its own harvest window, its own weather year and its own supply chain. Then add the fact that cannabis cultivar names are not reliable identifiers either. That variation is the raw input every terpene supplier is working with.

The question is what they do with it.

Recipe versus specification

This is the whole distinction, and it is the one thing to take away.

Formulating to a recipe means the blend is defined by its inputs. Take this much of feedstock A, this much of B, mix. When feedstock A comes in at a different concentration this quarter, the recipe still says the same thing and the output has moved. The supplier has done nothing wrong by their own definition, because they followed the recipe.

Formulating to a specification means the blend is defined by its output. The finished material must hit these compounds within these tolerances. When feedstock A comes in different, the formulation adjusts to compensate, and the finished blend lands in the same place it did last quarter.

This is how the Native® line is built, and it is why a cultivar profile ordered in March should match one ordered in November.

Formulating to a specification costs more. It needs analysis on every incoming feedstock, not just on the finished blend, and it needs the formulation capability to adjust rather than repeat. It is also the only approach that produces a consistent product from an inconsistent input.

When a supplier says their product is consistent, that is the question underneath: consistent inputs, or consistent output? Ask which, and ask what happens on their end when an incoming lot comes in off-profile.

The four places drift enters

Upstream, in the feedstock. Covered above. Controlled by incoming analysis and by formulating to a specification.

In blending. Weighing tolerance, mixing time, temperature during blending, and whether the batch was homogenized properly before filling. A large batch that was not fully mixed can be within spec at the top and out at the bottom.

In storage, at either end. Terpenes are volatile and they oxidize. A blend that sat warm in a half-empty drum for eight months is not the blend that shipped. This one is frequently the customer's, not the supplier's, and it is worth ruling out before making an accusation. Our guide on shelf life and storage covers the four controls that matter.

In your process. Fill temperature, mixing time, the distillate lot, the hardware batch, how long the product rested before it shipped. More "supplier inconsistency" turns out to live here than anyone likes to admit.

Working out whose drift it is

Before the phone call, spend an afternoon on this. It will either save you a difficult conversation or make it a much stronger one.

1. Compare the COAs, lot to lot. Same blend, three lots, dominant compounds side by side. If they are within a couple of percent of each other, the terpene is probably not your variable.
2. Check your retains. If you keep a sealed retain of each incoming lot, smell the retain of the batch that worked against the retain of the batch that did not. If the retains match and the products do not, it is your process.
3. Audit your own conditions. Pull the batch records. Fill temperature, loading percentage, mixing time, rest period, distillate lot, hardware lot. Look for the thing that changed. There is usually one.
4. Check the age and storage of the terpene you used. Date opened, container fill level, where it was stored.
5. Then call your supplier, with the lot numbers and the analysis in hand.

Suppliers take that call very differently when you arrive with data. So does the resolution.

What a supplier should be doing

The controls worth asking about, in the order they matter:

Incoming feedstock analysis. Every lot, not a sample of lots. Without this, formulating to a specification is impossible.

Analytical depth. A blend matched on four dominant compounds can still drift in character, because the compounds that make a profile recognizable are often present in fractions of a percent. Ask how many compounds they quantify, and ask to see a full panel rather than a summary.

Retain samples. Does the supplier hold a retain of every production lot? If they cannot pull the exact material they shipped you eight months ago and re-run it, they cannot investigate a complaint properly.

Documented tolerances. What is the accepted variance on each dominant compound, in writing? "Consistent" is a sales word. A number is a commitment.

Traceability from lot to feedstock. When something does go wrong, can they trace a finished lot back to the incoming materials that made it? That determines whether the fix takes a week or a quarter.

A stated response when a lot fails their own spec. Does it get reworked, rejected, or shipped with a note? The answer tells you a lot about who you are dealing with.

The Werc Shop side of the business has been testing cannabis terpenes since 2011, which is the reason this is the part of the operation we are most willing to be interrogated on. Our case studies cover what those controls look like across multi-state and Canadian license-holder production.

What consistency is worth

Your brand's reputation lives or dies on product consistency. A customer who buys the same cart twice and gets two different experiences does not conclude that agricultural feedstocks vary. They conclude your product is unreliable, and they buy something else.

The cost of drift is not the failed batch. It is the repeat purchase that never happens, and those do not show up in a quality report.

That is the case for paying for analytical depth rather than for the cheapest liter, whichever part of the range you buy from. The supplier who is running full panels on incoming feedstock and adjusting formulations to hit a specification is carrying a cost, and it lands in the price. What you are buying with it is a product that behaves the same way in March as it did in November.

FAQ

Why does my terpene blend smell different between orders? Most often either upstream feedstock variation that the supplier passed through rather than formulated around, or degradation in storage at one end or the other. Compare lot COAs and check your retains before assuming which.

How much batch to batch variation is normal? It depends on the compound and the blend, which is why the useful move is to get documented tolerances from your supplier in writing rather than accept "consistent" as an answer. A supplier who cannot state a tolerance is not measuring against one.

Can botanical terpenes be as consistent as synthetic ones? Formulated to a specification, yes, within a defined tolerance. That takes incoming analysis on every feedstock lot and the capability to adjust the formulation, which is where the cost sits.

Should I keep retain samples of every terpene lot? Yes. A small sealed retain, stored cool and dark, costs almost nothing and it is the only way to settle a drift question after the fact. Without retains you are arguing from memory.

What should I ask a supplier about consistency? How many compounds do you quantify, do you analyze every incoming feedstock lot, what are the documented tolerances on this blend, do you hold retains, and what happens on your end when a lot comes in off-specification.

Does drift matter if the difference is small? It matters if your customer can detect it. That is a sensory question rather than an analytical one, and the way to answer it is a blind triangle test on the finished product rather than an argument about decimal places on a COA.

Is drift more likely in some product formats? Yes. Formats where terpene does all the sensory work, so carts, tinctures and low-dose formats, show drift that a strongly flavored edible would mask entirely.

Where to take this next

Pull the COAs for your last three lots of the same blend and put the dominant compounds side by side. Then pull your own batch records for the same period. One of those two documents contains your answer.

If it turns out the variation came in the bottle, you have a decision about whether to move supplier. Talk to our formulation team about what formulating to a specification would mean for your product.

---

About the author: The Entour™ team is a division of The Werc Shop®, the first commercial cannabis laboratory to test for terpenes back in 2011. The team formulates True To Plant® terpene profiles for licensed cannabis brands across the US, Canada, and tribal markets.