The Hidden Crisis in Cannabis: Lab Shopping, Unreported Pesticides, and Consumer Risk

---

Key Takeaways:

• COAs aren’t always trustworthy: Lab reports, once seen as a reliable guarantee of product quality, are increasingly undermined by issues like THC inflation, unreported byproducts, co-elution errors, contaminated products passing safety checks, fake lab reports, and lab shopping practices.
• Consumers can’t rely on COAs alone: Even a legitimate-looking COA can be misleading, especially without deeper research into the testing lab’s reputation, lawsuits, sampling practices, and regulatory oversight.
• The industry urgently needs stronger regulation: True consumer protection requires tougher oversight, standardized testing protocols, mandatory full-panel safety screening, independent sampling, and more severe penalties for labs and brands that manipulate or falsify results.

---

“Today, many COAs are not worth the paper they are printed on, or the pixels that make up their digital image.” – Dr. Chris Hudalla, President and Chief Scientific Officer at ProVerde Labs

In the early days of the hemp and cannabis industries, lab reports – called Certificates of Analysis, or COAs – were seen as a gold standard for product quality.

Consumers quickly learned that the only real way to know what they were buying was to demand third-party testing, and COAs became the de facto assurance of safety and potency. For many products, the COA remains the only piece of information standing between consumers and blind trust.

But somewhere along the way, that assurance started to erode.

With problems like THC inflation, unreported byproducts, and a host of other hidden issues becoming harder to ignore, it raises an uncomfortable question: Can consumers really trust the COAs attached to hemp and cannabis products?

We spoke with lab testing insiders to uncover the deeper issues – and what consumers need to know.

---

The Role of COAs in Consumer Decision-Making

“COA” is the adopted term for a certificate of lab analysis, colloquially just called a “lab report.”

This is basically exactly what it sounds like: a company sends a product they made to a lab for testing, and the lab does the science and produces a report which basically tells you what’s in there.

These “what’s in there?” tests almost always include the potency of different cannabinoids (i.e., how much CBD or THC it contains), and generally include safety tests for pesticides, residual solvents, microbial contamination, heavy metals, and mold. Other tests, like terpene profiles, are not as commonly performed but are still widely offered by testing labs.

In ideal situations, this gives consumers a guarantee that the product offers what it claims to and that it doesn’t contain anything dangerous. It ensures that consumers can avoid unsafe products and offers an avenue for regulators to keep tabs on what’s being sold and ensure products follow applicable rules.

But standardization is lacking, with labs using varying methods and different states setting different rules for what a COA has to contain. Already, the issues with the system are showing: do you know if a HPLC-UV or GC-MS-MS test is more reliable for testing a delta-8 product? Do you think politicians know this?

And this is just the tip of the iceberg.

---

Key Problems Undermining the Reliability of COAs

Problem 1: Unreported Byproducts in Delta-8 THC and Other “Converted” Products

As we’ve discussed previously, one major problem with COAs is that they don’t always tell you everything they found.

Dr. Erik Paulson, lab manager at InfiniteCAL, explained to us:

Most labs test for what is required or what is requested, and do not look further into the potential problems. Certain chemical reactions, such as the chemical conversion of CBD into delta-8 THC, can produce several byproducts and use of different processing chemicals that are left in the final product. While the intended compounds are generally part of targeted screenings for determination of content, the byproducts and residual processing chemicals are not usually screened for.

A lot of the time, the byproduct compounds have not even been formally identified or studied, making it even harder to inform consumers.

Dr. Chris Hudalla, Chief Scientific Officer and President at ProVerde Labs, has been a prominent voice in this discussion for years now, and stressed the issues labs face in these situations:

So, if I, as a lab, document a signal from a contaminant that we never see in true cannabis products, but that contaminant does not even have a name yet, how do I call that out on my COA? Worse, if that contaminant does not have a name, then we certainly do not understand the toxicity of that contaminant.

And this last point is arguably the most crucial.

In terms of trust in the COA, the fact that these compounds are generally not listed despite being detected is a huge red flag. But it’s not just an unknown chemical; it’s an unknown chemical that could make the whole product much more dangerous than it seems.

ProVerde tackles this by adding a disclaimer to any COAs where byproducts were detected, but we’re unaware of any other lab that does this.  

---

Problem 2: THC Inflation – How High Can You Really Go?

“THC inflation” refers to the practice of reporting higher THC potencies than the actual product contains, and it is a big problem in cannabis testing.

One 2023 study sampled products sold in Colorado, and found that around 70% of samples were over 15% lower in THC than reported, and three products were less than 50% of reported values.

A study from 2021 looked at about 200,000 products and the THC levels they reported, finding an unusual spike in frequency as you cross the 20% potency threshold – a sure sign of a manipulated distribution.

Dr. Paulson explained to us:

Sadly, THC inflation is still a very large issue across the US. Some product types are worse than others with regards to the inflation of THC. For example, flower will routinely be labeled and sold with THC values listed 20-40% higher than their actual result (i.e. 25% THC flower being labeled at 37.5%).

---

Problem 3: Bait and Switch Sample Practices

“Bait and switch” sample practices are when a company sends off a sample to a lab for testing, but where the sample is not representative of the product on the shelves. The test can be totally accurate, but the product tested isn’t the same as the one sold to consumers.

Dr. Paulson pointed out that many state-regulated cannabis programs protect against this by requiring the lab or an independent sampler to sample the product for testing, but he notes that hemp does not have the same protections:

In the hemp product industry, testing is almost exclusively performed from samples that are sent in or dropped off to the labs, so a producer can certainly send in any sample for testing, receive a COA with data that represents the sample they sent in, then use that COA to represent a completely unrelated product.

---

Problem 4: Co-Elution Issues

“Co-elution” is a term in chromatography for when two or more compounds can’t be differentiated with the method being used.

The signal peaks in the chromatograph overlap, and it’s possible for one signal to completely “hide” behind the other. Worse, this means that they’re both counted towards the total for whichever compound the lab is expecting.

CBD Oracle had this issue recently, when we tested a full-spectrum CBD oil product for our Best CBD Oils for Sleep list. The original COA listed delta-9 THC, but our independent test identified that at least some of the peak was HU-331, basically the result of oxidizing CBD. The signal from HU-331 co-eluted with that from THC, leading to an overestimation of the THC on the original lab report.

This is interesting from a scientific perspective, but absolutely devastating from a consumer perspective. Think about it: if you had just read the original COA, there is no way at all you could have known this.

Dr. Hudalla commented that labs need to be part of the process, and that misidentification is “entirely possible” if a producer adds something like HU-331 but doesn’t disclose it. He notes that “most labs are under pressure to get results out as fast and cheap as possible.”

Dr. Hudalla told us that HPLC methods have been optimized well enough that this isn’t an issue with natural phytocannabinoids, but that things change when you add synthetic cannabinoids and the byproducts that generally come along with them. He showed us how common byproducts in delta-8 THC products can easily co-elute with THC under HPLC testing, but that gas chromatography (GC) testing resolves the signals more accurately:

---

Problem 5: Contaminated Products Getting a “Pass” on Safety

Safety testing for pesticides, mold, residual solvents, microbial contamination, and heavy metals is a crucial part of lab testing for both cannabis and hemp, but it isn’t always what it seems.

The cannabis industry has a big issue with contaminated products getting a “pass” on safety.

Dr. Chris Hudalla explained to us:

The biggest contaminant issue in many states is microbial contaminants. It is far easier to find a lab that will ‘PASS’ moldy product, than it is to resolve microbial contaminants in cultivation.

Pesticides are also a substantial issue, with an LA Times report finding widespread pesticide contamination in California cannabis. In particular, chlorfenapyr – a category 1 pesticide not allowed in cannabis cultivation – was the most commonly-detected pesticide, along with others like pymetrozine which is not even screened for in California cannabis.

Dr. Paulson commented, “The fact that contaminated weed is hitting the shelves is a travesty. While honest mistakes can happen, the intentional certification of clearly moldy or pesticide-laden material by certain labs is wholly not acceptable.”

---

Problem 6: Fake Lab Reports

This is an obvious one: you can’t always assume that a lab report shared alongside a product is genuine. All it takes is a small change or two to turn a non-compliant product into a compliant one, or to make your product look more potent than it really is.

We’ve investigated this issue as part of other work in the past. Notably, we found many fake COAs when investigating the hemp THCA flower market. One example we identified was doctored numbers on a COA to slightly boost the quantity of THCA, but did not make updates consistently across the document.

Ryan Bellone, commercial director at KCA Labs, told us they’d seen faked lab reports with KCA’s name on them: 

Yes, we see plenty, but they’re never perfectly similar to what we report out. QR codes are first to be removed, then client information. We have an effective watermark and an uncommon font on our reports to help protect this.

Ryan added that checking QR codes and emailing the lab involved are the “best defenses,” but also suggested some other checks, “Also using common sense. Does the font look like it differs throughout the report? Is the report really old, such as a year or more? Does the company/brand name match the material?”

---

Problem 7: Lab Shopping

All of the above is made worse by the practice of lab shopping.

This is the practice of sending samples to multiple labs and simply going with the one that offers the most favorable results. If one lab flags your product for pesticides but another doesn’t detect them, which report do you publish?

Even if you are honest, it’s very easy to see how people opt to publish the clean report and quietly shelf the other one.

This is also easy to justify and difficult to argue against, because who’s to say which report is accurate? As a consumer, even if you had access to both, it’s essentially impossible to tell. The labs would likely both tell you their test is correct.

---

What Consumers Can Do

The challenge for consumers is that most of these problems are not something you can notice by looking at COAs – it takes some more in-depth research into the labs themselves and careful web searches to find the problems.

It didn’t always used to be this way, as Dr. Chris Hudalla pointed out:

When cannabis was becoming increasingly accepted and legalized, state by state, I was a staunch advocate for testing, to ensure consumer safety. Back then, a Certificate of Analysis (COA) could be viewed as document with meaning, intended to help ensure consumer safety. But today, many COAs are not worth the paper they are printed on, or the pixels that make up their digital image.

Adding that, “Without a consumer digging into the history, performance and capability of an individual lab, how would they know how reliable a lab or their COA is? And most consumers don’t have the expertise to evaluate those labs. Regulators, for the most part, are certainly not doing it.”

As Dr. Hudalla argues, there isn’t too much you can do as a regular consumer. But that doesn’t mean there’s nothing you can do.

If you’re trying to evaluate a lab test, here are some things you should check:

• Scan the QR code (and don’t trust a report without one): Scanning the QR code on a COA should take you to the same COA on the lab’s website. This is a good way to spot obvious fakes and manipulation. You should also look out for changes in font and similar inconsistencies. 
• Google the lab: While you likely won’t turn up much from a web search, you can check Glassdoor and other sources of reviews from insiders for some idea of how the lab is operated and whether there are problems. You should also check the “news” tab to see if anything relevant comes up.  
• Check lawsuits and complaints: There have been some high-profile lawsuits and complaints involving cannabis testing labs that are worth checking. For instance, a lawsuit filed by InfiniteCAL and Anresco names several labs allegedly involved in deceptive practices, and another in Massachusetts from MCR Labs names eight. Likewise, there are other complaints that can be easily turned up via web search.
• Cross-check where you can: Searching some of the information from the lab report itself (such as sample IDs) and other information about the product including the lot/batch number can sometimes turn up alternative reports on the same product.

---

Industry Solutions and What’s Next

The responsibility for this continuing travesty ultimately falls on the regulators. We can point fingers at dishonest actors all we like, but there will always be dishonest actors in any industry. That is why we regulate industries, and all of these problems can ultimately be solved with more robust regulation.

Dr. Hudalla told us, “On the regulatory side, there has to be more consideration of consumer safety. But any changes can only come from better regulation, which can only happen when regulators hire appropriately trained staff.”

He pointed out that in Massachusetts – which is over a decade into a regulatory program – he is unaware of a single microbiologist on staff at the Cannabis Control Commission, despite microbial contamination being an ongoing issue in the industry.

He asked, “How do we express concerns to our regulators when they do not even understand the words coming out of my mouth? They are easily swayed by misinformation, and viewpoints that are presented which are skewed based on financial conflicts of interest.”

Dr. Paulson pointed out that, for hemp in particular, there is a lack of oversight and limited motivation to establish oversight:

What this lack of oversight also leads to is the absence of a requirement to test for certain things that can be harmful. While it is commonplace for producers to test and provide a COA for a standard cannabinoid panel to consumers, very often contaminants are not being screened for because the tests cost extra and neither the retailers nor regulators are telling them they need to perform the testing.

So how can these problems be solved? Based on comments received for this article and our own takes, there are some key steps that could help:

• Require reporting of common byproducts and unknown compounds: Unreported byproducts are pretty much guaranteed in delta-8 THC and other “converted” hemp products, but at the very least, regulators can require labs to look for common ones and report anything they find. A mandated disclaimer like ProVerde’s wouldn’t make the market safer, but it would keep consumers informed.
• GC testing for converted cannabinoid products: Byproducts are less likely to co-elute with gas chromatography (GC) testing, and this should be required for delta-8 THC and other hemp products made by converting from one cannabinoid to another.
• Contaminant screening should be required: Many hemp products in particular only include potency tests, without safety testing. It should be obvious to everybody that this is not acceptable, it should be “full panel” tested or not on the shelves at all.  
• Increase enforcement overall: This is simple. All of the problems identified in this post would be improved with more enforcement from regulators. You need to take an active approach to finding and correcting errors, not wait for other labs to blow the whistle.
• Hire (or work with) knowledgeable people: Analytical chemistry is complicated, but you can’t regulate labs if you don’t understand the issues. Regulators without expertise are easier to mislead, so hiring or consulting with knowledgeable people is essential.
• Test “off-the-shelf” samples: Companies sending in their own samples for testing enable bait-and-switch testing. Simply testing products actually being sold to consumers would fix this.
• Escalate penalties for repeat offenders: If being dishonest makes you more money than any fines you receive, there is no financial incentive for labs to fix the issues. Penalties should increase for each subsequent offense to create this incentive.
• Regulators need to vet testing labs and re-test samples to identify offenders: It’s better to have fewer testing labs doing a good job than more labs with some bad actors in the mix. Labs should be vetted by regulators. Take an off-the-shelf product and send it to multiple labs, then investigate any anomalies.

It should go without saying that this is not an exhaustive list: there is much more that needs to be done to ensure the safety of consumers, and ultimately the reputation of the industry.

---

Bottom Line

It’s important to note that we are not saying that lab testing is worthless – while flawed, it remains the best assurance consumers get that what they’re buying is safe. We need to improve COAs, not throw up our arms and give up on quality assurance altogether.

But as things stand, you cannot trust a COA if you haven’t done more research on the lab and the brand than simply reading the report. Consumer pressure helped to make COAs and industry norms, and consumer pressure can also encourage reliable COAs going forward.

Trust, but verify. Paranoia about fake reports or misleading numbers isn’t really useful – because you can easily distrust every report on these grounds – but skepticism is absolutely required. Consumers, responsible companies and reputable labs all need to stand together on this and demand more from regulators.