Patents can be valuable corporate assets, but they do not establish that a company’s product performs as advertised. Describing future variations before they have been built or tested can be good patent strategy, helping preserve protection options as technology develops. Read carefully, patent filings show what a company is seeking to protect and point investors to the right questions about its technology.
When we last revisited Elizabeth Holmes and the Theranos story, the television series The Dropout starring Amanda Seyfried was making waves. We’re back again to discuss Elizabeth and Co. as Lance Oppenheim and Nathan Fielder’s new documentary You Can See Everything dominates theatres and the cultural zeitgeist. This time it’s to explore the question “How could a company whose blood testing technology fell so far short of its promises secure patents?”
While watching the film, one scene stood out to me as someone who works regularly within the intellectual property space. In it, Holmes lays out the components of a Theranos device on a kitchen table and discusses the blood tests it can run. Fielder questions her use of the present tense, asking whether the device “can” perform any test or only “could.” Did she mean the device was able to perform the tests now, or that it had the potential to do so after further development? Holmes explains that running additional tests would require developing the necessary chemistries: the chemical methods used to carry out each test*.
That condition changes what the statement promises. Saying a device can run a test suggests that the capability already exists. Saying it could run that test if the necessary chemistry is developed leaves essential work outstanding. Someone hearing the first statement would reasonably expect more than the possibility described by the second.
Patent filings often describe inventions before a finished product exists, sometimes including variations that have never been tested. Looking ahead in this way can be good strategy, particularly when the technology may evolve considerably over the years ahead. If we were advising investors evaluating a company like Theranos, we would want them to understand why that is permitted (along with the value of that approach) and what they can reasonably conclude from the filings.
Theranos’s IP portfolio illustrates how easily the significance of a patent can be stretched beyond the invention it actually protects. Understanding that boundary is crucial for anyone developing, investing in, or communicating the value of proprietary technology.
What Did Theranos Promise?
Founded in 2003 by Elizabeth Holmes, Theranos promised to make blood testing cheaper, more convenient, and more accessible, with machines that could run a broad range of tests on a few drops of blood. According to the US Securities and Exchange Commission’s 2018 allegations, the company managed to raise more than $700M dollars from investors through false or exaggerated statements about its technology and business. In reality, its proprietary analyzer could only perform a small number of tests, while the company used third-party equipment for the vast majority of patient tests.
Theranos ceased operations in September 2018. In January 2022, Holmes was convicted of investor fraud and was later sentenced to 135 months in prison. The film follows her in the 34 days prior to starting her sentence.
In the wake of the collapse, commentators such as Daniel Nazer argued that the company’s patent portfolio gave it unwarranted credibility that helped to prop it up, making it easier to attract investors. In our original blog post, we discussed the risk of that misplaced confidence. Opinions such as these highlight the need for investors to know just what the filing and granting of a patent can tell you, and what it cannot tell you.
A company can hold patents while its product falls well short of what it promises. To understand how, we need to look at what the patents protect and what the patent office has actually examined.
What’s in a Patent Application?
A patent application includes a written description of the invention, called the specification, often with drawings. It also generally includes a set of proposed claims: the statements that define the scope of legal protection sought.
A pending application is a request for protection. Its proposed claims may change during examination or never be granted. In a granted patent, the issued claims define the bounds of protection and are central to determining whether a competing product or service infringes a patent or not.
Something described in a specification is therefore not necessarily covered by the issued claims. A specification may discuss a platform's possible uses extensively, while the granted claims protect a particular component or method. Reading the description alone can leave a very different impression from reading what was actually claimed and allowed.
For investors, this means looking beyond the patent count and asking how the protection relates to the technology the company is relying on to build its business.
Does a Patent Mean an Invention Works?
When the US patent office examines an application, it considers whether the claimed invention is new, non-obvious, and useful, among other requirements. It also examines whether the application adequately describes the invention and explains how to make and use it.
Some readers assume that the requirement to be useful means the patent office has verified that the invention works as advertised. It has not. The applicant must identify a specific, practical use that is credible to someone knowledgeable in the field. That assessment does not amount to independent testing of a company’s product or confirmation of every capability promoted to investors.
Canadian patent law applies similar principles. An invention’s utility must be demonstrated or soundly predicted as of the filing date, and the description must allow a skilled person to put the invention into practice.
There is a separate requirement called enablement. The disclosure must provide enough information for a person skilled in the relevant field to make and use the full scope of the claimed invention without an unreasonable amount of experimentation. This is assessed using the application and knowledge available at the relevant filing date.
An application cannot simply describe a hoped-for result and leave someone else to solve the essential technical problems. Equally, a patent covering one part of a device does not establish that the complete device delivers all its advertised functions.
Can a Patent Protect Something That Has Not Been Built?
Inventors are not generally required to provide working samples of their invention when they file. They may have figured out the key parts of their invention and how to implement them while still having testing, debugging, and product development to complete. The first-inventor-to-file system also gives them a reason to file early, once they have enough detail to support the application.
US patent practice also permits prophetic examples: descriptions based on predicted results rather than work actually performed. These must be clearly distinguishable from working examples reporting actual experiments, and the USPTO specifically warns against presenting predicted results as though they have already been achieved. Patent practice, in other words, draws the same line Fielder pressed Holmes on what has been done versus what is expected.
This helps explain why an untested invention can appear in a patent application. Early filing can support protection for an invention that reaches the market later, provided the application contains enough technical information to support the claims. It cannot reserve a desired outcome while leaving the essential work for future researchers, and a later breakthrough cannot fill a gap in what the application needed to explain when it was filed.
Consider the device Holmes describes in the film. She clarifies that the platform could run additional tests if the necessary test-specific chemistries were developed, which makes that unfinished work central to understanding her claim. A component of the platform might still be patentable while other parts require substantial development. A patent protecting that component would tell an investor something about the company’s IP position. It would not establish that the missing chemistries had been developed or that the platform could perform every promised test.
If the claims instead covered a testing method that depended on those chemistries, the question would be whether the application adequately explained how to carry out that method. That depends on the particular claims and disclosure. The scene alone cannot tell us whether any Theranos patent met those requirements.
For investors, hearing that further development is needed should prompt questions about what remains unresolved and how that work relates to the invention being protected.
Why Describe Untested Variations at All?
Patent practitioners often advise inventors to describe alternative implementations and anticipated developments in their initial filing, including versions that have not been tested. This can help a patent portfolio remain relevant as the technology and its market evolve. An application focused too narrowly on today’s product may leave out variations that become commercially important several years later.
A thorough specification can preserve options for later claims. In the US, a continuation application allows an applicant to pursue different claims based on the original disclosure, provided the required application chain remains pending. It cannot add new technical content. Describing alternative ways to implement the invention at the outset therefore gives the applicant more options than a filing limited to the version currently being built.
That is why describing alternative ways to implement an invention at the outset can be valuable. An inventor may already know how a variation would work even though they have not built it. Including that detail can support a later application seeking protection for the variation.
The limit concerns whether the disclosure supports the protection sought. An inventor may understand how a variation would work without having built it. Simply naming a desired capability, however, does not necessarily explain how to achieve it. Claims that depend on a later breakthrough cannot rely on the earlier filing date merely because the original application mentioned the goal.
For investors, the inclusion of untested variations is not itself a red flag. The useful questions are whether the disclosure supports those variations and whether the company clearly distinguishes future possibilities from capabilities it has already demonstrated.
What Does an Issued Patent Tell Us?
An issued US patent gives its owner a legal right to prevent others from making, using or selling the claimed invention in the US, subject to the patent’s validity and enforceability. It means the patent office allowed particular claims following examination. It does not give the owner an unrestricted right to commercialize; other patents or regulatory requirements may still stand in the way.
The significance of a patent depends on what its issued claims cover. For example, a patent protecting a sample-handling mechanism would not, on its own, substantiate a company’s claim that its complete analyzer performs hundreds of accurate tests. This is an illustration of the distinction, rather than a conclusion about a specific Theranos patent.
What Should Investors Ask?
Investors and acquirers can use patent filings to examine how the company’s protection relates to the capabilities driving its valuation. Useful questions include:
- Which applications are still pending, and which patents have been granted?
- What do the issued claims protect, and how important is that protection to the business?
- Which product capabilities have been independently demonstrated, and under what conditions?
- What essential development work remains, and which functions rely on third-party technology?
Patent professionals can help interpret the protection and assess its relevance to the business. Qualified technical experts can examine the evidence behind the product’s capabilities. Those assessments address different questions, and investors need to understand both.
Fielder’s “can versus could” question points to the gap investors need to examine. This helps explain how Theranos could seek and obtain patents while its technology remained under development, without establishing whether any particular patent was adequately supported. Filing ahead of product development is a legitimate practice, which is why investors need to read filings critically for what they establish. Confidence in the product still needs evidence that it can do what the company says it can.
If you are preparing for an investment or acquisition, Stratford can help you assess the scope and business relevance of an IP portfolio and identify questions that need closer examination.
* Former Theranos employee and whistleblower Tyler Shultz has written that the device shown in the film was a later, unfinished version, rather than the Edison machine used to test patient samples. That distinction may affect how viewers interpret the scene. Here, we use the exchange to explore how language conveys the difference between demonstrated capability and future potential, and what that means when reading patent filings. We draw no conclusions about Theranos’ earlier technology from the device shown. Read Shultz’s perspective on the film here.