Disclaimer: This interview is for informational purposes only and should not be relied upon as a basis for investment decisions. In Practise is an independent publisher and all opinions expressed by guests are solely their own opinions and do not reflect the opinion of In Practise.

{audio:0:04:18} Could you discuss your time at Digimarc?

{audio:0:04:28} My first contact with Digimarc was in 2005, which was when I started a company in Spain focused on digital watermarking, and they owned all the intellectual property there. I told them what I wanted to do and they decided to become shareholders of my company. Several venture capital firms joined and it survived for 10 years before we liquidated it. Some projects were acquired by one of the other founders of the company, and I was invited to kick start and run Digimarc's European operations. I moved to Germany at the end of 2015 to open an office in Cologne and hired the local team and triggered deals that are now operational. The relationship with GS1, in Germany, was a roll out of their existing US relationship. You understand the nature and importance of this relationship to engage on this side of the pond. I was a Digimarc employee until early 2018, then returned to Spain due to personal reasons. 

{audio:0:07:04} Since leaving the company, I have been in touch with stakeholders of the HolyGrail project, because that was one of the projects started in Spain from my old company, aquaMobile, and was in the framework of a different program, called REFLEX, which involved applying digital watermarks to flexible packaging to improve the rate of recycling at waste classification plants. During my time at Digimarc, I was adamant about going after that use case instead of focusing on what had been done in the States, which was the scanning rate at point of sale, applying watermarks to packages to improve the speed of scanning. In Europe it was not as compelling an advantage as the sustainability use case. I have been kept in the loop and have introduced Digimarc to some interesting people for rolling out the technology in other countries, only because of the good terms I kept with Digimarc executives.

{audio:0:09:39} That gives us many avenues to ask more questions. What was it like working for Bruce?

{audio:0:09:52} I did not report to Bruce as an employee; I reported to the VP of global sales, and there were different people in that position while I was there. I had known Bruce for many years and he was supportive throughout the life of aquaMobile, because we designed, developed and deployed commercial applications that had not been launched elsewhere. If you look at old Digimarc videos demonstrating mobile scanning technology, they use an app called CLIC2C, which was our app. Bruce was particularly supportive of everything we did up until we decided we would no longer fund the company. When I joined as an employee, I had contact with him occasionally and if I had to point out something I missed, it would be granting sufficient importance to the HolyGrail project earlier on. I pushed hard on that but did not find the right echo at Digimarc to go further or faster. He did change his mind once I left and they decided to heavily invest on it. Digimarc's vision has changed from five years ago, which is due to the great importance they are now giving this use case. Historically, Bruce was very good at creating value for shareholders, and companies who consistently lose money are seldom able to do that. They had both the IP and had identified some interesting use cases. I do not know Riley, nor do I have any references, but obviously he still needs to improve a lot.

{audio:0:14:52} What do you think of their current commercial implementation strategy?

{audio:0:15:08} From reading the transcript of their latest investor call, Riley announced they would discontinue their book piracy business, and I don't know how they're going to do that. They made that decision as they don't see great potential and the margins were very low. It continues their strategic mandate on focusing on use cases that have bigger growth potential. They are aligning resources to efficiently fuel HolyGrail and some other related use cases. Acquiring EVRYTHING shows they want to integrate vertically, because they have not been historically strong in back-end data management, which will be a critical piece in future. 

{audio:0:17:36} They will continue to place most of the attention on initiatives such as HolyGrail, supporting the packaging use case, in addition to their traditional business of bank notes with central banks. That is almost a separate company because the team working on the bank program has no contact with the rest of the company due to security procedures and confidentiality, but that business is still alive and generates a stable revenue stream for them which will continue.

{audio:0:18:47} Could Digimarc spin off or sell their currency business?

{audio:0:19:04} I wouldn't be surprised as they sold their ID business 15 years ago, but I am unsure of the limitations or who they would sell to, because the central banks, the CDs group that runs these security systems is very strict in how they incorporate new people, so it's not as easy as selling the other ID business. I am curious to see what they will do with the book piracy business, whether they will simply let it die or try to sell it, as I haven't read anything about it.

{audio:0:21:06} How did Digimarc replace FiliGrade in the HolyGrail initiative?

{audio:0:21:23} FiliGrade was a small company who were linked to HolyGrail from the beginning. They were invited to some workshops and claimed to own IP around laser edging of digital watermarks on plastics, but back then I did not agree they had a claim. Digimarc has 1,200 patents on watermarking, some of which include laser edging as a way to embed watermarks onto different physical materials. That was the elephant in the room when we collaborated in the HolyGrail project. We had to deal with that but never found the right moment. 

{audio:0:23:08} Back then, FiliGrade consisted of three people and had limited capacity to invest in the R&D required to make the technology work. There are very demanding prerequisites of the ecosystem, especially the detection process at waste sorting facilities where waste runs on conveyor belts at 3 m/s. To process an image and detect watermarks at that speed, while also making the watermark compatible with Digimarc's other packaging, was challenging. FiliGrade could not do that, so there was an arrangement to cooperate in the framework of the HolyGrail program. I was in charge of the relationship with FiliGrade as they were a Dutch company, and they became a licensee of Digimarc, to use our watermarks for both those tests and other use cases they had. Their customers wanted to use the technology for the protection of furniture designs. I don't know what the current commercial relationship is as I have no visibility on their current status.

{audio:0:25:50} How close are you to the current progress within HolyGrail?

{audio:0:25:56} I attended some open house events in Germany but missed the one in Denmark. Back in 2014, when I was still running my company, we also did some flexible packaging tests in Germany with TOMRA in a project called REFLEX, so I have been able to see the evolution as I receive the newsletters and bulletins they publish, which is all public. The German, Danish and French projects are advancing but remain in the pre-industrial phase. It still needs calibration and some tweaking, but the partial results we have seen are very promising. 

{audio:0:27:53} They have produced an 80% detection rate, which shows they are meeting the mark, so I am confident this will prove to be the best possible solution. Digital watermarks were compared to other technologies such as chemical markers, and it was obvious that detecting watermarks at speed, among many contaminated materials, would be more efficient than other solutions. So far that remains true, which is a sign for wide adoption.

{audio:0:29:39} You mentioned tweaking and calibration was still needed; did Copenhagen show this technology was ready or is more required for it to be truly deployable?

{audio:0:30:02} It is more about volume. The tests have been done under controlled conditions. They are trying to emulate what happens in the real world but they need more items on the conveyor belt, in addition to more dirt or contamination to make it hard for cameras to detect. That will encourage many other entities, not only in Europe but also abroad, to adopt it as a standard in their respective countries. That needs to be demonstrated and more stressful tests will happen after the summer, which will also be the moment of truth for the technology. 

{audio:0:32:07} With the new sorting equipment from Pellenc – the French manufacturer, which is relatively important in the European market, if both TOMRA and Pellenc successfully demonstrate the technology works in stressful conditions, it will be rolled out throughout the continent. The social economic context in Europe and the US is forcing CPGs to adopt these technologies, so money will be found to encourage solutions for improving recyclability. Digital watermarking is not as invasive or costly as other technologies, so it can scale well. Once the equipment is ready and waste sorting machines are prepared, all the pieces of the puzzle are there to make it work and roll it out to any country. 

{audio:0:34:05} The US has Federal programs but it is harder to run a pan-European program because national entities are usually in charge of the packaging recycling programs, and we have several GS1 organizations. Digimarc requires the cooperation of pre-press agencies who work on package design with the CPGs to embed the watermark. That process requires people to be sufficiently trained to avoid any quality costs. You do not want to embed a watermark and launch a product to market which is not readable by cameras in waste sorting facilities. It is important to identify the right partners to collaborate with the CPGs to ensure quality of products is above minimum levels. It is not only the national agencies coordinating packaging or recyclability but also the people who embed the watermarks.

{audio:0:37:11} Are these more stressful tests part of phase two of the HolyGrail experiment and are you disappointed it wasn't done in the initial phase two?

{audio:0:37:38} We began by validating the possibility of reading the watermark from a picture taken on a conveyor belt. Waste sorting equipment has cameras and already rely on image recognition to classify waste. We placed items with watermarks which cameras could read, but what about fragments of a package or those which are wrinkled or have dirt or food on top? Those situations required tweaking of the camera, not only how many pictures were taken but also the computing processing power required to process the images. In addition to identifying shapes, they need to identify specific patterns in small pieces of that shape. That meant building software capable of focusing on something other than the image itself, finding distortions of pixels in that image. 

{audio:0:39:59} Adapting to that environment required re-engineering of the software Digimarc already had. The packaging world is different to bank notes, which are specifically designed for security. Packaging has a great marketing value and you need a very strong case to convince marketers to change their marketing asset, which is the packaging. There are also different types of material as well as reflective and metallic inks, which all affect the scanning process and how cameras react. Metallic ink creates a reflection which makes the watermark unreadable, so the Digimarc R&D team have to diminish that impact and find ways to embed the watermark in those materials to be less vulnerable to reflections. 

{audio:0:42:06} We tried to make the watermark as imperceptible as possible. Ultimately, it is a variation in the pixels of the image, but some packaging is completely white, so how do you make that imperceptible to humans, but machine readable? Finding the right balance between imperceptibility and readability was challenging because you need approval from CPGs. All that required significant effort from an engineering standpoint, which will continue for some time because tweaks are still needed to develop a solution which can go to production. 

{audio:0:43:53} Everybody would have wished for this project to have gone faster, and Digimarc would have liked to see strong revenues from this program. Bruce would have liked to leave, proving it was both a commercial and financial success. I am confident they will prove it in two and a half years’ time. The trend is already there and more companies are joining the HolyGrail program. Broader adoption of the technology will come once these more stressful tests are proven successful.

{audio:0:45:49} To clarify, the test coming after the summer, would that be part of phase two, or phase two part two?

{audio:0:46:05} I cannot recall how they classified it, but it was the R&D phase, the pre-industrialized phase and industrialization. The second phase of pre-industrialization will show the capacity of detecting products in more stressful conditions, which is needed to trigger wider adoption in those countries. I can't recall the number of SKUs they've done in Denmark, but Germany already has thousands of products, and so should Denmark and France. 

{audio:0:47:09} One of the agencies who showed interest in researching this, thought they should start with a minimum of 1,000 products, and engage different retailers and CPGs in the country. They should choose a city to distribute these packages in several supermarkets and retailers, then specifically focus on waste collection and analyze the flow, testing in real conditions with real products which go to real garbage cans which are collected and processed among other waste, with no intervention throughout the process to see the results. That is what they will be doing in the next stage in other places, which should be enough for the ecosystem to adopt it, especially if you can demonstrate it in one of these environments. If the technology doesn't pass in that real test, it is either re-engineered or dismissed. We will see that later this year.

{audio:0:49:56} What you described is what they call phase three or the industrial trial; will that happen later this year?

{audio:0:50:09} Phase three is wider adoption whereas this is the final stage of phase two. They already have a significant number of SKUs which will be tested in a semi-real environment. Going from that limited test in a city and rolling it out to the whole ecosystem requires cap, and the adoption time is not immediate, they will need at least two years.

{audio:0:52:01} We heard that Digimarc is struggling to get enough SKUs on board to do the city trial you suggested. Have you heard that and do you have any fears about it?

{audio:0:52:29} I was more aggressive in the beginning but I will choose my words carefully because I don't want to say something I am not allowed to. The company has changed the way it approaches this business. There was no business model designed or agreed with all program participants; it was simply Digimarc saying their watermarks are $50 each per year and you can buy it from them or GS1. To embed the watermark, you would have to pay $400 to a pre-press agency who would do that for you. 

{audio:0:54:00} In the context of a European funded research project, that model will not work. They have to bear those costs to produce results, and all participants have their own cost to participate. TOMRA and Pellenc need to adapt their equipment and somehow get paid for all those investments, which is where it becomes tricky because once the research project is complete and there's a European directive saying all CPGs need to adopt technologies that improve the recyclability rate, they can suggest digital watermarking as one of the accepted solutions, but you need to ensure the program is open. With proprietary technology you need to ensure those who embed watermark have rights to. Digimarc cannot simply give watermarks to Henkel or Danone; it has to work like a standard. 

{audio:0:55:54} Digimarc has not been very clear on how they want to get compensated for this. Initially they were very aggressive in trying to get paid for the watermark for one year, which scared people off, but now they have to find a scalable model that makes sense for CPGs, retailers who produce white label products or retail brands. Digimarc are working toward a licensed model – all you can eat – and given the variable serialized projects coming up, it is the right approach. That will be more flexible and cater to the needs of CPGs and agencies so that they can put more of their SKUs into the program. Selling digital watermarks to CPGs and asking them to experiment with technology while committing to a subscription model per SKU, was not easy. I believe that is changing in a way that will make it easier for CPGs to adopt.

{audio:0:58:31} Do you think that by this fall, there will be enough brands on board to do the trial and keep this moving to that two and a half year target to wider adoption?

{audio:0:59:04} Many companies have been working with Digimarc in stealth mode. They were testing the technology in real SKUs on shelves, but you cannot see a watermark and they never publicly announced it. There are minimal costs for a company to adopt this. We need to think about how to put the watermark on the package before the design stage. If we decide to adopt the technology, we need to understand there will be some limitations in the design. Some companies have tried the technology, but are waiting to see who else participates, because they will only adopt it if they see a significant amount of other players adopting it. They don't want to be the only ones using the technology if it doesn't fly. 

{audio:1:01:12} Most big players in Europe have some of their SKUs watermarked. In my time, we were mainly focused on point-of-sale scanning. Procter & Gamble are involved with the HolyGrail project, which stimulates other companies to consider it, and some have joined as a result. There are still many stealth testers out there, because if you see the logos of the companies who publicly say they use the technology, there are not as many as the ones who already have processes in place to adopt it. Digimarc's current customer list is misleading as it doesn't show the full picture of the efforts and positive attitude towards the technology, which could eventually convert into customers.

{audio:1:04:00} Will they get enough brands on board to run the next test?

{audio:1:04:14} I have no reason to believe they cannot do it. Digimarc has been conservative in the past, especially towards very innovative use cases. The amount of money, effort and focus I have seen, tell me that more of their stealth mode results will surface at the right moment.

{audio:1:06:02} Will Digimarc ever reach a point where they can charge on a per item basis? Perhaps they could first roll this out, then get more aggressive on price?

{audio:1:06:31} How they share revenue with participants is where I saw a positive revenue model in the past, but I don't know where they are today. I know licensing was a flat fee, which was done before for other Digimarc products, but how would TOMRA or Pellenc be compensated? I go back to the issue of how to standardize this technology and ensure it's open and the price is the same for everybody. You can have some volume discounts but it should be sold not only by Digimarc, but also by GS1. If the conclusion is they have to go back to a per SKU model, then it will be the same as it was before, that GS1, Digimarc, or another entity could sell the digital watermarks at the same price. How they compensate the other players remains a big question. 

{audio:1:09:07} I know that there was a group of people, working on this, within the HolyGrail program, thinking about the model. I don’t think they have made any public statement as to how the model will develop. I think the participation of local entities who coordinate on a national level, such as GS1 or a different entity, will need to ensure the pricing makes sense for everybody. They have the experience and credibility, and in the end GS1 is participated by all CPGs and retailers. When we tested the all-you-can-eat licenses, they were attractive to CPGs and retailers as they wouldn't have to worry how many times they changed their packaging designs. There are seasonal packaging versions of an SKU which require different watermarks, so they could pay for four different watermarks because you need to differentiate one design from the other, which would dramatically increase costs. 

{audio:1:11:30} When dealing with a single SKU and applying a cost of $50, for an SKU that produces three million units, the impact is minimal, but when a CPG produces thousands of SKUs, the overall cost will be significant. In that case a CPG could end up paying millions of dollars for applying the watermarks. If they simply charge them a flat fee of half a million and let them do anything they like, that is a much more attractive model. For the big guys who are very seasonal, that would make more sense, but for the smaller guys, I have my doubts. They have less marketing power and would be less attracted to a flat fee. The HolyGrail project participants are all big companies, who are needed to lead these initiatives, but whatever they decide economically, doesn't mean it will be fair on the smaller guys. That is why an entity like a GS1 or similar needs to be sensitive to the needs of the whole ecosystem. 

{audio:1:14:17} Thank you so much, this has been incredibly helpful. We might ask Will to setup another call with you shortly, if that's okay with you?

{audio:1:14:26} Yes, I enjoyed talking about this. I will try to collect more information to be more up to date on the stages, to provide more insights, and I am happy to speak to you again.

Digimarc & Holy Grail 2.0 Strategy

DIGIMARC CORPORATION

Digimarc is a pioneer of digital watermarking technology. Their expertise and portfolio of patents allow them to apply imperceptible watermarks to product packaging containing data-rich, serialized marks far more advanced than the capabilities of barcodes. The invisible watermarks repeat many times throughout packaging artwork, increasing the efficiency of scanning relative to traditional visual barcodes and the serialized, connected capabilities allow more data that is updatable over the internet for each individual item or SKU. Digital watermarks provide many benefits, including sustainability, traceability and brand protection. Currency authentication for a consortium of central banks is a cornerstone use case of Digmarc's watermarks, providing the company with cash flow to fund a large portion of R&D and proving to customers the security of the technology itself. The company's watermark technology is being used in the Holy Grail 2.0 consortium to innovate and test scaleable recycling use cases for digital watermarks in Europe. Digimarc recently acquired EVRYTHING which gives digital identities for items through a digital product cloud.  

Digimarc & Holy Grail 2.0 Strategy

Former Director at Digimarc

inpractise.com/articles/digimarc

DIGIMARC CORPORATION

Why is this company interesting?

Executive Bio

Former Director at Digimarc

Interview Transcript

Related Content

Digimarc: Holy Grail 2.0 Developments

Former Director at Digimarc

Digimarc Watermarking Technology

Former VP at Digimarc

Partner Interview