From Discreet Actions to AI, The Journey of Anti-Counterfeit Strategies

Anti-counterfeit strategy is a relatively new area of business; 10 – 20 years for the majority of active companies and a few decades old for a few of the earliest entrants. There are even now very large businesses that have yet to start tackling counterfeits. 

The companies involved in the fight are typically international, with numerous staff in many countries involved in tackling counterfeits with an array of third-party suppliers across the globe, including law firms and investigators to assist them. Sometimes dramatic raids take place in the early hours with heavily armed police, and large amounts of fake goods are seized.

To date, anti-counterfeit efforts have typically been conducted discretely with a focus on efficient and targeted removal of bad actors online and offline, like an immune system fighting pathogens. 

It feels for experienced, battle-hardened brands that it is the end of a long first campaign with some introspection of what is proving effective. Lessons have been learned from numerous civil and criminal cases, working with law enforcement to stop key counterfeit sellers and manufacturers, shutting down factories, and seizing fake goods.

Here are a few key points that we have learned;

Anti-counterfeit programs have to deal with a high level of complexity in terms of geography, activity, and protagonists.

Bad actors are usually replaced quite quickly. Online sellers which are shut down can reappear in a different form. Even factories that are forced to close can remerge in a different location.

Fake goods successfully found can, unfortunately, end up back in the supply chain unless destroyed.

It is difficult for many brands to have cases taken up by law enforcement in many countries, given limited resources and higher priorities.

There is no suggestion of an end in sight.

Whilst these lessons indicate the difficulty of fighting counterfeits, this is not to say the work has been in vain. The efforts over the years have led to significant disruption and made life much more difficult for counterfeiters. They have had to come up with increasingly complex methods of avoiding anti-counterfeit measures, reducing their profitability, and deterring other opportunistic entrants.

However, there is a problem anti-counterfeit professionals face within their own companies – the benefits are often difficult to accurately quantify financially. When the wider business looks in on anti-counterfeit metrics, they can be skeptical of what they are getting for the spend.

The much larger traditional core business functions of operations, sales, marketing, and finance is where the budget for tackling counterfeits is generated and allocated. As such, the return on investment is watched closely, and when it is not always clear to assign a value to successful outcomes, the program spend can be questioned and be at risk of reduction. This is despite the wider problem of counterfeits being on the increase. 

This is providing a driver of change in how many anti-counterfeit programs operate. There are some companies that, in the face of this pressure, see no option to reduce the spend on anti-counterfeit activity. This is far from ideal but is justified in the face of unclear pay-offs and budgets, in general, being routinely scrutinized and under pressure. Time, it seems, is running out on some programs to prove a clear benefit to the wider business.

However, in other companies, this has been avoided by commercial pragmatism. Assumption-reliant ROI models are not required when there are tangible cashflows generated by anti-counterfeit measures.

Rather than acting independently of the core business, these successful anti-counterfeit programs are increasingly needing to mesh into other parts of the business but particularly the sales and marketing functions. It can be a symbiotic relationship with the sales function. The anti-counterfeit program focuses on identifying significant lost sales that can ultimately be then readily directed into legitimate sales channels. In return, the anti-counterfeit benefit can be much more clearly measured in an uplift in genuine product sales. Rather than a cost activity, it becomes an additional approach for ultimately increasing company sales.

Similarly with marketing, for those companies to openly declare war on counterfeits, it is a distinct opportunity to grab the attention of customers. In its most effective guise, this is to convey the unique value features of the genuine product over inferior fakes. 

This leads to another trend in anti-counterfeit programs – for them to come out of the shadows and be much more visible to key stakeholders. Historically most, but not all, brands were reluctant to talk openly about fake versions of their products and aimed to tackle the problem away from the wider view. This is changing. As an economic problem with high margins, some brand owners realize that focusing all their effort on criminal counterfeit production and supply is not going to provide the results they seek.  They must also look to tackle the problem from the demand side. 

As such, the engagement of stakeholders is now at the vanguard of anti-counterfeit programs. It is a natural progression to enhance and provide cost efficiencies to the existing areas and can be as discreet or as public as a brand sees appropriate. Stakeholder engagement itself has long been a part of anti-counterfeit, including mobilizing employees and external law enforcement such as customs. However, what is now growing is a willingness to engage with customers and distributors to raise awareness and mobilize a much broader combined front.

Developments in Artificial Intelligence (AI) for Anti-counterfeit

Another problem brand owners have is the growth in data they capture from the Internet with regard to suspicious activity. This is difficult to process quickly in order to prioritize targeted, evidence-based investigations. 

Artificial intelligence is likely to make significant progress in this area in the coming years. A decade ago, it was common for counterfeit products offered online to show clear indicators such as spelling errors in photographs. These types of defects are far less common, and online sellers tend to now use stock images in any case. It means that a range of other indicators has been sought out instead to identify high-risk sellers. These indicators lend themselves very well to being processed by AI tools. 

Whilst Amazon’s efforts to reduce fake offers to date have been frustrating, it is likely they are the front runner in using machine learning to identify fake sellers, along with the Alibaba Group. The volume of proprietary data they have to learn from is vast, and third-party technology providers cannot replicate their speed of learning and increasing effectiveness. Numerous other online platforms are no doubt progressing as well. Eventually, these AI tools may be made available for use outside the platforms they are developed in. 

Within the next three to five years, we should start to see a dramatic impact of AI on the availability of fake products being sold on major shopping and social media platforms. We should also see much greater efficiencies in identifying high-value targets that justify determined enforcement. However, brand owners will probably have to accept a lack of control or access to the most effective AI tools for some time and be heavily reliant on big data-rich companies. 

There will be numerous new AI tools from smaller technology companies that will try to mirror this learning. However, their progress will be limited by the volume of data they can gain access to. Some AI providers will offer to access a brand owner’s client’s data. However, we are already seeing many companies, such as Apple banning company use of ChatGBT because of the risk of it dispersing trade secrets and other valuable company data. The notion that a brand could give access to an AI tool that then learns from the data and takes that out into the wider market should pose too great a risk to be viable for many. 

In the near future, AI will at least clear the landscape of a great deal of counterfeit noise, automate and improve lower-value activities allowing anti-counterfeit professionals to focus their time on the highest-value processes. However, AI alone will not win the war against counterfeits.