If ad fraud were a country, its economy would dwarf the GDP of most of the countries in the world.

Specifically, the cost of online ad fraud could spiral to $16.4bn this year, according to a study published in March and commissioned by The&Partnership and m/SIX from Adloox.

The specifics of the report are not without contention, relying on a mixture of self-interest findings and extrapolations. But the trend is clear – despite efforts to clean things up, ad fraud remains a blight on our industry.

In ad fraud, nefarious actors – let’s be honest, let’s call them criminals – game the system by exploiting technical weaknesses, setting themselves up as fake publishers and invoking false ad views to self-generate advertising spend for ads that were never viewed, except by bots.

I was recently asked to name examples of advertisers that have been hit by the crime – but, so wide is the practice, it is probably easier to spotlight those which have not. Any advertisers that have not put some controls on their systems and partners are likely exposing themselves to some degree of ad fraud right now.

Ad fraud has been around for almost as long as online advertising. I remember looking at some of the first ad exchanges in the early noughties and seeing unprincipled sites gaming the system. But the massively increased scale of online advertising itself has inflated the scale of the problem, meaning many clients end up simply giving money away to fraudsters.

All is not lost. There are plenty of actions that smart advertisers can and should take to avoid falling victim to fake views and needless bills.

What all have in common is the need to ask tougher questions of their agencies or technology intermediaries. With several initiatives now in the market to counteract this scourge, the biggest weak point arises when clients don’t hold their suppliers to account for eradicating the problem.

Insist on verified inventory

Until relatively recently, brands’ ads were free to be placed across all manner of sites, respectable and otherwise, by spray-and-pray ad exchanges. But the problem of lack of control has given rise to third-party ad verification providers.

Now companies like Integral Ad Science, Moat, DoubleVerify and AdLoox are available to run pre-bid filtering on inventory in which a purchase is planned, returning with a recommendation on the quality of the placement. Brands should insist that their partners use tools like this to filter for quality in a way that human buyers could not.

Protect viewability

Viewability isn’t a proxy for ad fraud protection. Actually, the reverse can be true… Ad fraud bots are designed to do one thing really well; to simulate ‘real’ web users. To this end, they are very capable of tricking the scripts within an advert that measure viewability into believing that an ad has been genuinely viewed.

Unsophisticated machine learning algorithms have been known to get tricked by these ‘high viewability’ sites and users, and optimise towards them. What they are actually doing is honing in on fraudulent activity. Of course, more advanced machine learning platforms will quickly dismiss this suspicious traffic, but at the expense of apparent viewability.

Good platforms don’t require a proxy – they will offer ad fraud protection directly, blacklisting known bots and implementing processes and pre-bid measures to reduce exposure to risk.

Ask the right questions

The best way to ensure you don’t fall victim is to be robust with your agency and tech partners. That means asking them tough questions.

Many clients tell me they are frustrated that they don’t know the right questions to ask. After all, agencies are good at economically answering only the specific questions asked by clients. If a client says, “I want to know about your ad fraud policy,” they are likely to receive an insufficient one-liner that leaves much to be desired.

So play hardball to get to the heart of your suppliers’ protection measures, with questions like: “Are you transparent about media placement?”, “Do you buy media with your own technology stack?” and “How do you optimise campaigns?”

Trust the coalition

Somebody has to set a high bar for practices and platforms that are best placed to strike fraudulent inventory from your ad plan. In the UK, that is now JICWEBS, the joint industry committee for web standards.

Right now, JICWEBS is auditing ad tech platforms like ours for compliance with a host of criteria in buyers’ best interest. While the body has already begun accreditation, it will take some time for it to get through the host of partners in the market. By then, brands should select their stack of suppliers from those deemed trustworthy by the committee. Until then, buyers should follow JICWEBS’ Good Practice Guide to reduce their risk of exposure to ad fraud.

Keep your eyes open

Like the dog that is for life, not just for Christmas, the fight against ad fraud never ends. While ad buyers can make smart decisions upfront to minimise the chance of exposure, they should always be aware of the potential for change.

The only way to stay across the risk and the extent of your victimhood is to constantly monitor your ad placements, to constantly keep your partners on their toes and to routinely tell them what you need.