Alex Langshur, CEO at Incubeta US, explores the false promise of savings from AI.
With the major platforms loudly promoting their black box solutions, such as Google’s Performance Max (Pmax) and Meta’s Advantage+, alongside a slew of new companies bringing low-cost AI-based tools (for video and content production, measurement, modelling, analysis and workflow), you can see why advertisers might think that there are cost-savings to be had from their marketing agencies. The rationale seems fairly straightforward: these tools promise automation, significant time savings and scalability, all of which should reduce the effort involved in managing marketing campaigns. But, in my opinion, this is the wrong conclusion to draw.
In addition, as the cost of AI-based marketing & workflow tools continues to fall, it would be reasonable to expect that they become more accessible to price-conscious advertisers. As a result their adoption should spread, which would in theory level the playing field for brands of all sizes. Cost savings, speed and scalability are compelling capabilities to any marketer, so it’s understandable that there is tremendous interest in leveraging these tools to drive a competitive advantage.
But there’s a catch. The majority of a marketer’s toolbox is run on data, and high-quality data has become increasingly difficult to obtain. The data obstacles facing advertisers are fierce: 3rd-party cookie deprecation, the use of ad blockers and VPNs (particularly with younger demographics), and multiple, often colliding, country-level data privacy regulations. There is also the constant and very self-serving upping of the privacy ante by Apple (e.g. hide my email; link tracking and fingerprinting protection) to contend with.
If that wasn’t enough, the ecosystem fragmentation, of which the proliferation of retail media networks is a prime example, has created an additional layer of measurement and data extraction complexity.
The net result is that we have lost a significant amount of the signals that we’ve depended on in the past to build our audiences, track our media spend and evaluate the impact of our marketing efforts.
And we’re at a place where complexity in data collection, data management, performance measurement and channel management has never been higher. While this complexity may be viewed as an indication of the innovation and vibrancy of the sector, it can also be seen as the chaotic outcome of a sector in the throes of fundamental change.
In the face of so much change and such a complex mediascape, where should advertisers focus their efforts and how can they trust that their marketing budgets are being effectively invested? It’s always been my belief that we should optimise maximums as opposed to minimums, so what follows are the top five recommendations that we share with clients as they consider which investments to make that will generate returns over the long-term.
Make Pmax and Advantage+ work for you, rather than the platform:
Pmax and Advantage+ are powerful assistive tools for advertisers, and our testing has shown that, when deployed in the right way, they can help businesses enhance performance. However, what’s often overlooked is that these algorithms work in the same way regardless of the advertiser, which is problematic when two or more advertisers are seeking to appeal to the same client.
For example, when two market competitors (e.g. Widget-r-us and We-r-widgets) deploy Pmax across their campaigns, it’s essentially the same algorithm, running the same logic for optimisation. When viewed in this way, you can see how this can lead to increasing CPMs as each company tries to get Pmax to work better for them (cost being an important input). Our challenge is to tune these algorithms using different and proprietary signals to make them work harder and better for us, over our competitor, such as by using customer lifetime value as an early input signal. This is hard, complex, time-consuming, un-sexy work, but it’s absolutely critical if we want to succeed in a signal-poor, algorithmic-driven world.
The “all you need is first party data” half-truth:
For years we’ve been told that leveraging first party data is the key to success. The unfortunate truth is that for 95% of brands, their store of this valuable data represents just a tiny fraction (often less than <5%) of their overall market. So even if they knew everything about their customers, it’s a heavy assumption to make that this is representative of the entire category and overreliance on this should be treated with caution.
The case for data enrichment and/or access to third party consented data sources and ID graphs remains pertinent and must be a consideration within the marketers toolbox. And when it comes to ID (and device) graphs, scale is really all that matters.
Creative is the best top of funnel signal
Given that our first-party data is too small to fully represent our entire addressable market, we need another mechanism to turn heads, catch attention and expose potential clients to the brand. This is the role that good creative has always excelled at. Strong creative, built on the human truths that resonate with your audience, and convinces them to visit your owned digital assets, has never been more important, particularly in our ‘noisy’, distraction-filled and media-fueled world.
Once a visitor engages across any of your digital assets, you have the ability to start measuring and evaluating their path to purchase. Good creative is therefore a top-of-funnel signal, and this is incredibly important in helping tune the algorithms to work better for you versus your competitor. It also enables you to start filling in any gaps in your first party data stores.
To be able to use creative as a top-of-funnel signal you have to have diversity in your messaging. If you take a one size fits all approach to creative you can’t learn much, but if you create variants of your creative you can start to understand what resonates with different audiences.
Leverage bottom-of-funnel engagement:
With brands losing signal left and right, they need to maximise all of the high quality signals that they can. Fortunately, logged-in and bottom-of-funnel experiences offer one of the most valuable and informative sources of consumer data. Robust test-and-learn programs looking deeply at all aspects of EXperience Optimisation (EXO) across all assets & channels, as well as more granular Conversion Rate Optimization (CRO) are one of the most valuable sources of high-quality, high-intent signals available to advertisers. These signals are critical sources of input to the platform algorithms and we strongly recommend all clients to invest in EXO/CRO initiatives.
Value customer data like a balance sheet asset:
I look forward to the day that CFOs in all businesses start to include corporate first-party data as a tangible business asset on their balance sheet. Revenue comes from customers, and the ability to understand, segment, create and activate audiences (to include existing and new customers) based on first party data is a critical 21st century business function. To me, this data set is on the same level of organisational and balance sheet importance as physical assets or intellectual property.
Therefore, it needs to be managed as an asset where the necessary resources are allocated to manage and maximise its value to the business. It’s my opinion that the easiest and quickest way to determine whether an organisation properly values this asset is if it has a Customer Data Platform (CDP) and/or some form of customer data lake. The investment in building a CDP and training staff on its use and maintenance demonstrates a deeper level of understanding the critical and intrinsic properties of proper client data management.
With the effectiveness of marketing campaigns relying much more on strong creative, strategically wielding AI algorithms (rather than simply letting them take charge), and bottom of the funnel experiences, these five key focus areas are far more applicable and offer a real solution to improving return on investment (ROI) – as opposed to simply hoping that AI tools will solve the issue through automation and time-savings alone.
It’s my belief that these focus areas should be front and centre in the minds of every marketer and advertiser. They will maximise ROI in areas such as martech, training, and marketing, placing the advertiser in a strategic position to effectively exploit new technologies and tools, including those backed by AI, as the industry continues to evolve.
By embracing creativity and a human-centric approach, the smart utilisation of first-party data, and focusing on bottom-of-funnel engagement, we can gain invaluable insights into consumer behaviour and campaign success – allowing future strategies to be tailored more effectively. The key lies in recognising and treating our data as a critical business asset, innovatively leveraging technology, and maintaining a deep understanding of our audiences.
