It’s no secret that personalisation drives engagement. Today’s marketers also know that AI can power hyper-personalisation. But many are struggling to effectively deploy it, specifically for ecommerce. Learn how a unified data ecosystem can amplify marketing strategies, and get the checklist to help you align with your IT leaders.

Understanding trends in AI marketing

For as long as marketers have honed their craft, they’ve understood that the organisations that know their customers and most effectively delivers relevancy are the ones that can drive engagement, acquisition, and lifelong loyalty. In fact, research shows that a well-executed, hyper-personalisation of your marketing strategy can deliver 20x the return on investment (ROI) and lift sales above all other forms of marketing combined, SEO, CRO, social-media, cross-channel, advertising on and off-line etc. (Statistics courtesy of Forbes, McKinsey, Forrester et al).

Although marketers have relied on AI for some time now (maybe without even realising it), the generative AI revolution is creating lots of excitement, numerous questions, and some trepidation about what new use cases could mean for marketing. For many organisations, it can still be very difficult to power a customer-centric and hyper-personalised AI marketing strategy that effectively links back to and connects with its customers.

To understand why, we surveyed a diverse group of marketing and IT leaders throughout the world to:

• Identify the use cases that marketers prioritise but struggle to execute
• Understand the common challenges that marketers experience with generative AI use case execution
• Define what marketers truly mean by “personalisation” and “bringing their customer experiences to life”

See this crucial software comparison article: Distinctions between all the top hyper-personalisation software providers (Updated)

When exploring the major challenges that organizations run into when implementing or using artificial intelligence/machine learning (AI/ML) to support their marketing use cases, our research uncovered an impactful collection of barriers.

Limited data sharing an inoperable ability between systems

While many organisations collect and store first-party data (and may combine it with second- or third-party data) to use in marketing campaigns, a lack of connectivity and interoperability between the data can be problematic and limit the impact of marketing analytics or activation efforts. Although not the only culprit, siloed or independent data systems often contribute to an organisation’s lack of data unification and sharing, especially as the volume of collected data grows.

Based on our research, the top challenge organisations face when sharing data internally mirrors a top challenge associated with organisations using AI/ML for their marketing use cases: ensuring sufficient integrations or interoperability across data platforms.

There are many opportunities for marketers that are properly prepared to collect, process, and unify their data, particularly as the volume of gathered data grows. The ability to organise, access, and act on data is critical. Organisations that fail to address data management and unification requirements may find future data influxes to be more of a challenge than a chance for innovation.

Fragmented network of content channels

Consumers have seemingly endless ways to research their interests, be inspired, or make a purchase. Conversely, marketers are dealing with a larger and more fragmented network of content channels to reach and engage people. This only amplifies the challenges that organizations and marketers must navigate—especially since, as research shows, people expect to be consistently treated as individuals across all of these channels. Organisations need to adapt to consumers’ changing interests in real time and serve up relevant content no matter where they are.

Slow and complex tech adoption

The slow and oftentimes complex orchestration of technologies within organisations can hinder the development of a project before it even starts.

In recent years, marketing teams have advanced their tech fluency. Since 2015, the fastest-rising skills mentioned in chief marketing officer (CMO) job postings are “key performance indicators (KPIs)” and “cloud solutions”—two areas that fit comfortably in the world of data and technology. Additionally, our research revealed that 72% of marketers possess either primary decision-maker or executive approver authority within their organisations when it comes to choosing marketing or data technology solutions.

This shift in skill set and responsibility has moved the focus of IT more toward system integration support and away from its traditional role as the primary enterprise technology decision-maker. For a while, this worked out, but over time and with the growth of first-party data and AI, conflicting priorities and perspectives between IT and marketing teams surfaced.

On the IT side, there’s a strong will to lead enterprise data and AI projects. It makes sense: This team oversees the organisation’s data governance and security policies and often builds and trains AI/ML models. From the marketing team’s side, the desire to spearhead these data and AI projects stems from the fact that they are the ultimate consumers of these data assets (for instance, AI/ML model scores). Marketers need them to understand customers and build better marketing activation tactics, such as hyper-personalised content generation sustainability.

Ultimately, this can lead to misalignment on the core objective(s) for an AI project, which can encourage both teams to continue to work independently. This can create an environment for AI adoption projects to sputter along or fail altogether as both teams struggle to agree on important project components, such as the business case, technical and financial requirements, and evaluation criteria.

Through our research, we continue to see that today in how teams within an organisation believe first-party data is used to support an organisation’s marketing use cases.

Impact of data privacy policies

Ever-changing data privacy regulations require privacy leaders to constantly update and conform their organization’s data compliance and governance rules. Additionally, the expansive but often fragmented collection of data privacy laws propels many organizations to adopt internal policies that apply regulations, such as the California Consumer Privacy Act of 2018 (CCPA), General Data Protection Regulation (GDPR) Europe or Data Protection Act 2018 (UK), across any consumer that interacts with the organisation, regardless of that individual’s location. Additionally, heightened controls among internet browsers and hardware companies are impacting traditional data collection, and with these disruptions in gathering valuable data, marketers are facing significant disruption in how they drive business impact and measure ROI.

These changes can impact everything from campaign planning to post-campaign reporting, and they can affect the accuracy of third-party ad platforms and brand measurement systems. A marketer’s ability to target, measure, and understand its consumers is directly affected, which makes it vital that organizations align on and establish a privacy-centric approach to marketing.

This can be especially challenging for global brands that must untangle a web of regional, national, and international data privacy laws. Regardless of the complexity, owning and building a foundation of consented first-party data is crucial to AI-powered marketing. Our research indicates that high-growth companies are focused on using AI to support their marketing processes because AI is the business multiplier that can help organisations keep up with shifting consumer demands and gain necessary insights in an efficient and privacy-centric way.

As our research pointed out, many marketers are already cognisant of AI capabilities and the need to protect their customers’ data, but they may lack the necessary infrastructure, internal policies, or personnel to effectively use AI to support their marketing processes.

Finally the focus of implementation

Globally, people average 6 hours and 58 minutes of screen time per day. Daily screen time has increased by nearly 50 minutes per day since 2013. The average American spends 7 hours and 4 minutes looking at a screen each day. South Africans spend 10 hours and 46 minutes on screen per day.

But if you’re a retailer, don’t assume this is all on your website. The average is artificially high as inclusion in this is its adoption universally across business. Now also consider that shoppers you are ever likely to encounter carry there mobile phone with them everywhere, 365 x 24/7.

The not always obvious conclusion ,for those who consider their online store their primary revenue generation source, is that taking the products to the consumer is far more important than waiting for them to come to you. Translate that into marketing spend and it may need some re-adjustment.

Now add into that equation include that your AI hyper-personalisation can identify exactly when each product should be shown to each person, which you can’t do online until they come to you, and you have the maximum potential return. If nothing else remember if you don’t go to your consumers, someone else will.

See this crucial software comparison article: Distinctions between all the top hyper-personalisation software providers (Updated)