10 Barriers Preventing Marketing Adoption of AI
10 Barriers Preventing Marketing Adoption of AI
Inflated promises about generative AI have become so inflated that the reality of using it with marketing operations has been disappointing…
Inflated promises about generative AI have become so inflated that the reality of using it with marketing operations has been disappointing and frustrating. What’s causing this crisis? Last week’s No Brainer podcast discussed 10 common barriers marketing departments exeprience with their AI projects.
Overblown expectations are ushering in a period of adoption that Gartner calls the Trough of Disillusionment in its Hype Cycle Methodology. Gartner summarizes it: “Interest wanes as experiments and implementations fail to deliver. Producers of the technology shake out or fail. Investments continue only if the surviving providers improve their products to the satisfaction of early adopters.”
The generative AI hype cycle has dominated the technology news sector for nine months now. Epic promises of LLMs changing the very nature of work — led by Open AI’s GPT PR machine — have tantalized and terrified at the same time.
But as noted, the hype has not met reality. Enterprise marketers struggle to incorporate generative AI and the greater suite of AI technologies into their marketing efforts. To succeed, frank conversations about barriers and how to resolve them must occur.
The Next Stage — Resolving the Barriers
Vendor hype leads to more than big obstacles that prevent marketers from finding success with their AI initiatives. Problems can happen on a smaller level when experimentation with consumer-based solutions fails to meet promised use cases or meaningful functionality for a business. Disillusionment rises, and doubt about the AI revolution takes root.
We should expect maximum hype to be met with more frequent public skepticism and even scorn. The media might even start doing its job and offering conjecture to news releases.
In short, the trough of disillusionment will accelerate as tech start-ups and giants issue more overblown hype about AI. I expect that generative AI leader Open AI will become a focal point for blistering criticism.
The technology industry must address the barriers to adoption with meaningful solutions to get through the trough, from more thought-out corporate-grade technology to change management and education. That includes the 10 sales and marketing barriers Greg and I discussed.
Find the 10 points with a summary of each below. These summaries illustrate the barriers and can serve as starting points for those seeking answers. Future articles will expand on them individually and offer in-depth solutions to help overcome the obstacles.
Summarizing the 10 Barriers
1. Tactical Thinking (Shiny Object Syndrome)
Let’s be clear: Most barriers are human in nature. For marketers, that begins with a blind acceptance of media-driven hype that leads them to implement generative AI point solutions without intent. In the social media era, we called this shiny object syndrome. This led to Facebook-specific “strategies” instead of using Facebook marketing to serve larger corporate branding and lead generation efforts.
Today, shiny object syndrome blinds marketers who fail to tie AI to core marketing or business goals. These marketing attempts at generative AI hurt adoption by turning a blind eye to enterprise goals. Why invest in enterprise adoption of AI when it fails to deliver ROI, lead generation, and/or positive branding impacts?
Instead, outputs can be enticing with questionable results, including cute illustrations and hallucinated answers that incur risks from employee-driven data exfiltration through intellectual property infringement.
Questionable results in these instances do not result from a lack of AI capability, though many AI vendors selling individual licensed “solutions” have some blame to bear. The corporate marketer needs to better vet solutions, identify use cases that benefit the enterprise, resource them appropriately, and execute them.
Simply said, but not easy to do. Overcoming tactical thinking requires leadership.
2. Leadership Deficit
A lack of leadership vision, commitment, and support to implement AI from executives stymies adoption. From responsibility hand-offs to a failure to sponsor AI as a necessary technological evolution at all levels, CMOs tell their employees to experiment and adapt AI point solutions, then wonder why they do not see impacts.
This particular issue is grounded in a surprising lack of understanding in the C-suite about AI, much less generative AI. But then again, when you read Harvard Business Review’s recent piss-poor coverage of generative AI (I am still shaking my head at July’s cover story), executive misunderstanding begins with the media.
In the marketing suite, CMOs and executives must stop dictating clueless AI mandates and expecting magic to happen. Start rolling up your sleeves and get educated. Seek out examples of companies who have implemented and demand to see metrics.
It’s time to learn the technologies and use them to support core initiatives that will impact the bottom line. Most importantly, stay involved in the project with active sponsorship and strategic direction.
3. Employee Resistance
Well-discussed fears of widespread job loss have had their impact, with internal mid-level managers openly and passively resisting adoption. I have personally seen both. Journalists and pundits fuel this with conversations about AI’s existential and economic impacts. Of course, generative AI impacts journalists and content creators directly, so it’s understandable.
There’s not much to do about the media coverage other than to question sensational stories until further research. However, executives can proactively encourage employees to embrace AI as copilots that eliminate rote tasks and free them to focus on more creative strategic works.
Change always sparks resistance, spawning an entire communications discipline, change management. Enterprises often need to catch up on change management, but in the case of AI, failing to get in front of employee resistance will make success even harder.
Leadership goes beyond staying involved. Start by finding early adopters on teams and empower them to lead on the line. Highlight their successes at every opportunity. Pair team leadership with solid programs that train staff members on AI. Build communication programs that illustrate the personal benefits of using AI at work.
4. General Market Confusion
As referenced in the introduction and the above section, vendor-driven hype — amplified by weak journalism on all levels– has created incredible expectations, existential fears, and widespread confusion about AI. Marketing leaders are confused by nonstop hype and false promises.
Regarding hype, some vendors advocate for all-encompassing point solutions, while others emphasize intelligent integration within the martech stack. Leading the pack is Open AI, whose marketing strategy is led by dominating the media cycle with nonstop media stories and releases. They are followed by many Silicon Valley unicorns and every tech player who slaps a GPT integration into their product and claims they have an AI product.
The faster vendors selling to enterprises start building solutions that address real marketing problems, including working within existing marketing tech stacks. Unfortunately, many Silicon Valley startups will have to feel pain before they pivot or fail.
5. The Data Problem
Good AI works off of strong data to inform AI models comprised of one or multiple algorithms. Everyone who has spent time in the AI field knows that garbage in equals garbage out. Once marketing organizations decide to engage in AI, they struggle with contact, content, and customer behavior data quality, quantity, access, etc.
They face the cost of getting their data house in order. As my friend Jennifer Ives, CEO of Watering Hole, likes to say, “Your AI problem is a data problem.”
Unfortunately, cleaning up data is one of the costliest aspects of AI projects, from customized to bespoke solutions. Many companies need to improve the high price of cleaning their data by investing in staff, creating data governance policies, or investing in external vendors to do it for them.
Of course, some AI solutions focus on data cleansing, too. But that requires an investment in a data science team that can execute it.
6. Expense
After getting sold cost-effective point solutions for individual creators, marketing executives are shocked when confronted with the high cost of enterprise-grade AI solutions. These efforts often require procurement, inclusion of the CTO, CISO, and CFO teams, and larger executive approvals. Further, these solutions need to integrate better with the martech stack.
Given these challenges, many CMOs make their case internally and plan for procurement through normal budgeting processes. To do that, they need to justify their investment with pilots that promise tangible results and back them with examples of other enterprises that succeeded.
And that’s precisely the issue: After resolving their confusion and faced with the reality of AI costing, CMOs need to justify it. Money doesn’t grow on trees. Funding can be quite difficult in the trough of disillusionment with lots of hype and debatable outcomes. Companies are struggling to deploy AI due to high costs and confusion.
7. Quality Vendors/Tech Partners
The complex AI landscape makes it challenging to differentiate genuine capabilities and select the right vendor among numerous alternatives, including point solutions versus existing cloud services. The point solution problem has been discussed.
On the other side of the equation, established martech vendors — CRMs, data vendors, and creative suite solutions — have failed to make AI a core element of their DNA. Despite promised integrations and road maps, that failure has created additional resourcing challenges.
While hype would have you believe that every solution is a snappy fix, enterprises usually need at least some custom integration to make their preferred AI model work. To implement real AI, you still need strong data science teams. That requires vendors or staff with capable strategic and tactical professional services capabilities.
Companies must put vendors through the paces to meet their AI needs. Of course, this does dovetail directly back to point 6 and expense.
8. Talent Deficit
The significant talent deficit in AI is well documented, with organizations needing help to hire and afford data scientists. It’s so bad that Netflix is advertising for a $900,000 AI product manager position. Welcome to the AI talent deficit conundrum the IT sector has been embroiled in for the past five years.
Left to develop talent internally, leaders often do not invest in the necessary training for their teams to use AI tools effectively and creatively. Instead, they prefer to find vendors or attempt to hire. It’s a Catch-22 situation if there ever was one.
9. Governance and Workflows
Marketing leaders often overlook governance and policy in their organizations, neglecting implications for risks. Undefined loose culture can really hurt AI implementations.
In fact, this lack of technical hygiene is one of the biggest reasons marketing loses control of their stack to the CTO or IT operations departments. Just consider the data issue and how poorly marketing departments managed their channels, records, and respective analytics.
As a result, marketing departments need end-user guidelines and need to consider how actions and lack of processes affect AI workflows. Marketing leadership either needs to seize the initiative or have other departments guide the tactical execution of the AI effort, including governance and workflows are essential. Data hygiene, usage guidelines, use case vetting, and work processes are all part of running an effective marketing organization in the AI arena.
10. Lack of Named Case Studies
The AI industry needs to offer named case studies that provide tangible, measurable results. There needs to be actual results. When they are missing, it raises concerns about transparency and credibility when marketers consider building their case to procure AI solutions.
Some lack of public example lies in companies wanting to protect their strategic advantage. Most of it lies in vendor marketing hype and promises and a lack of pragmatic solutions for the enterprise.
The good news is that many companies will publicly back brands that do a great job. It just takes cultivation within the auspices of a strong case study program.
So those are the 10 barriers. What do you think is preventing the widespread adoption of AI?
This story is published on Generative AI. Connect with us on LinkedIn to get the latest AI stories and insights right in your feed. Let’s shape the future of AI together!
