Master1305/Getty Images
- A new McKinsey report finds that companies struggle to measure the impact of their marketing technology (martech) stacks.
- Businesses are forecast to spend $215 billion on martech by 2027, per McKinsey.
- McKinsey said AI agents can help fill the return on investment gaps.
Global businesses are forecast to spend $160 billion on marketing technology — or martech — this year, and roughly $215 billion by 2027, McKinsey estimates. So it must be helping these companies make boatloads of money, right?
No one really knows.
McKinsey recently surveyed 233 senior global marketing and technology leaders who each invest more than $500,000 annually in martech and adtech tools and services. As part of the study, published on Tuesday, the consultancy conducted in-depth interviews with 50 senior Fortune 500 marketers. There was a clear trend: Not one could clearly articulate how they were quantifying the return on investment of that spending, McKinsey said.
Regardless, more than a quarter of the marketing decision-makers McKinsey interviewed expect their martech spending to increase by up to 25% in the next three to five years, potentially exacerbating the return on investment black hole.
One of the core issues, the report’s authors said, is that all the investment in this tech over the years — from email campaign management software to website personalization and marketing analytics tools — means companies’ martech stacks have become bloated and the tools often operate in silos. In McKinsey’s survey, 47% of martech leaders said “stack complexity” and system and data integration challenges were preventing them from realizing the value of these tools.
Often, when trying to quantify the value of their martech investments, businesses are measuring the wrong things, McKinsey found.
Some marketing teams were simply measuring metrics like email sends and open rates, or impressions delivered, rather than trying to tie these numbers to the strategic business outcomes, such as incremental revenue growth or customer lifetime value. And many marketers were only factoring in the cost of license and subscription fees, and not the broader investments that go into integrating and maintaining those tools.
McKinsey
As a result, many C-suite leaders see spending on martech as a “cost of doing business” rather than a growth engine — meaning it often lacks strong executive sponsorship, McKinsey said in the report.
“The C-suite underestimates what’s really required to implement this and get value out of it — it’s not just hey, I write a check,” Robert Tas, a partner at McKinsey and coauthor of the report, told Business Insider in an interview.
Tas likened the martech dilemma to people buying new workout equipment, but not alternating their exercises to train different muscle groups. Similarly, companies need to invest in continuous training and integration of martech tools across the business. About one-third (34%) of martech buyers and decision-makers surveyed cited under-skilled talent as a hurdle preventing their companies from fully unlocking the value of their martech stacks.
“Most people end up buying this expensive tool and then they use 10 to 15% of its capability,” Tas said. “It’s like buying a car without snow tires and not driving it in the winter because you didn’t buy the right things for it.”
Martech: The AI agents will see you now
There’s hope — and it lies in artificial intelligence.
The McKinsey report gives an example of how marketing teams could use AI “orchestration” agents to autonomously manage tasks like collecting, cleansing, and integrating data. A “design agent” could generate personalized offers and other messaging. Another duo of agents could test and manage which marketing channels and media would work best. All of these agents would be managed by a “governing layer” to oversee the entire martech stack, McKinsey suggested.
The jury’s still out on the effectiveness of AI agents at their present level of sophistication.
The martech and customer-relationship-management company Salesforce is going all in on AI agents with its Agentforce platform, though its CEO, Mark Benioff, said this month that AI innovation is “far exceeding” client adoption.
While some companies, such as Virgin Voyages, have deployed AI agents to slash agency costs and boost marketing production speeds, others have found them to be less effective than they expected. Take the fintech firm, Klarna. It made early waves in citing how AI had helped it halve the size of its marketing team and replace workers in its customer support division. Klarna later reassigned some employees to customer service roles after its CEO acknowledged earlier cost-cutting had gone too far.
OpenAI’s cofounder, Andrej Karpathy, recently said in a podcast interview that it’ll take about a decade for AI agents to work through their cognitive issues before they can be functional. “They don’t have enough intelligence, they’re not multimodal enough, they can’t do computer use, and all this stuff,” Karpathy said.
Tas said that, when used correctly, AI agents can help companies get their martech stacks in order, eliminating duplicative tools and combining different data sets across various divisions in the business. Connecting the dots like this can be tricky when years of human governance and process get in the way.
“This is our opportunity to take all the sacred cows and challenge them and say, ‘Hey, it doesn’t have to be this way,'” Tas said.
