Technology and Digital Marketing: A Realistic Playbook for the Overbuilt Stack
The marketing technology market is now larger than most national economies, and most brands using it aren’t measurably better at marketing than they were a decade ago.
That’s not a hot take. It’s what the Scott Brinker Martech Landscape has shown for years. Tools have multiplied. Vendors have consolidated and unbundled and reconsolidated. Budgets have expanded. Reported marketing effectiveness has moved in a much narrower band. The problem isn’t the technology. It’s that most brands buy technology to solve problems they haven’t yet defined.
Here’s the working posture we take with clients trying to build a technology-enabled marketing practice that actually produces returns. It’s less about picking the right tools and more about which layers of the stack deserve the investment.
The Stack That Actually Matters
Most marketing technology conversations start with tools. That’s the wrong end.
The stack that matters is layered by function, not vendor. A serious modern marketing operation runs on five layers.
Data and identity. The layer that resolves who a user is across sessions, devices, and channels. Customer data platforms, identity graphs, and consent management sit here. Without a working identity layer, everything above it is guessing at attribution.
Content and asset management. Where the creative work lives, how it’s versioned, and how it flows into the channels that use it. This is often the least glamorous layer and the one that most often becomes the bottleneck when marketing wants to scale.
Activation and channels. Email, paid media, social, organic search, and the AI search layer that’s emerging alongside them. These are the tools that touch the audience directly, and they’re where most marketing spend goes.
Measurement and analytics. Attribution modeling, reporting, and business intelligence. This layer is where the work of measurement happens, and it’s often where the least experienced staff are asked to make the highest-stakes decisions.
Automation and orchestration. Workflow tools, marketing automation platforms, and increasingly, AI-powered orchestration that sits across the other layers. This is the layer that ties everything together, and the one most vulnerable to being oversold.
Each layer has its own decision framework. Confusing them, or over-investing in one while starving another, is the most common failure mode we see. Most stacks fail not because the wrong tools were selected, but because the right layers were underfunded.
Where Most Brands Overspend
The pattern is remarkably consistent across categories.
Brands overspend on activation tools. Every quarter brings a new channel, a new ad platform, a new social feature, and a new promise of untapped audience. The temptation is to buy in, because activation feels like doing marketing. It’s visible, it’s measurable in shallow ways, and it produces immediate outputs.
Brands underspend on data infrastructure. The identity layer, the consent framework, and the analytics setup are less visible, harder to demonstrate, and produce value that shows up two years later. So they get deprioritized in favor of the next campaign tool.
Brands underspend on content and asset systems. The bottleneck in most marketing organizations isn’t the ability to run campaigns. It’s the ability to produce enough good creative to fill the channels the brand is already paying for. Fixing that requires investment in content strategy and content production systems, not more media buying tools.
Brands overspend on tools that promise to replace judgment. The pitch is always compelling. The results are almost always underwhelming. AI-driven marketing tools help teams that already have judgment. They don’t help teams hoping to skip the judgment step.
Where Executive-Level Digital Marketing Actually Fits
There’s a specific version of this conversation that shows up at the executive level, and it deserves its own frame. See our take on digital marketing for executives for the detailed version. The summary is worth stating here.
Executives buying technology tend to buy for scale. That’s the right instinct at the wrong layer. The executive-appropriate technology investments are the data infrastructure, the identity layer, and the measurement framework. Everything else can be optimized by the team on the ground. The layers executives should own are the ones nobody below them can fix.
When executives get involved in tactical tool selection, the result is usually a longer sales cycle for the vendor, a more expensive contract, and a tool that solves the problem the executive was worried about six months before it got signed. That’s not a knock on executives. It’s an observation about where their attention is most valuable and where it’s least.
Ideas for Content Creation Technology Actually Helps With
There’s a legitimate role for technology in content creation, and it’s narrower than the pitch decks suggest.
Technology helps with research at scale. Analyzing search behavior, competitor content, and audience conversations across surfaces produces input signals that would be prohibitive to gather manually. The best content ideas often come from patterns the team couldn’t have seen without the data layer.
Technology helps with content production efficiency. Not with replacing the thinking, but with reducing the drag around production. Transcription. Translation. Format conversion. Tag application. Metadata generation. The team’s time gets redirected from mechanical work to editorial judgment.
Technology helps with distribution and repurposing. A single piece of long-form content can become a landing page, a series of social posts, a newsletter, and a video script. Doing that manually is expensive. Doing it well with AI-assisted tools like Descript, Opus Clip, or Otter is now realistic.
Technology doesn’t help with judgment. The content that actually works still requires a point of view, a specific audience understanding, and editorial discipline. The stack can amplify what the team is already doing. It can’t substitute for what the team isn’t doing.
Examples of Digital Marketing Strategies That Actually Compound
For teams looking for concrete patterns, three approaches have held up well over the past few years.
Topical authority via clustered content. Publishing depth on a specific set of topics until the brand becomes the reference source for that category. This requires content and SEO discipline, but it produces returns for years. Most brands quit before the compounding starts.
Audience-owned channels alongside rented ones. Building email lists, communities, and owned content platforms that don’t depend on paid media to reach the audience. Rented audience through paid media is necessary. It isn’t sufficient.
Brand and performance in the same plan. The separation of “brand marketing” and “performance marketing” as competing disciplines is one of the more expensive mistakes of the last decade. The strongest marketing operations treat them as connected, with brand investment producing search demand that performance marketing then captures. This connects directly to how branding and marketing work together rather than being separate practices.
None of these are new. All of them are still underused, because they require patience most marketing calendars don’t allow.
The AI Layer, Handled Honestly
AI-enabled marketing tools are the newest addition to the stack, and the one most likely to be misused.
The pattern that works: use AI to reduce the mechanical work around research, production, and personalization, so the human team can focus on strategy, judgment, and relationships. The pattern that fails: use AI to replace the mechanical work AND the judgment, hoping to produce marketing outputs at scale with fewer people.
The second pattern produces high-volume, low-quality content that competes with a lot of other high-volume, low-quality content. Some of it ranks. Very little of it builds brand or produces customer relationships that compound.
Teams thinking about AI in their marketing stack should also be thinking about generative engine optimization, because the same models that produce content are also the interface through which audiences discover brands. The two conversations are connected, and treating them separately misses the strategic point.
Where Watson Fits
Watson helps brands build marketing systems and digital experiences that connect strategy, creative, and technology in a way that produces measurable results. The technology decisions are downstream of the strategic decisions. When they get made in the right order, the stack works. When they get made out of order, the stack becomes an expensive documentation of past decisions.
Frequently Asked Questions
How much of the marketing budget should go to technology?
A functional benchmark is fifteen to twenty percent for growing brands, less for established ones with mature stacks. The exact number matters less than the distribution across layers. Overinvesting in activation while underinvesting in data infrastructure is the more common problem.
Should marketing technology decisions be made by the CMO or by IT?
Both, with clear ownership. The CMO owns strategy and use cases. IT owns integration, security, and data governance. Contested decisions usually mean the strategy and the infrastructure aren’t talking to each other early enough in the process.
Which marketing technology categories are overrated right now?
Point solutions that promise to replace strategy. Tools that automate personalization without connecting to a customer data platform. AI-driven content generators positioned as a substitute for editorial judgment. Each of these has a legitimate use case. Each of them is being sold beyond its use case.
Which categories are underrated?
Customer data platforms with strong identity resolution. Analytics infrastructure that supports incremental testing. Content and digital asset management systems that actually integrate with the workflow. These layers are boring, expensive, and where most of the durable value gets created.
How do we know if our current stack is working?
Three signals. Can we answer basic questions about customer behavior in less than a week. Is the team spending more time on judgment than on data pulls. Are the marketing outputs improving quarter over quarter, not just multiplying. If any of those answers is no, the problem is usually the stack architecture, not the individual tools.
