Marketing technology is in a strange spot: the landscape is bigger than ever, yet its share of budgets is shrinking. Scott Brinker’s latest count puts the market at 15,505 products, but Gartner’s 2026 CMO Spend Survey shows martech at 19.4% of marketing budgets, down from 26.6% in 2021. Paid media, meanwhile, climbed to 31.4%.
The working line from PPC Land’s explainer is useful: martech manages customer data, email, content and automation, while ad tech buys and sells paid media. The two touch constantly.
What a stack actually does
Enterprise stacks usually follow the same layering. A CRM is the system of record. A CDP or warehouse such as Snowflake or Databricks unifies behavioural and profile data. Engagement tools send email, SMS and push. Content systems, web analytics and attribution complete the picture.
The day-to-day flow runs in five steps.
- Collection: tags, SDKs, forms and server feeds record events like “product viewed” or “added to cart”.
- Identity resolution: hashed email, loyalty number, device ID and cookies are stitched to one person.
- Segmentation: rules or predictive models group profiles.
- Activation: segments are sent to email journeys or pushed into ad audiences.
- Measurement: opens, clicks and orders are written back to the profile.
Why this matters for performance marketers
First-party data held in a CRM or CDP now determines what an ad platform can target and measure as third-party cookies and device IDs thin out. The Hightouch and The Trade Desk integration reported in June 2026 matches DSP exposure logs to a brand’s own customer IDs inside the warehouse—work that previously took days through third-party identity providers.
But utilisation remains the old complaint. Gartner found marketers used 58% of stack capabilities in 2020, 42% in 2022 and 33% in 2023, before recovering to 49% in 2025. Data quality is also weak: in an Adverity survey, CMOs said 45% of marketing data used for decisions was incomplete, inaccurate or outdated.
Suites vs composable stacks
The architecture question is real. Suites from Salesforce, Adobe, Oracle and HubSpot bundle layers under one contract. Composable stacks keep customer data in a company’s own warehouse and plug specialist tools into it. With Databricks launching CustomerLake and Publicis agreeing to buy LiveRamp, the warehouse-native direction is strengthening.
AI is changing how the tools are operated
Gartner found 81% of martech leaders piloting or running agent initiatives, but 45% of those with agents said vendor-offered versions fell short of promised performance. Platforms are opening up: Salesforce’s Headless 360 exposes every capability as an API or MCP tool, and Meta released MCP connectors for its ad system. Integration traffic to Zapier, n8n and Make is showing double-digit declines as agent usage climbs.
What to do about it
Start with an honest utilisation audit before buying another tool. Map the identity path from collection to activation, because that is what makes Meta and Google targeting work. Renegotiate usage-based contracts, since half of CMOs in Gartner’s 2026 survey say they continuously renegotiate to avoid spikes.
Source: PPC Land
