Stacks on Stacks: the changing GTM tech stack

Oxygen Advisors’ Jacques Lourens, VendorSage’s Steve Baker, and Tracksuit’s Peter Sloan unpack how the GTM tech stack is shifting - from vendor consolidation plays and AI hype to what real stacks look like at different ARR stages. They focus on making staged, flexible decisions, the role of RevOps, and why data hygiene is the foundation for enablement, reporting and AI performance.

Key themes include:

• Cutting through the noise
  Vendors are pushing “one vendor, many use cases,” while buyers prefer flexibility and shorter terms. AI-native tools are shipping faster than incumbents pivoting to AI; adoption sticks only when driven by clear use cases. Ignore the hype and don’t be sold to - work backwards from business problems.

• Stacks by stage (patterns, not prescriptions)
  Early stage to ~$5m ARR: keep a lightweight, HubSpot-led stack, experiment freely, and turn on call/meeting recording early to capture context. Around ~$5m–$10m, formalise enablement, improve data quality, explore partners, and add CPQ carefully; by ~$10m–$20m, many shift to Salesforce for scale, add governance and ramp AI experimentation as data hygiene improves. In contracting, optimise for flexibility over discounts - long lock-ins can constrain you for years.

• AI as augmentation
  Let AI do 0–70%; humans own the last 30% for context, quality, and judgment. New capability mixes (for example, GTM engineer) and the idea that “everyone’s a context provider” across ops, data and frontline teams.

• Use cases before tools
  Tool-first buying drives failure; define the problem, map the workflow, then pick tech. Many record conversations, but few activate them. Turn recordings into assets, insights and model inputs.

• Keep it simple; protect data hygiene
  Don’t go too big too early: avoid multi-year contracts and seat bloat before processes mature. Clean, structured data underpins automation, trustworthy reporting, board visibility and AI effectiveness. Use tools as intended; run proofs of concept on small datasets before scaling.

Key takeaways:

• Invest in RevOps as soon as you have repeatability and build the muscle for non‑linear growth.

• Define clear use cases and processes first; evaluate vendors second. Run controlled experiments and proofs of concept.

• Prioritise data hygiene early and switch on call/meeting recording to capture high‑quality context for enablement and AI.

• Optimise for flexibility over headline discounts; avoid accidental complexity and premature enterprise tooling.

• Layer capability by stage. Copying a 10–20m ARR stack too early slows you down.

Watch the full discussion to explore these insights in depth.