Skip to content

08-tech-decisions.md

Latest commit

 

History

History
95 lines (68 loc) · 3.83 KB

08-tech-decisions.md

File metadata and controls

95 lines (68 loc) · 3.83 KB
Raw

Tech Decisions

What

Technology decisions — whether to build or buy, which vendor to choose, which technology to adopt — have lasting impact. This guide provides a framework for making these decisions with your eyes open.

Build vs Buy

When to Build

  • The software is core to your competitive advantage
  • No existing solution fits your specific requirements
  • You need full control over the technology and roadmap
  • You have the team to build and maintain it long-term

When to Buy

  • The problem is solved well by existing products
  • The software is not your competitive advantage (email, accounting, CRM)
  • Speed to market matters more than customization
  • Your team is small and needs to focus on core product

The Hidden Cost of Building

Building is not a one-time cost. You also pay for:

  • Ongoing maintenance and bug fixes
  • Security updates and dependency upgrades
  • Feature requests and changes
  • Onboarding new developers onto the custom system
  • Infrastructure and hosting
  • Opportunity cost (what your team could have built instead)

Rule of thumb: Building costs 3-5x the initial development over 5 years.

Evaluating Vendors

Functional Fit

  • Does it solve your actual problem? (Not the demo problem — your problem)
  • Can you test it with your real data before committing?
  • What's missing from the demo that you'll need?
  • How customizable is it?

Technical Fit

  • Does it integrate with your existing systems?
  • What are the API capabilities?
  • What's the data migration path?
  • What's the vendor's uptime SLA?
  • Where is data stored? (Important for compliance)

Financial Fit

  • Total cost over 3 years (not just the first year discount)
  • Cost at scale (pricing often changes as you grow)
  • Hidden costs (implementation, training, add-ons)
  • Exit costs (how do you get your data out?)

Vendor Health

  • Are they profitable or burning investor money?
  • What's their customer retention rate?
  • How responsive is their support? (Test this before buying)
  • What's their product roadmap? Does it align with your needs?
  • What happens if they get acquired or go out of business?

Understanding Trade-offs

Every technology choice involves trade-offs. There is no best option — only the best option for your specific context.

Common Trade-offs

Trade-off Question to Ask
Speed vs Quality Can we ship fast and iterate, or does it need to be right the first time?
Flexibility vs Simplicity Do we need to handle many scenarios or just the common ones?
Cost vs Reliability What's the cost of downtime?
Custom vs Standard Does customization give us a real advantage?
New vs Proven Are we comfortable with the risk of new technology?

How to Decide

  1. List the options — Usually 2-4 realistic choices
  2. Define criteria — What matters most? Cost, speed, reliability, flexibility?
  3. Weight the criteria — Not everything is equally important
  4. Score each option — Against each criterion
  5. Check for deal-breakers — Any option that fails a non-negotiable requirement is out
  6. Make the call — With imperfect information. Waiting for perfect information is itself a decision.

Common Mistakes

  • Choosing technology because it's trendy, not because it fits the problem.
  • Not planning for the exit. If you can't migrate away from a vendor, they have leverage over pricing and roadmap.
  • Over-optimizing for the current state. Choose for where you'll be in 2 years, not where you are today. But don't over-engineer for hypothetical scale either.
  • Deciding by committee. Collective input is good. Collective decision-making is slow. Assign a decision maker.