Vision vs. Velocity: The Hidden Trade-offs That Stall Engineering Progress

David Dieruf2025-10-13

There's a false choice that's killing engineering teams — it's vision vs. velocity. The real struggle isn't deciding between the two; it's surviving the pressure to do both at once.

Optimize for velocity (speed) and you risk building the wrong thing. This means rebuilding, rearchitecting, and (in the process) crushing morale. Fixate on the vision, and you spend too long perfecting it, and you miss the window of opportunity. Someone else captures the market while you're still whiteboarding.

This isn't a theoretical debate. This is the reality of every leader caught between executive pressure and engineering reality.

The truth is, vision and velocity aren't trade-offs. If you neglect either one, the other becomes irrelevant.

When Velocity Becomes the Enemy of Progress

In our combined 45 years of software engineering experience, Wael and I have seen the false choice play out over and over again in various companies. One recent example was in 2020.

Prior to founding Kisasa, Wael was working with a public health agency to develop a self-service app that would guide people through at-home COVID testing. Things like including how to use the test packets, what symptoms to watch for, and next steps.

A VP gave the executive order, "Ship something fast." The goal was clear: To get a working version of the COVID self-testing app into people's hands before infections spiked again.

A small team of developers sprang into action, gathering the details available at the time, and started building. They bundled up triage, kit guidance, appointments, and notifications into one monolithic app and shipped it - in 12 weeks flat.

Then, reality hit.

As the rollout expanded beyond the US to the UK, Japan, and other countries and regions, the design collapsed under jurisdictional nuances and uneven team maturity. Health data required country-specific consent, retention, and storage. Guidance and eligibility were updated weekly by region, turning hard-coded flows into release bottlenecks. Multiple regional teams needed independent releases, but the monolith forced serial merges, blocking each other.

Within three months, they had to rebuild everything. Ultimately, shortcutting the vision for the sake of velocity didn't save any time at all. Instead, it doubled the project's cost, and worse, delayed the release of a much-needed application into a world in crisis.

Market Reality: Why Vision Alone Fails Too

On the other side of the spectrum sits vision without velocity.

Here's a well-known yet very relevant story: Google's researchers had state-of-the-art large language models years before OpenAI launched ChatGPT. In terms of raw capability, Google was ahead. But internal debates over safety, morality, brand risk, and long-term implications slowed their rollout. Committees reviewed use cases, PR teams weighed reputational fallout, and legal teams scrutinized liability. The result? Progress stalled.

Meanwhile, OpenAI took a different path. Their bet wasn't on perfection, but on timing. They shipped a "good enough" model, opened it up to millions of users, and learned in the wild. Each release fueled a feedback loop: adoption drove hype, hype attracted capital, capital funded iteration, and iteration improved the model. Within months, OpenAI captured both the narrative and the market.

The lesson is clear: Vision without velocity isn't leadership; it's hesitation.

You can have the best technology in the world, but if you miss the window, someone else with less polish but more urgency will own the market. In today's environment, market momentum compounds faster than technical advantages. Vision sets direction, but velocity decides who gets there first.

Balancing vision and velocity

You can't treat vision and velocity as opposites. Vision without velocity is a stalled idea; velocity without vision is wasted energy. Real breakthroughs, the kind that change markets and build profitable, lasting businesses, come only when the two move in balance. You need clarity on where you're going and the ability to move fast enough to get there before the window closes.

But here's the catch: whether you achieve that balance has less to do with technology itself and more to do with people. The way teams are structured, how decisions are made, and how communication flows — all of it shapes the systems you end up building. That's Conway's Law in action: your organization chart becomes your architecture.

The People Problem: When Politics Shapes Architecture

Conway's Law is brutal in its simplicity: Systems mirror the communication structures of the organizations that build them. If your teams are siloed, your architecture will be siloed. If your org chart is a monarchy, your code will be a monarchy.

We've seen this play out countless times:

  • **Siloed teams, siloed systems: **Each department builds its own "mini-platform" for local needs. At first, it looks efficient — until integration means stitching together spreadsheets, APIs, and manual workarounds. The architecture mirrors the org chart instead of the business.
  • Political bottlenecks: Architecture decisions don't stall because the code is complex — they stall because leaders are stuck in cold political handshakes. Every integration requires sign-off across territories, and what should take days drags into weeks as executives negotiate turf, budgets, and credit.
  • Duplicated effort: Two teams solve the same problem in isolation. One builds a customer portal; another builds a near-identical portal under a different VP. Months later, leadership realizes they're paying to maintain two codebases that should have been one.
  • Integration as negotiation: When it's time to connect these systems, engineers aren't debating the best architecture — they're navigating organizational politics. The question isn't "What scales best?" but "Whose roadmap wins?"
image1

Conway's Law in the wild

Take something as basic as user authentication. One enterprise we worked with had two major product lines, each run by separate divisions. Both teams needed login and identity management. With no alignment, Team A built an SSO solution tuned for enterprise customers; Team B built a lighter-weight login service for consumers.

For two years, they ran in parallel. Customers who bought both products had to manage two sets of credentials. Support tickets exploded. Meanwhile, engineering had to patch security vulnerabilities in two codebases and maintain two roadmaps.

On paper, this looked like a "technical problem." In reality, it was Conway's Law: two siloed org units, two siloed architectures. The cost wasn't just duplicated systems — it was lost customer trust, slower delivery, and higher security risk.

So many "technical problems" aren't technical at all. They're alignment problems disguised as architecture. Until you fix the communication and decision-making structures, the system will continue to reflect the dysfunction of the organization that built it.

You can hear more about Conway's Law in the Going to Production podcast episode.

Kisasa Blueprint: Aligning Vision and Velocity

The good news: This false choice isn't inevitable. You *can *have both vision and velocity if you align the two from the start.

That's the heart of the Blueprint Methodology. It's Kisasa's way of pulling organizations out of the trap and giving engineering leaders a way to move fast in the right direction.

Instead of asking teams to choose between speed and strategy, the Blueprint reframes the problem: Get everyone aligned on the destination first, then design the path that makes velocity sustainable.

  1. Getting Aligned — Vision Before Velocity

Most teams start with tools. The Kisasa Blueprint starts with questions:

  • What are the business goals over the next 6–12 months? Not features, not tech, but outcomes. Are you trying to expand into a new region? Reduce churn? Ship a new product line?
  • Which problems are you actually solving? VPs know "feature X" isn't the same as solving "customer retention." This phase draws the line clearly.
  • What does success look like for both teams and customers? Success isn't shipping a dashboard; it's reducing support tickets by 40% or doubling daily active users.

Phase 1 of Kisasa's Blueprint process is less about engineering and more about alignment. Product, engineering, architects, and executives sit at the same table. Without this step, everything that follows is noise — activity without direction.

  1. Discover What's Working Today — Current State

Before you can chart a path forward, you need to know where you are:

  • People: Team structure, skills, communication patterns, and politics. Who talks to whom? Who blocks whom? This is Conway's Law in practice.
  • Technology: Systems, deployments, data flows, technical debt. Which parts are fragile? Which are solid?
  • Business: Goals, constraints, budget, and market pressure. Can we afford a complete rebuild, or do we need phased improvements?
  • Optional: Automated Source Code Analysis. When legacy systems are involved, the Blueprint flags complexity hotspots, dependency risks, and maintainability scores. This shortens discovery and reduces guesswork — so conversations focus on facts, not politics.

The Kisasa Blueprint deliberately removes judgment from the conversation. This isn't about blaming teams for "bad code" or "slow delivery." It's about achieving clarity.

  1. Create the Blueprint

    Now we need a map of how we are going to achieve those business goals.

  • Tailored architecture: Every org is different; the goal isn't "best practice" but "best fit." What works for Netflix or Google may be disastrous for a mid-sized bank or a SaaS startup. The Blueprint takes into account the specific realities of your organization — team maturity, regulatory constraints, market pressures, budget, and appetite for risk.

        Instead of asking, "What's the textbook best way to build this?" the Blueprint asks, "What's the best way for us to build this, given our current state and future goals?"

        That's how you avoid the trap of over-engineering a platform your teams can't maintain or under-investing in a foundation that collapses under growth. "Best fit" means designing architecture that your people can actually run, scale, and evolve, without grinding delivery to a halt.

  • Balancing quick wins with long-term stability: Leaders get a plan that builds momentum now, without sacrificing the foundation. \

  • Trade-off analysis: Every architectural decision is explained in business terms: why this approach saves money, reduces risk, or accelerates time-to-market. \

  • Implementation strategy. A step-by-step bridge from today's systems to tomorrow's architecture. Who does what, when, and why.

Kisasa's Blueprint turns abstract principles into concrete choices — decisions made with full context, not in isolation.

Get your Blueprint started with Kisasa today

Whether you're advocating for modernization or evaluating strategic options, our Blueprint gives you the technical clarity and business case you need to move forward with confidence. Explore the engagement details and get started with Kisasa Labs.