Find Your Bottlenecks

Pinpoint what slows you down as volume grows.

What this recipe is for

Identifying the single points of friction—people, process, systems—that throttle throughput or scale, so you can fix the constraint and unlock more capacity without burning more resources.

What you’ll get

Visibility into your current capacity constraint

A prioritized list of fixes that unblock growth

A repeatable diagnostic for future scaling

Faster throughput with the same or less effort

Key inputs

Process cycle times (how long key tasks take)

Work-in-progress and queue lengths

Lead times (order-to-delivery, decision-to-execution)

Resource utilization (people, tools, systems)

Failure/rework rates

Volume vs. output over time

Core logic

Every system has a bottleneck—the slowest, most constrained part that determines overall capacity. Improving anything else without addressing that constraint delivers diminishing returns. Find the tight spot, fix or buffer it, then reassess the next constraint.

Step-by-step actions

Step 1: Map your core flow

Outline the end-to-end process for your primary value delivery (e.g., lead → sell → deliver → collect).

Step 2: Measure cycle times and queues

Track how long each stage takes and where work piles up. Look for disproportionate delays or growing backlogs as volume rises.

Step 3: Identify the constraint

The stage with the longest delay, highest queue, or most frequent failure is your current bottleneck.

Step 4: Diagnose root cause

Is it:

People overloaded or improperly trained?

A manual step that can be automated or standardized?

Waiting on external input (approval, materials)?

System/tool latency or poor integration?

Step 5: Apply the fix or buffer

Resolve the constraint (add capacity, simplify, automate, remove handoffs).

Or buffer it temporarily (work-in-progress limits, staging) while a permanent fix is implemented.

Step 6: Re-measure and repeat

Once the bottleneck shifts, repeat the process. Continual improvement builds scalable throughput.

Decision thresholds / guardrails

Work queues growing at a specific stage while others idle → Constraint exists there.

Throughput not improving despite added input upstream → You’re pushing into a bottleneck.

Fix causes downstream overload → Introduce pacing or buffers; don’t just shift the choke point without smoothing flow.

Multiple “almost” constraints → Triage by impact; fix the one causing the biggest delay first.

Examples

E-commerce fulfillment: Orders pile up because packing is manual. Fix: standardize SKUs, introduce batch workflows, or add a part-time packer to clear the constraint.

Service delivery: Client onboarding delays new project starts. Fix: create a streamlined intake form and delegate initial data collection to an automated system.

Agency: Creative approval waits block all downstream work. Fix: set fixed review windows and pre-approved templates to reduce decision latency.

Thinking checks

Where does work accumulate as volume increases?

If you double input, does output double—or does it stall?

Are you improving the real constraint or just adding pressure elsewhere?

Is the fix sustainable, or does it shift the bottleneck without addressing root cause?

If the answer is no…

Start with a simple time-and-queue audit for your top 3 processes.

Pick the biggest delay and apply a targeted fix (simplify, delegate, automate).

Measure throughput improvement, then move to the next constraint.

What to track (minimum)

Cycle time per stage

Queue/backlog size

Throughput vs. input volume

Frequency of rework/failures at each stage

Time to resolve the identified bottleneck