Operational bottlenecks are the steps in your delivery process that consume time and money without adding anything the customer would pay for. The fastest way to find them is to draw your process end to end, mark every step as value-adding or non-value-adding, and put a number on the non-value-adding ones. That number tells you what to fix first.
Most founders never do this. They feel the friction — the late nights, the re-done work, the orders that slip — and they respond by working harder inside a process they have never actually looked at. Below is the full method with a worked example, so you can run it on your own business this week.
Key takeaways
- A value stream map lists every step from customer order to delivery, and the only question that matters at each step is whether the customer would pay for it.
- Non-value-adding steps are not automatically waste; some are necessary controls. The waste is the waiting, rework, chasing, and duplicate entry around them.
- Costing waste means multiplying the time a step consumes by the person who does it, then adding the cost of the problems it creates downstream.
- Prioritization is not about fixing the biggest category; it is about fixing the bottleneck where wasted time blocks the most other steps.
The method in five steps
| Step | What to do | Why it matters |
|---|---|---|
| 1. List the steps | Write every action from order received to order delivered, in order | You cannot see waste in a process you hold only in your head |
| 2. Mark value | Label each step value-adding or non-value-adding | Separates customer-facing work from internal friction |
| 3. Tag the waste type | Waiting, rework, chasing, duplicate entry, handoff | Tells you what kind of fix is needed |
| 4. Price the waste | Multiply time spent by who does it, weekly, then annualize | Turns annoyance into a decision-grade number |
| 5. Rank and cut | Fix the step that blocks the most other steps first | Effort spent elsewhere is often invisible |
Step 1: List every step from order to delivery
Take a real order — not a typical one, an actual one from last week — and reconstruct it. Write down every single action, including the ones nobody thinks of as steps: the customer email sitting in an inbox, the spreadsheet update, the stock check done by eye, the message to a supplier, the label printed twice because the first address was wrong.
Do this on paper or in a doc. Aim for fifteen to thirty lines. If you finish in five, you are describing the process as it is supposed to work, not as it works.
Step 2: Mark each step value-adding or non-value-adding
The test is simple: would the customer pay for this step if they could see the invoice line? Picking the item adds value. Printing the label adds value. Retyping the same address from the storefront into the shipping tool does not — the customer would be baffled to be charged for it.
Be careful here. Some non-value-adding steps are regulatory or protective: checking a payment for fraud, confirming an age restriction, logging a return. Mark them as non-value-adding but necessary. The waste is not the check; it is the fact that the check takes six separate lookups because nothing is connected.
Step 3: Worked example — a five-person e-commerce company
Here is the map for a sample business selling a physical product. Five people: the founder, two warehouse staff, one customer service person, one marketing contractor.
Value-adding steps: pick and pack, quality check, carrier handoff, delivery confirmation.
Non-value-adding but necessary: payment verification, stock reconciliation, returns logging.
Pure waste:
- The founder manually copying each new order from the storefront admin into the inventory spreadsheet, because the two systems are not connected.
- Customer service emailing the warehouse to ask whether an order has shipped, because the customer asked first.
- Re-picking orders when the spreadsheet stock count disagrees with the shelf.
- The founder approving every refund personally, regardless of size, which creates a queue that stops on weekends.
- Searching three folders for the supplier's current price list because there is no single source.
None of these appear on the customer's invoice. All of them appear on the founder's calendar.
Step 4: Put time and cost on the waste
Now estimate. Roughly how long does each waste step take, and who does it? A founder spending twenty minutes a day reconciling orders is a founder losing over an hour and a half a week on work that produces nothing. Multiply that by the founder's effective hourly value, then by the weeks the business operates.
Do the same for the customer service person chasing shipping status, and for the warehouse re-picking orders. Add one more line: the revenue cost of the queue that stops on weekends, because refund approvals are almost always tied to a customer who will not buy again until they are resolved.
The goal is not precision. The goal is to stop calling it "just how we do things" and start calling it an annual figure. You can sanity-check your estimate against the free calculator at https://tools.exiuss.com/radar, which asks for six inputs and breaks the total down by category — weekly chaos, urgent problems, missed revenue, and wasted tools.
Step 5: Rank by blockage, not by size
Here is the mistake almost everyone makes. They total the waste, find the biggest line, and fix that first. But the biggest line is often not the bottleneck — it is downstream fallout from a smaller blockage upstream.
In the example, the manual order copy looks trivial: minutes per order. But it is the reason the stock count is wrong, which is the reason for the re-picks, which is the reason for the shipping status emails. Fix the copy, and three other waste lines shrink without being touched.
So rank by dependency. Ask of each waste step: how many other steps get slower or break because this one exists? The step that blocks the most is the one you fix first, even when it looks small on the ledger.
What to do differently than "automate it"
Founders jump straight to buying a tool. But automation applied to a broken sequence just makes the wrong process faster and harder to change. Run the map first, then match each waste line to one of four moves: systemize it, delegate it, automate it, or eliminate it. Some lines — the weekly report nobody reads, the approval nobody needs — should be eliminated, not automated.
Quick checklist
- Reconstruct one real order, step by step, on paper
- Mark every step value-adding or non-value-adding; flag necessary non-value-adding steps separately
- Tag each waste line by type: waiting, rework, chasing, duplicate entry, handoff
- Estimate weekly time per waste step and who does it, then annualize the cost
- Rank by how many other steps each waste blocks, not by size
- Assign one move per line: systemize, delegate, automate, eliminate
When you have your map and your annual figure, you are no longer guessing about what to fix. You are making a budget decision. Start with the free RADAR diagnostic at https://tools.exiuss.com/radar — it takes about five minutes, gives you a cost breakdown by category and a chaos archetype, and tells you which bottleneck to hit first.