The 15-Minute H2 Load Test

Five questions that test your second-half plan against the operation that has to carry it.

This week we have been making one argument. The mid-year review scores outcomes, the operating machine rarely gets examined, and so most H2 plans inherit the H1 bottlenecks. A plan written on an unexamined machine is a forecast of the constraint, not the ambition.

Today is the practical part. Below is the full load test: five questions you can run against a finished H2 plan in about fifteen minutes. The whole thing is on this page, free, with no gate and no email capture. There is nothing to download and nothing to sign up for. Read it, run it, keep it.

You need your H2 plan, your H1 results, and something to write with. Take each question in order, and write the answers down rather than answering in your head. Vague answers evaporate. Written ones argue back.

1. Which line item assumes capacity that does not exist?

Go line by line and ask what each commitment costs in hours and attention, and whose. Then ask where those hours come from.

Most plans contain at least one line that assumes capacity nobody currently has. The new service launch that needs your two best delivery people, who are already fully booked. The marketing push that needs writing time the calendar has never once produced. The plan rarely says "we will find the time somewhere." It just assumes it, in the gap between the commitment and the roster.

Mark every line where you cannot name the specific hours that will fund it. Those lines are not plans yet. They are invoices the second half has not agreed to pay.

2. Which H2 commitment shares a root cause with an H1 miss?

We covered this one in depth yesterday, so here it is in brief.

Put your H1 misses next to your H2 commitments and look for shared roots. Not shared topics. Shared causes. If a launch slipped because delivery consumed every spare hour, then every H2 line that depends on spare delivery hours shares that root cause, whatever its subject matter.

A new promise drawing on the same well as an old miss usually fails the same way. The fix is not a better date. It is naming the root cause in the plan and changing something about it, or shrinking the promise to fit the well.

3. Where does work queue when volume rises?

The plan presumably involves more of something in H2. More clients, more projects, more proposals. So ask: when volume rose in the past, where did work pile up?

Every operation has a place where work waits. A person everything routes through, a review step that only happens weekly, a handoff that depends on someone's inbox. At normal volume the queue is invisible. At higher volume it is the whole story, because the queue grows faster than the throughput does.

Your H1 has the evidence. Think of the busiest stretch of the half and where things sat. If the plan increases volume without changing that spot, the queue is in the plan, just unlabeled.

4. Who decides what changes without you?

Pick the most important recurring decision in your delivery work. Scoping a project, approving a deliverable, handling the awkward client conversation. Then ask one question: who decides that without you?

If the honest answer is nobody, then you are a line item in every commitment the plan makes, whether your name appears or not. The plan's real capacity is not the team's hours. It is the number of decisions that can route through one person in a week, and that number does not grow because a document asks it to.

One question is enough here. You do not need a full audit of founder dependence today. You need to know whether the plan assumes a version of you with more hours than you have.

5. What does the plan assume stays quiet that was loud in H1?

Every plan assumes a baseline of calm. The tools keep working, the workaround keeps holding, the client keeps being patient, the recurring fire keeps not recurring.

So ask: what was loud in H1 that the plan assumes will be quiet in H2? The thing that ate three days a month. The process everyone routes around. The duct tape that has become the process, the pattern we wrote about in when workarounds become the way the business runs. Noise like that rarely silences itself. If the plan has no line addressing it, the plan is betting against your own recent evidence.

List the three loudest things from H1. Then check whether the plan mentions any of them. The gaps are the assumption.

What to do with the answers

Score it simply. Each question either found something or it did not.

Zero or one finding, your plan and your operation are reasonably aligned, and Q3 2026, which starts Wednesday, July 1, can begin on schedule with a clear conscience. Two or more, the kindest thing you can do for the second half is spend a day revising the plan now, while it is still paper. Usually that means shrinking a line, resequencing a commitment, or adding the operational change the plan was silently assuming.

Notice what is not on that list. Not a new tool, not a new hire, not a standing meeting. When a plan fails its load test, the instinct is to add something. Most findings are fixed by removing something instead: a line, a dependency, a promise the operation never agreed to. The plan usually needs to lose weight, not gain headcount.

The test is yours either way. A diagnostic you have to trade your email for is already a pitch. This one is just a test. Fifteen minutes now is cheaper than discovering the same answers in October.

One more thing for the week. The question that tends to sting most is the first one, the capacity that does not exist. Tomorrow we will look at where that capacity usually went, and it is probably not where you think.

Continue the series

This is part 3 of a 5-part series on Load-Test the Plan. Then Look at the Price.. The full arc:

• Monday: Your H2 Plan Is a Forecast of Your Bottleneck
• Tuesday: The Line You Copied From January
• Wednesday: The 15-Minute H2 Load Test (this post)
• Thursday: The Workflow Is Not Broken. It Is Executing the Price.
• Friday: Utilization Is High Because Pricing Is Broken

Content and Accuracy Disclaimer

This article was drafted with AI assistance and reviewed by the Altvina team. We rigorously fact-check all content to ensure reliability.

Should you notice any inaccuracies or outdated information, please contact us so we can correct it. Your feedback helps us maintain high standards of accuracy and transparency.