A two-week CCDV-F study plan, allocated by published weight

A two-week CCDV-F study plan that follows the blueprint rather than your interests: day one is a cold timed mock, days two to twelve are eleven study days allocated in proportion to the published domain weights, and days twelve to fourteen are two fresh papers and a review of what to study next. Three of the eleven study days go to a single domain, because that domain is a third of the exam.

The shape of the fortnight

One cold practice paper, eleven study days, two mocks, and a final study review. The cold paper comes first because a score by domain offers evidence for adjusting the days below. Spending one of fourteen days on that baseline can make the remaining study time more focused. The free CCDV-F practice paper runs at full length under the exam’s clock and needs no account.

The two mocks are deliberately at the end and deliberately fresh. A paper you have already seen measures recall of that paper. If you have only three unseen mocks, this plan uses all three and keeps none in reserve. Each fresh form adds a new set of mistakes to analyse.

Why three days go to one domain

Applications and Integration is 33.1% of CCDV-F, roughly 17 of 53 items, and larger than the next two domains combined. Eleven study days split by published weight comes out at 3.6 for it, 1.8 for model selection, 1.6 for agents, 1.2 each for prompting and tools, 0.9 for security and 0.3 each for the two smallest, which rounds to the plan below. The full breakdown is in the post on that domain.

The contrast worth holding is the architect foundations plan, whose blueprint is flat enough that weighting produces a near-even fortnight.

The uncomfortable part is what that leaves. Claude Code is 3.1% and eval and debugging is 2.6%, three items between them, and a shared day at the end. They are the most enjoyable material on the paper to study and they are worth about a twentieth of it.

The day-by-day plan

Planning at skill level

This is the one track where a study plan can go a layer deeper than domains, because CCDV-F publishes weights for all 25 of its skills. That is what days two to four are doing: not three days of “applications”, but application design at 8.6%, then software engineering foundations at 7.4%, then API mechanics at 6.8% and configuration management at 4.1%, in descending order of what they are actually worth.

The same instrument is worth using inside the model-selection days, where it produces a surprise: the skill the domain is named after is its smallest, at 2.7%, while technical fundamentals is 6.1%. Details in the model selection post, and the full weight table for every track in the domain weights post.

What one study day looks like

Ninety minutes, in three unequal parts. Twenty on reading the objective and its skills, and writing each one down as a decision rule in your own words: a trigger and a response, not a summary. Fifty on practice items in that domain only, answering closed-book. Twenty on the misses, naming which rule would have caught each one.

A miss that maps to no rule you have written is the useful result, because it means there is a rule you have not found yet. That repair loop matters more than the reading, and the twelve recurring decision rules are a reasonable starting library to write against.

The review on day 14

No guide publishes a raw-to-scaled conversion for the 720 standard, so a practice percentage tells you nothing certain about the official result. Day 14 reads Cred Farmer's readiness score instead. It uses our own 100 to 1,000 practice scale and counts the first sitting of each full form, guided or timed, never a drill, a re-sit or the free no-account paper.

A CCDV-F form has 53 items, so 720 means 37 correct (70%) and 800 means 42 correct (79%). The score reaches its Exam-ready band (90) only once your first sittings average 800 or more, and since no real exam result has ever been compared with that scale, it stays a signal rather than a promise. The real exam does not judge domains one by one either, so any lower domain result or repeated miss simply hands that topic the next block of study. For context on the official scaled score, read the passing score post. The repair loop is in the study guide.

Key takeaways

• Cold practice paper first, on day one. A cold score by domain is what tells you which days you can shorten.
• Three of eleven days on one domain. Applications and Integration is 33.1%, so 27% of the study time still under-serves it.
• Allocate at skill level. CCDV-F is the only track publishing all 25 skill weights; use them inside the big domains.
• The interesting domains are the smallest. Claude Code and eval are three items between them and share one day.
• Both mocks fresh, both at the end. A paper you have seen measures recall of that paper, not your broader command of the objectives.
• The readiness score is a signal, not a promise. No guide publishes how raw performance becomes a scaled 720, and no real exam result has checked our practice scale.