The first unit plan I ever wrote was just five lesson plans in a folder with a cover sheet stapled on top. It looked complete. It wasn't, because none of the lessons knew about each other. Day 3 assumed something day 2 never actually taught. The test on day 10 asked about a skill we'd touched for eight minutes on day 1 and never came back to.
A real unit plan isn't a stack of lessons, it's a sequence with a spine: one standard, broken into the sub-skills a student needs in order, each lesson building on the last, ending at an assessment that actually matches what got taught.
Start from the end, not day 1
Before you sequence anything, write the summative assessment, or at least sketch it. If you don't know what day 10 looks like, you can't know what day 3 needs to accomplish. This is backwards design in one sentence: the test tells you the sequence, not the other way around.
The four things a unit plan needs that a lesson plan doesn't
- An essential question. Not "what is photosynthesis" but something a student could argue about across the whole unit, revisited on day 1 and day 10.
- A skill progression, not a topic list. "Introduce, practice, apply, assess" for the actual sub-skills inside the standard, in the order a student needs to encounter them.
- A formative checkpoint mid-unit. Around day 5 or 6, something quick that tells you whether to move on or reteach, before the summative assessment makes that decision for you.
- Resource continuity. If day 2's vocabulary doesn't show up again until day 9, it's gone. Plan for repetition, not just introduction.
A prompt that gets you a real sequence
"I teach [grade/subject]. My standard is [paste the code and language]. My unit is 10 class days, 50 minutes each. Give me a day-by-day skill progression (not just topics) that builds toward a summative assessment on day 10, including where a mid-unit formative checkpoint should sit and what it should check." You'll get a spine. You still need to write the actual lesson content for each day, but you're not guessing at the order anymore.
Where the model runs out of usefulness
- Your students' actual background knowledge. It doesn't know that last year's class never really got fractions and this year's needs an extra day of review before the unit even starts.
- Pacing around real school life. Fire drills, assemblies, the week before winter break, none of that is in the prompt unless you put it there, and it changes what's realistic.
- The essential question's honesty. A generic question ("why is history important?") won't hold a class's attention for ten days. You know your kids well enough to pick one that will.
A short example
Standard: analyze how an author develops a central idea (a common ELA middle-grades standard). A topic-list unit might go: "intro central idea → practice → essay." A skill-progression unit goes: day 1-2, identify explicit textual evidence for a stated idea; day 3-4, distinguish central idea from topic (the trap most students fall into); day 5, formative check, can they tell the difference on a new text; day 6-8, trace how the idea develops across a full text, not just one paragraph; day 9-10, write the analysis, using the vocabulary and moves practiced every day before it. Nothing on day 9 is new. That's the point.
The part worth doing on paper first
Before you touch any tool, sketch the sequence by hand, one line per day, in whatever order feels right. Then check it against your essential question and your summative assessment. Half the time you'll find a day that doesn't actually build toward either one, easier to cut on paper than after you've written the full lesson.
Our [unit plan builder](/unit-plan) turns your standard, grade, and duration into that day-by-day sequence, essential questions, aligned assessments, and key resources included, so you start from a real spine instead of a blank page. Pair it with our [lesson plan generator](/lesson-plan-generator) once the sequence is set, to build out the actual day-by-day lessons.
