Do the math on your Sunday sometime. Not the guilt-math ("I should be faster"), the actual math. If you teach five sections and you spend even 25 minutes tailoring your lesson for each one, that's over two hours — on top of writing the lesson in the first place. Add materials, and you've lost an afternoon you'll never tell anyone about.
Here's the thing no one says out loud: the reason planning takes all weekend usually isn't that you're slow. It's that you're doing the same job five times.
You're not writing one lesson. You're writing five.
When teachers say "planning takes forever," they picture the writing — the objective, the activity, the exit ticket. But that part is rarely the bottleneck. The bottleneck is the versions.
You write a strong lesson. Then you rebuild it for the class with eight ELLs. Then again for the section with six IEPs. Then you add an extension so your honors kids don't finish in four minutes and stage a quiet coup. By the time you're done, you didn't plan a lesson. You planned a lesson, then re-planned it four more times, by hand, one class at a time.
That's not a speed problem. It's a structure problem. And you can't out-hustle a structure problem — you can only change the structure.
Why the usual "time-saving" tips don't work
Most lesson-planning advice tackles the wrong minute:
- "Use a template." Templates speed up the shape of a lesson, not the five versions. You still differentiate each one by hand.
- "Reuse last year's." Helps until you hit a class with a different mix of students — which is every class.
- "Just use a generic AI tool." Now you get one bland lesson fast, and you spend the saved time rewriting it to fit real students. You moved the work; you didn't remove it.
Every one of these leaves the actual time-sink — building the versions — sitting on your plate.
The fix: separate the lesson from its versions
There's a genuinely different way to think about this. A lesson has a core that should never change between your classes — the standard, the objective, the essential question, the way you check for understanding. What changes is the access: the reading level, the scaffolding, the language supports, the extension.
Once you see it that way, the goal is obvious: build the core once, and generate the access-versions automatically. You make the professional decisions — what to teach, how to assess it — and something else handles the mechanical multiplication of turning one lesson into five.
That's the difference between a two-hour Sunday and a twenty-minute one.
This is the whole idea behind Tangram
Tangram is built on exactly this principle. You build one strong lesson with Tan — our AI planning partner powered by a methodology our founders developed across hundreds of districts — and then you tell it who's actually in each class: your ELLs, your IEP accommodations, your reading levels, your MTSS tiers.
It adapts every section and hands you a version built for each group you teach. Same objective, right access, every time. Because the personalization is rule-based, there's no extra cost per adapted copy — one build becomes five ready-to-teach versions. And every version is observation-ready, aligned to Danielson, Marzano, and Understanding by Design.
You're still the one directing the lesson. You've just stopped doing the same job five times.
Get your Sunday back
Build one lesson, get a version for every class you teach. 3 free lessons — no credit card.
Try 3 lessons free →Your weekend isn't the price of teaching well. Build once, teach every class, and take the afternoon back.