The Index Card Tower Challenge: Design by Iteration

The measurement that matters in this activity is not the height of the tower. It is the difference between round one and round three. Run the index card tower challenge once and you get a tower. Run it three times with a timer and a two-minute review between rounds, writing the height down each time, and you get a graph of a child getting better at something in forty minutes. Most families see round one at about 30 cm and round three somewhere past 70 cm, with the same twenty cards. Nothing about the materials changed. The child changed.

Key Takeaways

• You’ll run three ten-minute builds with the same twenty index cards, measuring height after each round and reviewing what failed in between.
• The concept is iterative design: a structure improves through cycles of build, fail, diagnose and rebuild, which is how engineering actually works.
• Two numbers to track: tower height per round, and how much mass the top can hold before collapse.
• Best for ages 6 to 13. Six-year-olds need the rounds shortened to five minutes.
• The usual trap is a child defending their first design instead of replacing it. The two-minute review exists to prevent that.

What You’ll Need for the Index Card Tower Challenge

Item	How much	Where to get it	If you don’t have it
Index cards, 3 × 5 in (7.6 × 12.7 cm)	20 per round, unused	Stationery shop	Cut cereal-box card or thin card into 7.6 × 12.7 cm rectangles
Masking tape	Exactly 30 cm per round, cut in advance	Hardware drawer	Any tape, measured out and cut before the timer starts
Scissors	1 pair	Kitchen drawer	Tear along a folded line
Tape measure or metre rule	1	Toolbox	A broom handle marked in 10 cm steps
Timer	1	Phone	A kitchen timer
Ping-pong ball or 20 g weight	1	Games cupboard	A small bag of coins, weighed
Notebook	1	Anywhere	Back of an envelope

Time: prep 5 min, three rounds plus reviews 40 min, cleanup 5 min Cost: a packet of index cards Mess level: low

Do the Experiment

Step 1 — Set the rules before the first round

Twenty cards, 30 cm of tape pre-cut into six 5 cm pieces, ten minutes, free-standing tower, measured from the table to the highest point. No leaning on walls, no taping to the table. Write the rules down, because mid-round rule changes destroy the comparison.

Step 2 — Round one, with two designs instead of one

Ask for two different towers in the first ten minutes, each using ten cards, rather than one careful tower. This feels wasteful and it is the most valuable instruction in the activity. Measure both and record both heights.

Step 3 — Run the two-minute review

Stop building. Ask three questions and write the answers: where exactly did it fail, was it a buckle, a tip, or a joint, and which of the two designs would you keep? Do not let the review run long. Two minutes is enough and it keeps the pace.

Step 4 — Round two: rebuild from the better idea

Fresh cards, fresh 30 cm of tape, ten minutes, one tower this time. Measure and record. Then add the ping-pong ball to the top and see whether it survives, recording that as a separate result.

Step 5 — Round three: one change only

Before building, have your child name the single change they are making and why. One change, stated in advance. Build, measure, record, test with the ball again.

Step 6 — Plot the three heights

Draw a simple graph with round number on one axis and height on the other. Three points. Then ask which round produced the biggest jump, and why that might be.

What you should see: Round one typically lands between 20 and 40 cm, often with a wide flat base and flat stacked cards. Round two usually jumps substantially, to 45 to 70 cm, because almost every child discovers rolled tubes or folded triangular columns on their own after watching flat cards fold. Round three gains less, often only 10 to 20 cm, which is a genuine and useful result: the early gains are the cheap ones. Towers that hold the ping-pong ball almost always have a flat card taped across the top as a platen, which spreads the load. Expect at least one dramatic collapse, and expect it to be the most informative moment of the afternoon.

What the Index Card Tower Challenge Measures

Round	Typical height	Typical design	What usually fails
1	20–40 cm	Flat cards stacked, wide base	Cards fold under their own weight
2	45–70 cm	Rolled tubes or triangular columns	A joint peels, or the stack leans and tips
3	55–90 cm	Tapered tower, braced tiers	Buckling of the longest column
With a ball on top	Any	Needs a card platen on top	Point loading crushes one column

The gap between round one and round two is the measurement that matters. Record it, because that number is the child’s learning rate on this task, and it is the one thing a single-round challenge cannot show you.

What’s Actually Going On

Two things are happening at once, one structural and one about process.

Structurally, a flat index card is almost useless in compression. It has a huge area but almost no thickness in the direction that resists bending, so it folds at a few tens of grams. Roll the same card into a tube and it becomes a column that can hold a kilogram or more, because the material has moved away from the centre line where it does nothing. The governing quantity is the second moment of area, which rises steeply as material moves outward from the axis. This is the same principle covered in the paper column experiment and in standard mechanics teaching such as MIT’s Mechanics of Materials.

Once your child is building with tubes, the failure mode changes to buckling, where a column suddenly bows sideways rather than crushing. Buckling load falls with the square of the column’s length, so three short tiers beat one tall tube every time, which is exactly the discovery most children make in round three without being told. Thin rolled tubes also buckle locally, crumpling in a small zone rather than bending as a whole, and real thin shells fail well below the ideal theoretical load because of small imperfections. That effect is important enough that NASA published design criteria for it in SP-8007 on the buckling of thin-walled circular cylinders.

The process half is where the research lives. Dow and colleagues’ 2010 study found that people who built several prototypes in parallel ended up with better final designs, explored more divergent ideas and reported greater confidence than those who iterated on a single design. That is the reason round one asks for two towers rather than one good one. Kapur’s 2008 work on productive failure adds the second half: struggling with a problem before being taught the method can outperform being taught first, provided there is a proper consolidation step afterwards. The two-minute review is that consolidation step. Skip it and the activity is just three collapses.

One claim worth correcting, because it circulates widely. You may have heard that kindergarteners outperform business school graduates at the marshmallow challenge. That comes from Tom Wujec’s talks about his workshop observations, not from a controlled study, and it should be treated as an anecdote rather than evidence. The real research, cited above, supports parallel prototyping and structured failure, which is a narrower and more useful claim.

How to Explain It to Your Kid

Ages 5–8: flat falls over, rolled stands up

Hand them one flat card and ask them to stand a book on it. It folds. Then roll the same card into a tube, tape it, and try again. It holds. That single contrast is the whole structural lesson at this age, and once they have it, the five-minute rounds become about using it.

Ages 9–12: name the failure before rebuilding

Teach three words: fold, tip and peel. After every collapse, ask which one it was. Folding means the column was too weak, tipping means the base was too narrow, peeling means the joint failed. Each diagnosis points to a different fix, and matching the fix to the failure is the core engineering skill on display here.

Ages 13+: make it quantitative

Have them measure how much mass one rolled card can hold, in grams, and then design a tower whose columns are each loaded well below that figure. Introduce a safety factor: design for half the measured failure load. Then push the height until it fails anyway, and see whether it failed where the maths said it would.

The question to ask: “If I gave you forty cards instead of twenty, would your tower be twice as tall? What would actually limit it?”

Why Teaching This Early Pays Off

Children get very few chances to see their own improvement as a measurement. School mostly measures outcomes against a standard, not progress against your own last attempt. Three recorded heights in forty minutes is the cheapest possible version of that, and children tend to find it unexpectedly motivating because the comparison is with themselves.

It also builds a diagnostic habit. The question “what exactly failed” is more useful than “it fell over”, and naming the failure mode is a skill that transfers directly to debugging code, fixing a bike, and working out why a recipe did not rise. The National Research Council’s framework places designing solutions and optimising them through iteration among the core engineering practices, and a card tower is about the lowest-cost way to practise both.

What this does not do is prove anything about engineering aptitude, and one afternoon of towers will not change a trajectory. What it reliably produces is a child who, next time something collapses, asks where rather than why me. Pair it with an engineering design notebook if you want the records to accumulate over months.

Before You Start: What Parents Should Know

Best age	6–13. At 6, shorten rounds to five minutes and drop the ball test. At 12 and up, add the safety-factor version
Supervision	Light. The adult’s job is the timer, the tape ration and insisting on the review between rounds
Real hazards	None beyond normal care. A collapsing card tower is harmless. The real risks are scissors and the temptation to stand on a chair to measure a tall tower, so measure from the side with a tape instead
Don’t	Don’t allow extra tape mid-round or let a tower lean on a wall, because both quietly destroy the comparison between rounds. Don’t rebuild the same design three times and call it iteration

When It Doesn’t Work

• Every round gives the same height. The child is rebuilding the same design. Require one named change in advance of each round, and ask them to say out loud what it is.
• The tower tips rather than folds. The base is too narrow for the height. A rough working rule for a free-standing card tower is a base at least one third of the planned height.
• Joints peel apart. Tape sticks poorly to card edges. Fold a small tab on each column end and tape the flat of the tab, not the edge.
• The child gets upset at a collapse. Reframe before it happens, not after. Say at the start that the tower is supposed to fall in round one, and that the job is to notice where. Then treat the collapse as the data point it is.
• Cards are too floppy to hold a roll. Humidity. Index cards absorb moisture and lose stiffness noticeably. Use fresh cards from a sealed packet, and in a humid room expect lower heights across every round.

Take It Further

1. Load capacity instead of height. Same cards, same tape, but the goal becomes holding the most mass at a height of exactly 30 cm. Different constraint, completely different designs, and it introduces the idea that a specification shapes a solution.
2. Material comparison. Run identical rounds with index cards, printer paper and thin cereal-box card. Measure heights and work out height per gram of material, which is roughly how structural efficiency is actually compared.
3. Build a shake test. Place the tower on a tray resting on two marbles and nudge the tray. Towers tall enough to win on height often fail immediately, which is why real tall buildings are designed for lateral loads rather than weight alone. For a materials-first version of the same challenge, the spaghetti and marshmallow tower uses sticks instead of sheets.

What to Watch For Over the Next 3 Months

• Week 4: Rolling things into tubes. Paper, foil, a magazine. The rolled-tube insight is sticky and it shows up in unrelated crafts.
• Month 2: Naming failures. “The joint came apart” rather than “it broke”. That precision is the habit you were building.
• Month 3: Asking for another go, with a stated change. If your child starts saying what they will do differently before trying again, iteration has become a method rather than a word.

Frequently Asked Questions

How tall should a 20-card tower actually get?

There is no official record to chase, and chasing one misses the point. Most children reach 20 to 40 cm in a first attempt and 55 to 90 cm by a third attempt with the same materials. The useful number is your child’s own improvement between rounds, not a benchmark from someone else’s kitchen.

Is tape cheating?

No, but ration it and cut it before the timer starts. Unlimited tape turns the activity into a tape sculpture and removes the structural problem entirely. Thirty centimetres, pre-cut into six pieces, forces decisions about where connections matter most.

Can we do it without index cards?

Yes. Cut thin card or cereal boxes into 7.6 by 12.7 cm rectangles and you have the same material problem. Keep the size identical across rounds, because card dimensions change which designs work.

My child builds one tower and refuses to start again. What now?

Make the second round a different challenge rather than a repeat. Ask for the tallest tower that can hold a ping-pong ball, or the tallest using only ten cards. A changed goal licenses a new design without it feeling like their first attempt was rejected.

---

About the author

Ricky Flores is the founder of HiWave Makers and an electrical engineer with 15+ years of experience building consumer technology at Apple, Samsung, and Texas Instruments. He writes about how kids learn to build, think, and create in a tech-saturated world. Read more at hiwavemakers.com.

---

Sources

1. Dow, S. P., Glassco, A., Kass, J., Schwarz, M., Schwartz, D. L., & Klemmer, S. R. (2010). “Parallel prototyping leads to better design results, more divergence, and increased self-efficacy.” ACM Transactions on Computer-Human Interaction, 17(4), 1–24. https://doi.org/10.1145/1879831.1879836
2. Kapur, M. (2008). “Productive Failure.” Cognition and Instruction, 26(3), 379–424. https://doi.org/10.1080/07370000802212669
3. National Research Council. (2012). A Framework for K-12 Science Education. National Academies Press. https://doi.org/10.17226/13165
4. Roylance, D. (1999). Mechanics of Materials (3.11). MIT OpenCourseWare. https://ocw.mit.edu/courses/3-11-mechanics-of-materials-fall-1999/
5. NASA. (1968). Buckling of Thin-Walled Circular Cylinders (NASA SP-8007). NASA Technical Reports Server. https://ntrs.nasa.gov/citations/19690013955
6. White, R., & Gunstone, R. (1992). Probing Understanding. Falmer Press.