Big benefit, small risk. This bias appears even without bad intent. Designing balance is the skill of keeping good and cautionary information at equal weight, down to word count, placement, and the reader's eye path. This time we face the psychology of evasion through silence.
Level the two pans of the scale
Picture a butcher weighing meat on a balance. Slip a finger quietly under one pan, and equal weights look unequal. The same thing happens in material design. Efficacy data gets large type and wide space; the side effect sits in a corner as one small line. You may think you placed no finger, yet the design of layout and word count tips the scale.
Designing balance is the skill of placing benefit (the good side) and side effect (the caution) so they reach the reader with equal weight. Here, weight is not only factual accuracy. It is type size, position, order, and number of lines. Whatever hits the reader's eye first is the heavy thing. Fair balance means balancing this visual weight too.
Loading the correct data is not balance. Balance happens only when it reaches the reader's eye at equal weight.
Tipping by not writing
Imagine a health check result sheet. Good numbers line up in bold; the flagged item gets a tiny note saying only "consider re-testing." Not one lie. Yet the reader concludes "mostly healthy." This is the danger of omission — producing a result by not acting.
In reported cases, a presenter prepared no material at all for the primary endpoint (the result decided up front as most important) and explained only a secondary endpoint (a side result) that showed a significant difference (a difference hard to explain by chance). No false figure was shown. They simply did not present the important one. In another case, a screening test (a sifting examination) was mandatory before dosing, yet the product summary stated "requires no screening or testing" as if it were a product strength. A safety step was actively erased and shown as a benefit.
What both share is this: they did not raise the volume of the benefit — they lowered or removed the volume of the caution. They did not lift the right pan; they quietly pulled a weight from the left pan. Tipping by subtraction is harder to notice than exaggeration by addition, which is exactly why it must be the first target of balance design.
Why makers shrink the cautions
No one studying for a driver's license practices only the accelerator and leaves the brake for later — it leads straight to crashes. Yet in material design, a bias appears: polish the benefit (accelerator), postpone the side effect (brake). Why?
A maker's psychology drives this. One force is the sin of omission. Speaking and showing are actions; not speaking and not showing feel like "doing nothing," so guilt is faint. "I wrote no lie" becomes a shield. Another force is motivated reasoning: when the desired conclusion comes first, cautions start to look like things that "need not be stressed." The maker feels sincere — the most troublesome part. The table below lines up reported deviations with the psychology behind them and the design that stops them.
| Reported deviation | Psychology behind it | Design that stops it |
|---|---|---|
| No primary-endpoint material; only the secondary explained | Sin of omission (staying silent on the important one) | Fix the order primary then secondary; place primary first, equal volume |
| Mandatory screening framed as "not needed" | Motivated reasoning (selling the saved effort) | Put contraindications and required tests at the same size and on the same page as benefit |
| Side effect shrunk to one corner line | Local rationalization (just here, make it small) | Keep a "weight ledger" counting word count and area against the benefit side |
Make weight visible as numbers
A proofreader marking up a galley (a pre-print test sheet) measures letter height with a ruler, not by feel. Balance is the same: "roughly the same" is not design. Counting what can be counted is the surest way to prevent tilt.
Concretely, count three things. First, word count: if the benefit runs twenty lines and the side effect runs two, the look has set a tenfold gap before content. Second, placement: what sits on the first page, at the top-left where the eye lands first. Third, order: telling the good results first and tacking the caution small at the end pulls the closing impression toward benefit. These are not preferences; they are measurable facts. Measure, and you can fix.
Balance is not kept by feeling. It is kept by countable rulers — word count, placement, order.
Do not turn non-inferiority into superiority
For an added indication (a new use added to an approved drug), a case is reported where only non-inferiority data (showing the drug is not worse than the existing one) existed, yet one slide of the old indication's comparison data was shown to claim "superior to the existing drug." This is also a balance problem. "Not losing" and "winning" are different things, yet the display design built the latter impression. The very content of the weight — the type of data — was swapped.
This skill deepens across levels L1 to L4. L1 just includes the side effect in this one case as told. L2 learns the procedure of placing benefit and side effect at equal volume as a pattern. L3 understands why balance is needed and can judge not to show non-inferiority as superiority, to present the primary first. L4 builds the "weight ledger" that checks word count and placement as a system, a standard that stays level no matter who makes it. If last time's grounding in sources (returning claims to approved evidence) is the floor, balance is the work of keeping the delivery scale level on top of that floor.
What a Good Creator Brings ── Map of all 10 episodes
- Vol. 1: The Core Question — The Maker Carries Both "Accuracy" and "Clarity" ── An introduction showing that the maker of promotional materials must carry both fidelity to facts and the skill of clear design at the same time.
- Vol. 2: Two Axes for Reading Skill — Fidelity to Facts x Craft of Delivery ── We map the skills of materials-making onto two axes — fidelity to facts and craft of delivery — into four types, and show why persuasive-but-inaccurate work is the most dangerous and why fidelity sets the ceiling for design.
- Vol. 3: The Power to Always Return to the Source: Tying Every Claim to Approved Evidence ── On grounding: can every number, figure, and phrase in a material be traced back to its approved source data, catching secondhand citation and embellishment.
- Vol. 4 (this episode): Designing Balance — Giving Benefit and Risk the Same Weight ── The skill of keeping benefit and caution at equal weight through layout, word count, and the reader's line of sight.
- Vol. 5: The Power to Anticipate Misreading — Imagining How Your Reader Goes Wrong ── The skill of finding, before release, where your reader will misread the material — and heading it off.
- Vol. 6: Persuasion Within the Bounds of Accuracy — Putting a Factual Brake on the Urge to Sell ── The fifth skill: designing persuasion that reaches readers at full strength without exaggeration, keeping a factual brake on while making the message land.
- Vol. 7: The Power to Translate Rules into Form — Turning Regulation from "Forbidden" into Design ── Treating regulation not as a list of bans to memorize but as design guidance — reading the reason behind each rule and turning it into how a material is built, across four levels.
- Vol. 8: The Power to Review Yourself First — Become the Strictest Reviewer Before You Submit ── The ability to doubt your own work and become its strictest reviewer before anyone else does.
- Vol. 9: The Power to Take Feedback ── Turning a Rejected Draft into Precision, Not a Verdict on You ── A returned draft is information that sharpens accuracy, not a judgment of your worth; separate feeling from fact and turn each comment into a future standard.
- Vol. 10 (final): Building Trust ── Toward "This Person's Materials Are Safe", and the Integration of All the Skills ── Trust is built by accumulation, not by a single good piece. When reviewers and requesters come to feel "this person's materials always trace back to the source", checking gets lighter and the eight skills work as one. The final installment.
Balance is most often missed by people who tell no lies, because most tilt comes from subtraction — not writing, shrinking — and stirs no sense of guilt. So do not rely on feeling: hold countable rulers of word count, placement, and order. Before releasing, the maker measures whether benefit and side effect sit on the scale at equal weight. That is the surest brake against the quiet tilt of omission.
- Suspect tilt by subtraction. Balance breaks less from exaggeration than from omission — not writing, shrinking. With no lie present, even the maker barely notices.
- Guard weight by counting. Word count, placement, and order are measurable facts. Balance benefit and caution by numbers, not feel.
- Stop the swap of kind. Do not show non-inferiority as superiority or a required test as "not needed." Deliver the data's content at equal weight too.
- Ministry of Health, Labour and Welfare, Compliance and Narcotics Division (commissioned project). Report on the Monitoring of Promotional Information Activities for Prescription Drugs (March 2024 and prior years). Flagged cases are published with company names anonymized; the deviation patterns cited here are generalized from this report.
- Ministry of Health, Labour and Welfare. Guidelines on Promotional Information Activities for Prescription Drugs. The principle of fair balance in presenting efficacy and safety.
- Japan Pharmaceutical Manufacturers Association. JPMA Code of Practice. Principles of accuracy and fairness in information provision.
- Ministry of Health, Labour and Welfare. Standards for Fair Advertising of Drugs and Related Products. The requirement to describe efficacy and side effects in proportion.