To advance the progress completion percentage in a SmartBook assignment, only one question per concept must be answered correctly because McGraw Hill's adaptive engine counts coverage of concepts, not raw question volume; once a concept is marked as mastered, additional correct answers on that same concept do not raise the percentage further. The meter is a coverage score: it rises as distinct concepts cross the mastery threshold and stalls when you keep drilling concepts you have already cleared.
What the percentage actually measures
SmartBook is an adaptive reading and study tool built on McGraw Hill's LearnSmart engine. The percentage you see is not a quiz score. It is an estimate of how much of the assigned concept set the system believes you have learned, weighted by the number of distinct concepts in the assignment (McGraw Hill Education, SmartBook overview).
The engine tags each question to one or more learning objectives or concepts drawn from the chapter. When you answer a question correctly with sufficient confidence, the system updates its mastery estimate for that concept. Once that estimate crosses the mastery threshold, the concept is treated as covered. Your percentage is essentially mastered concepts divided by total concepts, adjusted for partial credit on concepts you have partially demonstrated.
This is why the "one question per concept" behavior appears. Answering ten questions on the same concept correctly does nothing beyond the first mastery event. Answering one question correctly on each of ten different concepts moves the meter ten times. The lever is breadth of correct coverage, not repetition.
Why the engine resists grinding
The adaptive algorithm uses a spaced-repetition model related to the Ebbinghaus forgetting curve. Each concept carries a memory-strength estimate that decays over time and strengthens with successful recall at spaced intervals. When you answer correctly, the engine may schedule that concept for later review rather than immediate repetition.
The practical consequence is counterintuitive. If you answer a concept correctly once and then never see it again, the system may still schedule a recheck later in the assignment to confirm retention. If you answer incorrectly, the concept returns sooner and more often. The engine is allocating your question budget toward concepts it believes are weak.
A second mechanism reinforces this. The engine tracks confidence separately from correctness. Rapid-fire correct answers can be treated as lower-confidence evidence than answers that take normal time, and some deployments ask you to self-rate confidence before submitting. High-confidence correct answers advance mastery faster; guessed correct answers advance it more slowly or trigger a recheck.
Plain language: the system is trying to find out what you don't know, not to reward you for re-proving what you do.
Quantified context: how much each answer is worth
The exact point value per concept varies by assignment size and publisher configuration, and McGraw Hill does not publish a fixed conversion. What can be stated qualitatively and with ranges from user-reported behavior:
Small assignments of roughly 20 to 40 concepts tend to move in visible jumps, often several percentage points per newly mastered concept. Large assignments of 100 or more concepts move in smaller increments, typically around 1 percentage point or less per concept. Partial mastery on a concept can register as partial progress, so the meter may creep without a full concept clearing. Re-answering a mastered concept correctly generally produces no change in the percentage.
These are behavioral ranges, not manufacturer-published constants. Treat any specific "points per question" claim as unverified unless it comes from your instructor's configuration. The safe conclusion is directional: more distinct concepts cleared per session means faster percentage growth.
Edge cases that stall the meter
Several situations make the percentage appear stuck even when you are answering correctly.
You are re-drilling mastered concepts. The engine keeps serving questions on concepts you have already cleared, especially near the end of an assignment, and correct answers there are inert for the percentage.
You are answering correctly but with low confidence flags. If the system records a guess, it may withhold mastery credit and schedule a recheck.
You are hitting the same learning objective through multiple questions. Several questions can map to one concept, so ten correct answers might cover only two concepts.
You are in a review phase. Some assignments include a reassessment pass at the end that can temporarily depress the displayed percentage as the engine re-tests decayed concepts.
You are misreading partial credit. A concept answered correctly once but flagged for later review may show as in progress, not mastered, so the meter lags your perceived effort.
Red flags to watch:
Percentage unchanged after five or more correct answers in a row. The same concept reappearing repeatedly after correct answers. Question count climbing while percentage is flat. Self-rated confidence prompts appearing frequently. A sudden percentage drop late in the assignment.
Misconceptions worth correcting
The most common false belief is that answering more questions always means more progress. The engine's coverage model breaks that assumption. Volume without breadth is wasted effort.
A second misconception is that guessing strategically is harmless. Correct guesses can advance the meter, but the confidence tracking and recheck scheduling tend to claw that progress back, and the retained knowledge is weaker for exams.
A third is that the percentage equals exam readiness. It measures concept coverage, not depth or transfer. A student at 100 percent has touched every concept but may still struggle with synthesis questions.
A fourth is that the meter can be rushed in a single sitting. Spaced repetition scheduling means some concepts are designed to resurface, and compressing the timeline can leave concepts in a partial state.
The professional caveat: instructors can see completion and time-on-task data, and some configure the assignment so that the percentage only counts toward a grade at a deadline. Chasing the meter without learning the material is visible in the analytics and counterproductive for the exam.
Comparison: strategies and their effect on the meter
| Strategy | Effect on percentage | Effect on retention | When to use |
|---|---|---|---|
| Answer one question per new concept, move on | Fastest coverage growth | Moderate | Early in a large assignment |
| Re-drill mastered concepts repeatedly | Near zero | Slight | Avoid; wasted budget |
| Slow down and answer with confidence | Moderate, steadier | Strong | Concepts you find hard |
| Guess quickly to clear the meter | Temporary gain, often reversed | Weak | Avoid |
| Use "read" and highlight features before answering | Slower start, faster later | Strong | Dense chapters |
| Review flagged and decayed concepts first | Recovers stalled meter | Strong | Mid to late assignment |
Timeline: how a typical assignment progresses
| Phase | What the engine does | What you should do |
|---|---|---|
| First 10 percent | Samples broad concepts, few repeats | Answer honestly, don't rush |
| 10 to 40 percent | Serves new concepts, occasional rechecks | Prioritize breadth |
| 40 to 70 percent | Mixes new and decayed concepts | Clear flagged items |
| 70 to 90 percent | Heavier recheck load, slower meter | Slow down, confirm mastery |
| 90 to 100 percent | Sparse new concepts, review pass | Expect a temporary dip |
| Final review | Re-tests decayed concepts | Treat as exam prep |
The dip near the end is normal. The engine is confirming retention, not punishing you.
Practical actions that move the meter
Work in concept-first passes. Skim the assignment's learning objectives if your course shows them, then answer one question per objective before returning to any concept.
Answer deliberately. Take normal reading time per question; rushed correct answers can be logged as lower-confidence evidence.
Read the explanations after wrong answers. The engine resurfaces missed concepts, and understanding the rationale shortens the recheck cycle.
Prioritize flagged and decayed concepts. When the engine offers a review, take it; clearing decayed concepts restores percentage faster than new coverage late in an assignment.
Stop drilling after mastery. If a concept has been answered correctly and does not reappear, leave it alone.
Use the assignment in spaced sessions when time allows. Two or three shorter sittings align better with the retention model than one marathon.
For thresholds: if the percentage has not moved after roughly five consecutive correct answers, assume you are re-drilling mastered concepts and switch to unfamiliar material. If it drops late, treat it as a review pass rather than an error.
Limitations and what the meter cannot tell you
The percentage is an internal estimate, not a validated measure of competence. McGraw Hill does not publish the exact mastery threshold, the point weighting, or the confidence algorithm, so any precise "X correct answers equals Y percent" rule is inference, not documentation.
The engine's behavior also varies by publisher configuration, course setup, and instructor settings. Two students in different courses can see different meter dynamics on identical effort.
Finally, coverage is not comprehension. The meter can reach 100 percent while weak spots remain in application and analysis. Use the assignment as a map of what to study, not as proof of readiness.
This article is general educational information about how an adaptive learning tool behaves, not a guarantee of grades or outcomes specific to your course. Assignment configurations vary, so check your instructor's rubric and the platform's help documentation for the settings that apply to you.