My CAT Mock Scores Are Not Improving — What Should I Actually Change?
There is a particularly frustrating stage in CAT preparation where you are clearly studying more, but your mock score refuses to move.
You revise Quant. You solve RCs. You watch DILR solutions. You take another mock.
The result looks almost identical.
This is usually the point where students respond by adding more material.
That can be exactly the wrong response.
A flat score does not automatically mean you need more knowledge. It often means you have not identified the constraint that is limiting your score.
Reason 1: You are analysing answers, not decisions
A typical mock review looks like this:
Wrong answer → read solution → “understood” → next question.
That is solution checking, not mock analysis.
For every wrong or unattempted question, ask:
- Did I know the concept?
- Was the question worth attempting?
- At what point should I have left it?
- Did I misread something?
- Did I calculate incorrectly?
- Did I choose the wrong method?
- Was I rushing because of a previous time loss?
The answer tells you what to change.
Reason 2: You are trying to increase attempts before improving selection
“Attempt more questions” sounds like progress.
But CAT is not a volume contest.
Suppose Student A attempts 20 questions at 60% accuracy while Student B attempts 15 at 85% accuracy. The second student may generate a much better score despite attempting fewer questions.
The important question is not simply “How many did I attempt?”
It is:
How many of my attempts were good decisions?
Start tagging attempts as:
- Easy conversion
- Reasonable attempt
- Risky attempt
- Should have skipped
If the last two categories are large, the first improvement may come from attempting less, not more.
Reason 3: One bad DILR set is destroying the section
DILR is brutally sensitive to selection.
You can be perfectly capable of solving two sets and still end with a poor score because you invested the first 12 minutes in a set that never opened up.
When analysing DILR, do not only resolve every set after the mock.
Recreate the first three to five minutes.
What signals suggested that the set was manageable or dangerous?
Could a table have simplified it?
Was there a clear starting point?
Were there too many cases?
The skill you are training is not only set solving. It is set selection under incomplete information.
Reason 4: VARC errors are being treated as random
If your VARC accuracy swings heavily, look for patterns.
Are you getting inference questions wrong?
Are extreme options attracting you?
Are you selecting options that sound intelligent but are not supported by the passage?
Are you reading too quickly after falling behind?
Do not write “RC mistake” in your error log.
Write the actual reason.
For example:
“Selected an option that was partly true but answered a different question.”
That sentence is useful in the next mock.
Reason 5: QA preparation is chapter-wise, but CAT is mixed
At home, you know you are solving Time & Work.
In CAT, the exam does not tell you which tool to use. You first have to recognise the structure, estimate difficulty and decide whether the problem deserves your time.
That is why mixed timed sets become increasingly important as the exam gets closer.
After concept learning, practise identifying and solving questions without a chapter label.
Reason 6: You keep changing strategy after one bad mock
One poor score can make students redesign everything:
“Maybe I should start with VA.”
“Maybe I should attempt four DILR sets.”
“Maybe I need 22 QA attempts.”
This creates noise.
A strategy should be changed because a repeated pattern supports the change, not because one mock felt bad.
Track at least three or four mocks before making large conclusions.
Reason 7: You have no repair cycle
This may be the biggest one.
Mock → score → analysis → next mock is incomplete.
The missing step is repair.
A better loop is:
Mock → Diagnose → Repair → Retest
If a mock shows that ratios are weak, spend time repairing ratios.
If it shows that you repeatedly waste time on difficult QA, practise quick difficulty identification.
If it shows that reading errors are costing marks, slow down deliberately in selected practice.
Your next mock should test whether the repair worked.
A simple score-stagnation audit
Take your last three mocks and calculate:
- Accuracy by section
- Number of negative marks
- Time lost on abandoned questions
- Number of solvable questions left unattempted
- Number of errors caused by concepts
- Number caused by execution
- Number caused by selection
The largest bucket is your current constraint.
Work on that first.
Match the solution to the reason your score is stuck
If your analysis shows that concepts are mostly known but speed, question selection and time management are repeatedly costing marks, the CAT Speed & Shortcuts Live Program is the OYM course built specifically for that problem.
If the bigger issue is that you are taking mocks but not converting them into a repair plan, the On Your Mocks Test Series is the more natural fit. Its testing philosophy is built around identifying why marks were lost, repairing the weakness and retesting rather than simply moving to the next score.
Neither is automatically the right choice. If your analysis reveals a basic concept gap, fix the concept first. The resource should follow the diagnosis.
Your next mock should have one purpose
Before you begin the next test, write one sentence:
“In this mock I am testing whether I can ______.”
Examples:
- leave bad QA questions earlier
- choose a DILR set faster
- reduce RC overthinking
- keep accuracy above my recent average
- avoid revisiting doubtful questions
That turns a mock from an assessment into an experiment.
And that is usually when scores start moving again.