Start with control, not speed
The fastest way to waste Agoda's reported 5 days completion window is to begin with full timed simulations. A simulation measures the habits you already have; it does not automatically improve them. First identify why answers go wrong, then introduce the 36 minutes constraint.
For analytical roles using Verify G+, prepare for numerical, deductive and inductive reasoning in one adaptive assessment. The fact sheet allows up to 30 questions in 36 minutes. One Agoda candidate reported a 24 questions version, answered 18 questions and still progressed. Those numbers should shape your strategy: accuracy and adaptive performance matter more than forcing a fixed completion count.
My strongest recommendation is to spend the first 40% of preparation time untimed. That is contrarian because countdown clocks feel like the obvious threat. Yet speed work only helps after you have a repeatable method worth speeding up.
Verify your exact route before practising
Agoda does not give every applicant the same assessment. Its official careers guidance separates role families:
- Non-Tech applicants may receive an aptitude assessment.
- Marketing and Product can be assessed on business analytical acumen; Marketing may add a recorded communication stage.
- Supply, People, Finance and Legal can be tested on stakeholder management, data literacy and job expertise, with numerical, statistical and logical thinking.
- Strategic Partnerships can involve graphs, charts and logical reasoning.
- Tech roles may receive coding tests in Python, SQL or other role-relevant technologies instead of Verify G+.
- Customer Service assessment can cover language, communication, problem solving, numerical thinking and occupational personality.
Read the assessment name in your invitation before buying or opening a question bank. If it says Verify G+ or General Ability, mixed reasoning is the relevant route. If it names a coding platform or a job simulation, a generic plan for the wrong format is the wrong preparation tool.
Also confirm the deadline. Agoda recommends completing the assessment within 5 days, and the link can reportedly take 1 to 2 hours to become active after issue. Do not treat the send time and the usable start time as identical if the invitation says activation is pending.
Build an error taxonomy in 60 minutes
Complete a short untimed diagnostic and label every miss. Do not write “careless” beside all of them; that word hides the fix. Use categories that point to an action.
For numerical items, useful labels are unit error, wrong denominator, percentage-direction error, table lookup, multi-step arithmetic and unsupported inference. For deductive items, use reversed condition, confused “must” with “could,” ignored exception and introduced outside knowledge. For inductive items, record wrong feature, rule fits only 2 frames, direction error or failure to test the distractor.
A compact log can have 4 columns: item type, chosen method, error label and correction. After 20 questions, patterns become visible. If 6 errors involve the denominator, more random questions will reproduce the same weakness. Ten minutes relearning percentage bases is more valuable.
The insider point is that review quality separates productive practice from busy practice. Solving 100 questions and reading only the answer letter builds recognition, not reasoning. Reviewing 20 questions until you can explain every option produces a reusable method.
Numerical reasoning: write the relationship first
Agoda's official assessment guide specifically mentions graphs, charts, numerical thinking, statistical thinking and data literacy across several non-Tech teams. Expect the numbers to sit inside a business context, not appear as isolated arithmetic.
Before calculating, write a 3-part note: target, relationship, unit. If the question asks for percentage growth, the relationship is change divided by original; the unit is percent. If it asks for revenue per booking, divide revenue by bookings and keep the currency. This takes seconds and blocks many denominator mistakes.
Use estimation before exact calculation. If revenue rises from roughly 80 to 100, growth is around 25%, so an option near 60% deserves immediate suspicion. Estimate first, calculate second, then verify that the result answers the asked direction.
Practise table scanning deliberately. Read the row label, column label and unit header in that order. A table marked “USD thousands” changes an answer by 1,000 units. On multi-period charts, point to the correct year before extracting the value.
Your calculator should reduce clerical work, not replace structure. Enter one clean expression and perform a magnitude check. Repeatedly tapping intermediate figures increases transcription risk and consumes time.
Work a full example the way you would under the clock. Suppose a table shows quarterly bookings of 1,240, 1,310, 1,455 and 1,600 (in thousands), and the question asks for the percentage growth from Q1 to Q4. Target: percentage growth. Relationship: (new minus original) divided by original. Unit: percent. That gives (1,600 minus 1,240) divided by 1,240, or roughly 29%. Before selecting an option, sanity-check the direction — growth should be positive and the answer should sit near your estimate, not an order of magnitude away. If the closest option is 2.9% or 290%, you have misplaced a decimal or divided by the wrong base, and it is faster to redo the 3-part note than to trust a number that does not match your estimate.
Deductive reasoning: separate must from might
Deductive questions punish plausible storytelling. The answer must follow from the stated rules even if the real world suggests something else.
Translate each statement into a minimal rule. “All A are B” allows A to B; it does not allow B to A. “No C are D” forbids overlap. “Some E are F” establishes at least one shared member but says nothing about the rest. When a conclusion adds a relationship that was never stated, reject it.
Try the counterexample test. If you can imagine 1 arrangement that obeys every premise but makes the conclusion false, the conclusion is not logically necessary. This technique is particularly useful when 2 options both sound reasonable.
For ordering or scheduling problems, draw only the constraints that matter. Six names and 5 rules can look dense in prose; a short sequence or grid makes contradictions visible. Do not rewrite the entire prompt.
Under time, distinguish uncertainty from complexity. If the reasoning chain is clear but arithmetic or placement is long, continue. If you still do not know what relation is being tested after a focused read, make the best supported choice and preserve time for later items.
Inductive reasoning: test one feature across every frame
Visual patterns invite overfitting. A candidate spots rotation in the first 2 frames and commits before noticing that shape count also alternates. Use a fixed feature scan: count, position, direction, fill, size and interaction.
Start with the simplest rule that explains all frames. Test it against every transition. If it fails once, revise it; do not invent an exception merely to protect the first idea. Common structures include alternation, addition or subtraction of elements, movement by a fixed number of positions, independent changes in 2 features and combinations of previous panels.
When answer options are visible, use them as experiments. If 2 choices differ only in fill, settle the fill rule rather than re-solving position. This is not shortcut reasoning; it is efficient hypothesis testing.
Spend 15 minutes practising verbal explanations of visual rules. “Black triangle moves 2 places clockwise while the dot alternates” is more reliable than “it feels like B.” Clear language makes the rule easier to verify under pressure.
Switch between the 3 abilities on purpose
Verify G+ is mixed, so the hidden challenge is changing mental mode, not just solving each domain. Practising 40 numerical questions in a row inflates confidence because it removes that switch cost. Build mixed sets of 12 questions instead, naming the mode — calculate, prove, or infer pattern — before each item. Run at least 2 mixed timed sets before the live assessment; a third is only worth it if you have time to review it properly.
A realistic 5-day preparation schedule
Use Agoda's recommended 5 days window as a planning boundary, not a reason to delay until the final evening.
Day 1: diagnose
Confirm the assessment name and complete 20 questions untimed across the relevant domains. Build the error log. Choose the 2 error patterns costing the most marks.
Day 2: repair numerical method
Spend 45 minutes on ratios, percentages, rates, tables and charts. Finish with a short timed block, then review every wrong or guessed response.
Day 3: repair logic and patterns
Divide 60 minutes between deductive and inductive work. Explain each answer aloud or in one sentence. Add new error labels only when they point to a different correction.
Day 4: rehearse mixed conditions
Complete 1 mixed simulation under the published timing for your exact test. Review for at least as long as you tested. Decide on one pacing rule, such as moving on when an item stops producing useful information.
Day 5: stabilize
Do a light 20 minutes warm-up, re-read the invitation and prepare the environment. Do not cram a new technique immediately before the assessment. Begin while rested and well before the deadline.
If you have only 2 days, compress Days 1 and 2 into the first day and use the second for a mixed run plus review. Never skip the invitation check.
Set up the assessment environment
Agoda's published advice is straightforward: minimize distractions and use a stable internet connection. Apply that literally.
Choose a desk, connect power, silence notifications and close heavy downloads. Have permitted scratch materials ready before starting. Open the link early enough to handle login or compatibility issues without spending deadline time. Agoda says its assessments typically last from 30 minutes to 1 hour, while the reported Verify G+ format itself allows 36 minutes.
Read every instruction screen. Confirm whether you can return to previous items, whether a calculator is supplied or permitted and whether examples are timed. Do not transfer rules from a practice site to the official session.
Use the practice round if offered. Its value is interface fluency: learning how selections register, where progress appears and how to submit. Saving even 5 seconds per interaction protects attention for reasoning.
Pacing without chasing a completion number
The 30 questions ceiling over 36 minutes suggests an average near 72 seconds per question, but adaptive items vary. Treat that figure as an awareness tool, not a per-item command.
Use checkpoints based on attention rather than a rigid quota. Every few items, ask whether you are still making evidence-based decisions. If one problem consumes several minutes, its opportunity cost is larger than its position on screen suggests.
The candidate who advanced after answering 18 questions in a 24 questions session illustrates that unfinished does not automatically equal failed. It does not mean slow is always safe. The defensible lesson is to avoid trading several solvable later items for one stalled item.
Accuracy is also not an excuse for perfectionism. Make the strongest justified selection available and move. The right balance comes from rehearsal, which is why your final mixed practice should include the same time pressure as the real invitation.
What not to copy from generic advice
Ignore any universal claim that Agoda demands a specific percentile or number correct. No public company-wide cutoff is verified. Ignore advice to practise only numerical reasoning if your invitation names General Ability; the reported G+ route includes 3 abilities. Ignore a software candidate's HackerRank plan if you received a reasoning assessment, and vice versa.
Also reject the idea that more material is always better. In a 5 days window, switching among 6 providers can create interface noise and contradictory explanations. Choose one clear source for method, official examples for familiarity and a small error log for continuity.
Preparation should narrow your decisions. By the final day, you should know how you start each item type, when you stop an unproductive path and what mistake you check before submitting.
Related Agoda guides
Provider & company guides
Frequently asked questions
Can I prepare properly if Agoda gives me only 5 days?
Yes. Use Day 1 for a 20 questions diagnostic, Days 2 and 3 for targeted repair, Day 4 for a mixed simulation and Day 5 for a light review. The short window rewards focus more than volume.
Should I try to answer all 30 questions?
Verify G+ is described as having up to 30 questions in 36 minutes, but adaptive delivery can vary. One Agoda candidate progressed after answering 18 questions in a 24 questions session, so a fixed completion target is not a verified pass rule.
How many full practice tests should I take?
For a 5 days plan, 2 full mixed simulations are usually enough if each receives detailed review. Add a third only when it tests a specific pacing change rather than repeating the same errors.
What should I do if the Agoda link is not active?
The link can reportedly take 1 to 2 hours to activate. Recheck the invitation, wait within that stated window and contact the recruiter if the issue persists rather than letting the 5 days deadline approach.
Is numerical practice enough for an Agoda analytical role?
Not when the invitation names Verify G+ or General Ability. The format combines 3 abilities—numerical, deductive and inductive—and can run for 36 minutes, so mixed switching practice matters.
See TestSolve in action first
Watch a 2-minute walkthrough, then download when you're ready. No subscription, no signup.
TestSolve is independent and not affiliated with Agoda or any employer or test provider named on this page. All product names and trademarks belong to their respective owners.