11.3 105-Minute Exam Timing, Flagging, and Review Workflow
Key Takeaways
- Sixty questions in 105 minutes creates an average pace of 1.75 minutes per item, so flagged-question discipline matters.
- Scenario questions should be triaged by layer: user, object, field, record, data, report, automation, application, or Agentforce trust boundary.
- Multiple-select items require independent evaluation of each option against the prompt, not selection of every true Salesforce statement.
- Because the exam is closed reference, final timing practice must be done without notes, browser tabs, Trailhead pages, or outside help.
Managing the 105-Minute Attempt
The opened Salesforce guide in the source brief lists 60 multiple-choice or multiple-select questions and 105 minutes. That creates an average of 1.75 minutes per item. The average is useful, but it is not a command to spend exactly the same time on every question. Some items will be quick if the layer is obvious. Others deserve a slower read because the wording asks for the best administrative action in a mixed security, data, automation, or application scenario.
Build a timing plan before test day. A practical target is to complete a first pass with enough time to revisit flagged questions. For example, aim to be near question 20 by about 35 minutes, near question 40 by about 70 minutes, and through question 60 with roughly 15 minutes left. Adjust to your own style during practice, but do not discover your timing plan during the real appointment.
| Exam phase | Target behavior | Risk to avoid |
|---|---|---|
| First pass | Answer clear items and flag uncertain ones | Turning one hard item into a five-minute drain. |
| Scenario triage | Identify the affected layer before comparing answers | Fixing object access when the issue is record sharing. |
| Multi-select handling | Test each option against the exact prompt | Selecting true but irrelevant statements. |
| Review pass | Reopen flagged items and check wording | Changing answers from anxiety without new evidence. |
| Final minute | Confirm every question has an answer | Leaving accidental blanks if the interface allows them. |
The first read should classify the scenario. Is the problem about a user entering data, a user seeing data, a report returning rows, a dashboard presenting aggregates, a flow firing unexpectedly, an import failing, a page not showing a field, a queue routing work, or an Agentforce trust boundary? Classification reduces panic. Once you know the layer, many distractors become easier to remove.
For access questions, separate app access, object permissions, field-level security, record sharing, page layout visibility, and report or dashboard access. For data questions, separate import tool choice, external IDs, validation rules, duplicate rules, backup exports, ownership, and delete behavior. For automation questions, separate manual process, Flow, approval, assignment rules, escalation rules, and whether the requirement is stable enough to automate. For Agentforce, separate user need from data permission, grounding, testing, monitoring, and inappropriate AI use.
Flagging Workflow
- Answer immediately if the scenario layer and best action are clear.
- If two options remain plausible, flag the item and write the mental distinction in a few words.
- If the prompt is long, identify the business goal, constraint, and current failure before reading options again.
- For multiple-select items, evaluate each option as yes or no against the prompt, then verify the required count if the interface provides one.
- During review, change an answer only when you find a concrete reason, such as a missed word or a layer mismatch.
Words such as best, most appropriate, least effort, most secure, and maintainable matter. An option may work technically but be too broad. Giving View All Data to solve one report visibility complaint is usually excessive. Rebuilding an object model to solve a page-layout issue is usually excessive. Using Agentforce where deterministic validation or a guided flow is required may create trust and support risk. The exam often asks for the administrative choice that fits the requirement with the least unnecessary blast radius.
Multiple-select questions create a different trap. Do not treat the answers as a popularity poll of true Salesforce facts. If the prompt asks for two actions to let a user edit a field on visible records, object edit permission and field edit permission may matter, while role hierarchy may not if records are already visible. If the prompt asks why a dashboard shows more data than the viewer expects, the dashboard running user may matter even though the viewer's own record access is also a real concept.
Practice timing under closed-reference conditions. The official guide states that no hard-copy or online references are allowed during the exam. That means no notes, Trailhead tabs, documentation searches, phone messages, browser lookups, or personal reference sheets. Final practice sets should match that reality. If every practice answer depends on opening Setup docs, your timed score is measuring lookup habits rather than exam readiness.
Use review time for quality control, not wholesale second-guessing. Re-read flagged prompts for qualifiers: all users, selected users, external users, before save, after save, visible records, report folder, dashboard running user, private sharing model, required at all entry points, online or test center, or current registration details. Look for the exact pain point. Then ask whether your chosen answer changes the correct layer and respects least privilege, data quality, maintainability, and Salesforce exam security rules.
The exam is not a place for outside content. Salesforce prohibits sharing, using, or requesting certification exam questions and answers. This timing workflow assumes original practice and official study, not memorized dumps. Memorized alleged answers are fragile because scenario wording changes the administrative layer. Professional preparation builds the reasoning to handle a new scenario ethically.
A candidate is 25 minutes into a 105-minute practice set and has answered only 8 of 60 questions because they keep debating uncertain items. What adjustment is strongest?
In a multi-select item, one option states a true Salesforce feature but does not solve the scenario. How should the candidate treat it?
A prompt says a user can open an account report but cannot see some rows another user sees. What layer should be considered before changing report formatting?