Appian ACD301 Practice Questions with Explanations
Free Appian ACD301 practice questions. 22 of them, each with the correct answer, a full explanation, and the reason every other option is wrong.
ACD301 practice questions
Question 1
A junior developer is trying to make a reusable object so that a record status will show up as an icon with a specific color.
If no status is designated, then "N/A" should display. They have already created relevant constants for all status values and colors needed. This status will be used both on the record list and within the record summary view.
Which strategy should you recommend?
- A. Create an interface that includes an a!match(), which is used to select a a!iconIndicator() or defaulted a!richTextItem("N/A") based on a rule input made for the value parameter of the a!match().
- B. Create an expression that includes a a!richTextDisplayField(), where a!match() is used to select a a!richTextIcon() or defaulted a!richTextItem("N/A") based on a rule input made for the value parameter of the a!match().
- C. Create expressions for each a!richTextIcon() option with appropriate constants. Use an a!match() in the target interface to choose from the icon expressions or a!richTextItem("N/A"), based on the target data's status.
Show answer and explanation ▾
Correct answer: A
One reusable interface that takes the status as a rule input and uses a!match() to return the matching icon, with N/A as the default, keeps the logic in a single object. The record list and the summary view can both call it.
Why the other options are wrong:
- B. Wrapping a whole display field in an expression is a less flexible building block to place in both the record list and the summary view.
- C. Separate expressions per icon spread the logic out and push the a!match() into every target interface, so nothing is truly reusable.
Question 2
You have an external system where data is exposed through a REST API endpoint. You have created a service-backed record type that connects to this external system, and configured a record data source expression and a sync expression. You have also configured syncing in batches.
When initiating a full sync of the record type for the first time, the sync fails and returns the following error:
“Unable to sync due to an error in the Record Data Source during batch number 1,026.” After performing a direct query with an API analyzer such as Postman, you see that the external system returns an “416 Undefined” error at 1,026, indicating that there are no more records to be synced.
When the API returns a “416” error, how should you handle this error in the record data source expression and ensure the full sync on the record type completes successfully?
- A. Return an empty record type constructor.
- B. Return an empty list.
- C. Return an empty integration response.
Show answer and explanation ▾
Correct answer: B
When the external system answers 416 because there are no more records, the data source expression should return an empty list. An empty list tells the sync that the batch is finished, so the full sync can complete instead of failing.
Why the other options are wrong:
- A. An empty record type constructor is not a list of records, so it does not signal the end of the data.
- C. An empty integration response is not the list the sync expects, so the sync still reports an error.
Question 3
A large manufacturing company is looking to build their first Appian application.
They want to improve efficiency and productivity across various departments, but they also want to start with a manageable project that will quickly demonstrate the value of Appian.
Which idea should you recommend for the company's first Appian project?
- A. A comprehensive enterprise resource planning (ERP) system that coordinates all business processes and data across the company.
- B. A customer relationship management (CRM) system that centralizes customer data and automates sales and marketing processes
- C. A departmental expense approval application that streamlines the expense submission and approval process for employees and managers.
Show answer and explanation ▾
Correct answer: C
A first project should be small enough to finish quickly and show value. A departmental expense approval app has a clear process, a limited group of users and fast payoff.
Why the other options are wrong:
- A. An enterprise-wide ERP system is far too broad to be a manageable first project.
- B. A full CRM covering sales and marketing is a large, cross-department build, not a quick win.
Question 4
A large retail organization recently adopted Appian as an enterprise platform to replace their existing support ticket management system.
The Support Tickets source table has approximately two (2) million rows of data over the current sync limit, with statuses of either “Open” or “Closed.” The organization needs to view, report, and take action on “Open” tickets in Appian, and have the ability to retain access to “Closed” tickets for audit purposes.
Which data structure should you recommend?
- A. Two synced record types, one for “Open” tickets and one for “Closed” tickets, with sync filters enabled.
- B. One synced record type for “Open” tickets, with sync filters enabled, and one unsynced record type for all tickets.
- C. One synced record type on the Support Ticket source table, with the setting Keep data available at high volumes enabled.
Show answer and explanation ▾
Correct answer: B
Syncing only the open tickets keeps the synced record type under the sync limit for the data people act on every day. A second, unsynced record type over all tickets still gives access to closed tickets for audit.
Why the other options are wrong:
- A. Two synced record types still try to sync the closed tickets, and the closed set alone is too large for the sync limit.
- C. One synced type on the whole table keeps all two million rows in the sync, which is the volume problem the company has.
Question 5
You are designing a process model where a user's loan application needs to be simultaneously reviewed by both an underwriter and a credit analyst before moving on to a decision node based on the outcomes of the reviews.
Which two node configurations should you recommend to ensure that both reviews are completed before the process proceeds? (Choose two.)
- A. Activity chaining disabled
- B. Synchronous subprocess
- C. Multiple node instances (MNI)
- D. Activity chaining enabled
Show answer and explanation ▾
Correct answer: B, C
Multiple node instances run the review once per reviewer at the same time, so the underwriter and the credit analyst work in parallel. A synchronous subprocess holds the parent flow until the review work has completed, so the decision node is reached only after both reviews finish.
Why the other options are wrong:
- A. Activity chaining only affects how the next task is presented to a user; turning it off does not make the process wait for both reviews.
- D. Activity chaining moves one user straight to their next task; it does not make the process wait for both reviews.
Question 6
You are managing an employee onboarding process, which involves passing employee custom data types (CDTs) between parent processes and subprocesses, that will return output to a parent process.
You initially passed the CDTs by reference in subprocess. However, a recent update to the employee CDT structure caused some in-progress subprocesses to fail because they expect the new CDT version.
How should you handle this situation going forward?
- A. Cast the data in the subprocess so that versions will align.
- B. Pass the employee CDT by value in the subprocesses
- C. Use a Start Process smart service to run the subprocess.
Show answer and explanation ▾
Correct answer: B
Passing the CDT by value gives each subprocess its own copy of the data and returns results through output mapping. The subprocess no longer shares the parent's variable by reference, so a change to the CDT structure is less likely to break instances in flight.
Why the other options are wrong:
- A. Casting inside the subprocess does not remove the dependency on the parent variable that is passed by reference.
- C. Start Process runs the process asynchronously, so it cannot return output to the parent.
Question 7
Towards the end of a sprint planning session, you discover that the business stakeholders have differing priorities on certain features, leading to confusion about what should be included in the upcoming sprint.
How should you manage this situation to ensure the team can proceed effectively?
- A. Engage the stakeholders in conversation to express their priority concerns and underlying timelines. Then, allow the product owner to decide on the priority.
- B. Prioritize the features that are technically easier to implement, allowing the team to make progress while waiting for the stakeholders to resolve their differences offline.
- C. Proceed with the sprint planning based on the original backlog to ensure the current sprint stays on schedule. Plan to resolve the prioritization conflict in the next sprint.
Show answer and explanation ▾
Correct answer: A
Conflicting priorities are resolved by bringing the stakeholders' concerns and timelines into the open and then letting the product owner decide. The product owner owns the order of the backlog, so the team leaves planning with a clear and agreed scope.
Why the other options are wrong:
- B. Technical ease is not a measure of business value, and the conflict stays unresolved.
- C. Carrying on with the original backlog ignores the conflict and risks building the wrong features.
Question 8
You are reviewing the health check report for your application server. You notice that server heap utilization often averages above 80% over 10minute intervals. Which two factors are most likely to contribute to this elevated heap utilization and can potentially impact performance? (Choose two.)
- A. Web APIs with a high response time on querying the data.
- B. Memory leaks due to inefficient Java plug ins.
- C. Suboptimal process designs using large/complex database queries.
- D. High generation of large objects, such as documents or reports.
Show answer and explanation ▾
Correct answer: B, D
Application server heap is consumed by Java objects held in memory. Plug-ins that leak memory keep objects from being released, and frequent creation of large objects such as documents or reports fills the heap quickly, so both push utilization up and can slow the server.
Why the other options are wrong:
- A. Slow query responses in web APIs point to database or query time, not to application server heap.
- C. Heavy database queries in process designs load the database rather than explain sustained heap use on the application server.
Question 9
You are building a complaints management application for a customer.
When an employee receives a client complaint, the employee will create a new case record in Appian with the complaint details.
Depending on the complaint details, the Legal team may be required to take action on the case. If the Legal team fails to take action on those cases, then the customer will be out of compliance. Currently, only the Legal team is able to assess whether or not a case requires action from their team.
The Legal team is overwhelmed by the high volume of cases that require their assessment to determine if action is required. Compliance issues are increasing, and the Legal team wants suggestions for how Appian can reduce their case assessment workload.
Which action would best assist the Legal team and mitigate compliance risks?
- A. Use a!sentimentScore() on case creation to assist the employee in determining if Legal action is required
- B. Display the Records Chat component to the Legal team so that they can quickly review case summaries and mark cases that require action.
- C. Create an AI Skill to determine if a new case may require Legal action and, if so. assign it to the Legal team for review.
Show answer and explanation ▾
Correct answer: C
An AI skill can classify each new case and flag the ones that may need legal action. Only flagged cases are routed to the Legal team, which cuts their assessment workload and lowers the chance that a case needing action is missed.
Why the other options are wrong:
- A. Sentiment measures tone, not whether a complaint requires legal action.
- B. The Legal team would still have to review every case, so their workload is not reduced.
Question 10
Your customer is a large university that uses an Appian application for student data. The main users of the application are student advisors who examine student metrics and receive biweekly emails about their student's progress. A parent process model has already been built to fulfill this requirement.
The university's new initiative is to improve graduation rates. To do this, they want to update the biweekly email to include a report for each student identified as having a high risk of not graduating.
For example, if a student advisor has 15 students assigned, and two (2) students are identified as high risk, there should be two (2) reports attached to the update email. If no high-risk activities are identified, the process sends the email update without any attachments.
How should you add the smart services nodes to calculate risk and generate the report prior to sending the email?
- A. Create a new synchronous subprocess and insert it into the parent process model.
- B. Create a new asynchronous subprocess and insert it into the parent process model.
- C. Create a new gateway flow in the parent process for the smart services.
- D. Incorporate the smart services directly in the parent process with no extra gateway flow.
Show answer and explanation ▾
Correct answer: A
A synchronous subprocess can run once for each high-risk student, calculate the risk, generate the report and pass the documents back to the parent. The parent waits for it to finish, so the reports are ready before the email node runs.
Why the other options are wrong:
- B. An asynchronous subprocess does not make the parent wait or return the reports, so the email could go out without them.
- C. A gateway flow in the parent does not cleanly handle a varying number of students, including none.
- D. Placing the nodes directly in the parent does not handle a varying number of reports per advisor.
Question 11
You are integrating external data sources in Appian applications to store and process protected personal information.
What should you consider when processing personal information in Appian?
- A. Design a conservative data retention policy of five (5) years for backups to satisfy personal information and business processes audits when needed.
- B. Maintain uniqueness and data integrity by using usernames as identifiers from external systems and make comprehensive updates across all systems when usernames change.
- C. Use of personal information from external data sources should be traced through related Appian objects to understand where else that same data may be stored.
Show answer and explanation ▾
Correct answer: C
Personal information brought in from external sources can end up in process variables, record data, documents and logs. Tracing it through the related Appian objects shows every place it may be stored, which is needed to protect, retain and delete it properly.
Why the other options are wrong:
- A. Keeping backups for five years holds personal information longer than needed, which is the opposite of a careful retention policy.
- B. Usernames can change and are themselves personal information, so they are poor identifiers across systems.
Question 12
A food distribution company has a website for new clients to sign up and begin using their food services. Nearly 20 clients sign up per day.
The website allows for data download, but lacks the business process ability to automate an onboarding process using this data. The company is considering Appian as a workflow tool to handle the onboarding process for their new clients.
Appian needs to receive the data in real-time from the website containing client information once it is submitted and kick off a new onboarding process for each client.
Which must you clarify before you can recommend an Appian integration approach?
- A. VPN configuration of the website.
- B. Average latency of the website server
- C. REST capabilities of the website.
Show answer and explanation ▾
Correct answer: C
Real-time delivery means the website must send each submission to Appian, or Appian must be able to call the website. Both depend on whether the website can make or expose REST calls, so that must be clarified before choosing between a web API and an outbound integration.
Why the other options are wrong:
- A. VPN configuration is a network detail and does not decide the integration approach.
- B. Server latency does not decide the integration approach, especially at about 20 sign-ups a day.
Question 13
You are designing an application to be used as a point-of-sale system.
You are designing the Enter Transaction process model where a user will enter in the transaction details from a front-end action.
Each day there may be thousands of transactions processed, all of which require this process to be executed.
Each transaction requires a receipt document to be generated and emailed to the customer, a process that could take up to 60 seconds.
The receipts do not need to be accessed within the Appian system again.
Which two recommendations should you suggest? (Choose two.)
- A. Use a single process for the Enter Transaction process model where possible, to minimize the impact of that process on engine memory.
- B. Increase the archival settings of the Enter Transaction process model to ensure receipts can be regenerated in the event of an error.
- C. Activity chain the Enter Transaction process up until the receipt generation and send email node, so that the customer experience is not impacted.
- D. Conduct periodic archival of documents into an external document storage system.
Show answer and explanation ▾
Correct answer: C, D
Activity chaining only up to the receipt step lets the user finish the transaction at once, while the 60 second receipt generation and email run afterwards. Because thousands of receipts are created daily and never read again in Appian, they should be archived regularly to external storage.
Why the other options are wrong:
- A. Keeping the slow receipt work in one process makes every transaction instance larger and longer lived in engine memory.
- B. Longer archival settings keep more process data in memory, and the receipts do not need to be accessed again.
Question 14
Your customer's use case relies heavily on plug-ins and automation capabilities, such as document extraction for unstructured data from legal contracts. Due to the difficulties that compliance provides, the customer must be self-managed.
The customer is interested in using Appian's document extraction AI skill as it seems like it could be a good fit. Also, it fits within their use of AI policies.
What should you suggest?
- A. Use the non-generative Al skill for document extraction.
- B. Use the generative AI skill for document extraction.
- C. Recommend an alternate solution, as the AI skill will not work for this use case.
Show answer and explanation ▾
Correct answer: A
The document extraction AI skill pulls fields out of documents without generating new content. It meets the extraction need for the contracts and stays within the customer's policy on the use of AI.
Why the other options are wrong:
- B. A generative skill is not what the customer's stated policy and extraction need call for.
- C. The extraction need can be met with an AI skill, so an alternate solution is not required.
Question 15
On the homepage in a new Appian application, a dynamic Ul component displays a student's weekly calendar.
The component contains a grouped listing of activities broken out by the seven (7) days of the week. The code for the component lives in the homepage interface and directly references record types for Student, Course, and Location.
The product owner loves the component and wants to reuse it as much as they can in any future Appian applications. You decide to pull the code for the component out and put it in a new interface rule.
Which two steps should now be taken to make this code a completely reusable object for other applications within the same Appian environment?
(Choose two.)
- A. Create a generic record type for calendar content, including person, activity, and location.
- B. Replace all record type references with rule inputs representing the calendar fields.
- C. Set the data type to "Any Type" for the rule inputs for the reusable calendar fields.
- D. Create a local variable which maps input record types to record fields to be used in the component.
Show answer and explanation ▾
Correct answer: B, D
Replacing the direct references to Student, Course and Location with rule inputs removes the tie to one application's record types. A local variable then maps whatever record data is passed in to the fields the calendar uses, so any application can feed the component.
Why the other options are wrong:
- A. A generic record type would still bind the component to one specific record type.
- C. Any Type inputs remove type checking and are not needed to make the component reusable.
Question 16
You have just completed your first Appian application for a new customer. It is a simple case management application containing several record types, with the Appian Cloud database as their data source.
The customer now has a business requirement to integrate all data stored in Appian with their existing external data visualization and reporting system, updated nightly. The external system has a requirement where the data it visualizes must be warehoused within the external system.
What is the best technical approach to exposing all Appian data to this external system?
- A. Create a VPN tunnel that allows for direct external connections to the Appian Cloud database, then point the external system to the relevant database tables.
- B. Create a process to dump relevant database tables into CSV files nightly, then move the CSV files over to the external system using SFTP tools.
- C. Create a web API that returns all records by their record ID, their relevant service accounts, and expose the Web API to the external system.
Show answer and explanation ▾
Correct answer: B
The external system must hold its own copy of the data and only needs it nightly. A scheduled process that exports the relevant tables to CSV files and transfers them over SFTP delivers a full batch that the external system can load into its warehouse.
Why the other options are wrong:
- A. Pointing the external system at Appian tables does not warehouse the data inside the external system as required.
- C. A web API that returns records by ID is an inefficient way to move all data every night.
Question 17
You are required to set up an integration to a customer relationship management (CRM) that implements OAuth 2.0 Authorization Code Grant for authentication.
You need to request for a number of parameters to establish connectivity.
Which parameters, in addition to Client ID, Client Secret, and Scope, should you request?
- A. Token Request Endpoint. Token Refresh Endpoint
- B. Authorization Endpoint, Token Request Endpoint
- C. Token Request Endpoint, Additional Parameters
Show answer and explanation ▾
Correct answer: B
The Authorization Code Grant has two steps. The user is sent to the authorization endpoint to sign in and consent, and the returned code is then exchanged at the token endpoint for an access token, so both URLs are required.
Why the other options are wrong:
- A. Without the authorization endpoint the user consent step cannot start.
- C. Additional parameters are optional, and the authorization endpoint is still missing.
Question 18
You are optimizing a database for a loan management system that has a large table named Loan_Handlers.
The table has columns such as Ticket_Id, Employee_Id, Status, and File (stored as BLOB).
The table is queried often on Ticket_Id in various places within the application.
The Ticket_Id is currently stored as text, even though it only contains numeric values.
The Employee_Id and Status columns have inconsistent data types across related tables in the schema.
Users in the system also frequently upload files.
Which two steps will improve the query performance on the Loan Handlers table? (Choose two.)
- A. Ensure that Employee_Id and Status columns have the same data type across all related tables
- B. Implement full-text indexing on the Ticket_Id column to speed up searches.
- C. Convert the Ticket_Id column from text to an integer type
- D. Store the File column in a separate table
Show answer and explanation ▾
Correct answer: C, D
Storing Ticket_Id as an integer makes comparisons and index lookups faster and the index smaller than with text. Moving the BLOB column to its own table keeps the Loan_Handlers rows small, so frequent queries read far less data.
Why the other options are wrong:
- A. Matching data types across related tables improves consistency, but it is not what speeds up queries on this table.
- B. Full-text indexing is for searching words inside text, not for lookups on a numeric identifier.
Question 19
You need to implement data fabric for a client, where the required database contains two million rows beyond the supported limit. Your client also indicates that the amount of data is set and will not increase with time.
What should you recommend to ensure that all key data is constantly available?
- A. Split up data into more than one synced record type, and use sync filters on each so that only a certain data subset is available for each record type.
- B. Use a single synced record type, and configure record-level security so that only a certain data subset is available.
- C. Use one or multiple synced record types, and configure a rate limit so that the batch limit is not reached.
Show answer and explanation ▾
Correct answer: A
The row limit applies to each synced record type. Splitting the data across more than one synced record type, each with a sync filter that selects its own subset, keeps every record type under the limit while all the data stays synced.
Why the other options are wrong:
- B. Record-level security hides rows from users but does not reduce the number of rows that are synced.
- C. A rate limit does not change the row limit of a synced record type.
Question 20
You are building an application for university students to manage their academic registrations and post their achievements on their LinkedIn profiles.
The Appian application must request permission from the students to access their LinkedIn account for posting the achievement.
Which authentication type should be used?
- A. OAuth 2.0: Client Credentials Grant
- B. OpenID Connect
- C. OAuth 2.0: Authorization Code Grant
Show answer and explanation ▾
Correct answer: C
The Authorization Code Grant sends each student to LinkedIn to sign in and approve access. Appian then receives a token that lets it post on behalf of that student, which is the permission step the requirement describes.
Why the other options are wrong:
- A. Client Credentials authenticates the application itself and has no user consent step.
- B. OpenID Connect verifies who the user is; it does not grant permission to post on their behalf.
Question 21
You are the sole system administrator of an Appian Cloud development environment, managing two teams of developers.
In Appian, you've created a new group for each team and added them to the Designers system group. You've created Basic user accounts for each developer and added them to their respective group. You want to ensure that the two developer teams do not have access to each other's data.
You call the createNewSchema() stored procedure twice to create a dedicated schema for each team. Then, you copied down the credentials that the Appian Cloud database returned to you.
What should you do next?
- A. Add the new credentials to the database configuration in the Administration Console. Then, configure the database's security rules to ensure both groups have Editor permissions.
- B. Use the credentials provided to create two connected systems Ensure that both groups have Viewer permissions to their connected system
- C. Create two new user groups for controlling schema access and add the credentials under each group’s properties. Then, add both teams to their respective groups and ensure that both groups have Editor permissions.
Show answer and explanation ▾
Correct answer: B
Each new schema has its own credentials, and a data source connected system is created for each one with those credentials. Giving each team's group access only to its own connected system keeps the two teams out of each other's data.
Why the other options are wrong:
- A. Giving both groups Editor permissions on one shared configuration would not separate the two teams' data.
- C. Group properties do not hold database credentials.
Question 22
You are working with a manufacturing company that is transitioning functionality from an outdated legacy system into new Appian applications. The company has the following timeline: Phase 1: Add the sales team as Appian users and migrate their functionality to Appian. The go-live date is six (6) weeks from today. Phase 2: Add the customer service team as Appian users and migrate their functionality to Appian. The go-live date is to be determined. When phase 1 goes live, the sales team will be creating hundreds of new records each day in Appian. The company has requested these new records also be created in the legacy system so that the customer service team can continue assisting their customers with correct and up-todate data until phase 2 is completed. The records are relatively simple, but currently the only integration option available within the legacy system is a nightly CSV file ingestion. What should you recommend to minimize the risk of delaying the phase 1 go-live date and best meet this company's needs?
- A. Use RPA to create the records in the legacy system once they are created in Appian.
- B. Implement nightly CSV file ingestion and email the customer service team to notify them of new records.
- C. Work with the company’s IT team to create API endpoints in the legacy system.
Show answer and explanation ▾
Correct answer: B
Nightly CSV ingestion is the one integration the legacy system already supports, so it adds the least delivery risk in a six week window. The records are simple, and an email tells the customer service team when new records have been loaded.
Why the other options are wrong:
- A. Building and testing automation against the legacy screens adds delivery risk within six weeks.
- C. New API endpoints in the legacy system are extra development that could delay go-live.
That was 22 of 71.
The full Appian ACD301 pack has all 71 questions, each with the answer, the explanation and why the other options are wrong, plus a questions-only copy for timed runs. $39.00 USD, paid once, with free monthly updates and a pass-or-your-money-back guarantee.
Get the full packNot ready yet? Get the first 24 free as one PDF by email. Get the Half Pack →
About the ACD301 exam
These 22 Appian ACD301 questions are free here. The free PDF has 24, free by email.
They are drawn from the same bank as the full ACD301 pack, which has 71 questions in total.