20 free Salesforce Platform Integration Architect practice questions from the CertStash pack of 175, each with the correct answer and a full explanation. The rule is the same for every exam: ten free on the product page, 20 free here, all 175 in the pack, and each step includes the one before it.
Get all 175 questions (US$39) · Download the first 10 as a PDF
Question 1
A company is designing a Lightning Web Component (LWC) to display transactions aggregated from different sources. Their current system landscape is as follows:
1. Transactions are created at any time through their various on-premise and cloud-based systems.
2. All necessary transactions are replicated to a custom Transaction object in Salesforce. It is updated periodically so it only has a subset of the necessary transactions between updates.
3. Middleware supports publish-subscribe interactions and provides RESTful Enterprise APIs that can retrieve transactions from on-premise and cloud-based systems.
The company wants to address a usability concern regarding incomplete data displayed on the LWC component.
What should the integration architect specify so the LWC will be able to display all the required transactions?
Show answer and explanation
Correct answer: C. Use the Continuation class to call the Enterprise APIs and then process the response in a callback method.
The custom Transaction object only ever holds a subset of the data between replication runs, so the component must fetch the complete set from the systems of record at view time. The middleware already exposes RESTful Enterprise APIs for this. An Apex Continuation makes that long-running callout asynchronously from the Lightning component's controller and processes the response in a callback method, so the user sees every required transaction without the request tying up a synchronous Apex thread while the middleware gathers data from on-premise and cloud systems.
Why the other options are wrong
- A. Asking middleware to refresh the custom object is still an asynchronous replication, so the component can display data before the update arrives.
- B. Calling the APIs straight from client-side JavaScript runs into browser security restrictions and bypasses server-side credential management.
- D. Lightning Data Service only reflects the incomplete custom object, so it cannot show transactions that have not been replicated yet.
Question 2
A global financial company sells financial products and services that include the following:
1. Bank accounts
2. Loans
3. Insurance
The company has a state-of-the-art core banking system that is the master system to store financial transactions, financial products and customer information. The core banking system currently processes 10 million financial transactions per day. The CTO for the company is considering building a community portal so that customers can review their bank account details, update their Information and review their account financial transactions.
What should an integration architect recommend as a solution to enable customer community users to view their financial transactions?
Show answer and explanation
Correct answer: A. Use Salesforce External Service to display financial transactions in a community Lightning page.
The core banking system stays the system of record and processes 10 million transactions a day, so the transactions should be shown from the source rather than copied into Salesforce. External Services registers the core banking API from its OpenAPI specification and exposes its operations as invocable actions, which a flow on a community Lightning page can call to retrieve a customer's transactions on demand and display them. Customers see current data from the master system without it being stored or synchronized in Salesforce.
Why the other options are wrong
- B. External objects need an OData or custom adapter in front of the core banking system, while External Services calls its existing API directly from the community page.
- C. Replicating 10 million transactions a day into a custom object duplicates the master data and consumes heavy storage.
- D. An iframe embeds another application's UI without integrating the data into the Salesforce security and sharing model.
Question 3
Universal Containers (UC) is a large printing company that sells advertisement banners. The company works with third-party agents on banner initial design concepts. The design files are stored in an on-premise file store that can be accessed by UC internal users and third party agencies.
UC would like to collaborate with the third-party agencies on the design files and allow them to view the design files in the community.
The conceptual design files size is 2.5 GB.
Which solution should an integration architect recommend?
Show answer and explanation
Correct answer: C. Create a custom object to store the file location URL: when a community user clicks on the file URL, redirect the user to the on-premise system file location.
The design files come to 2.5 GB, and the files already live in an on-premise store that both internal users and third-party agencies can reach. Storing the file location URL in a custom object lets community users find the right design in context and then be redirected to the on-premise location to open it. This avoids moving very large files into Salesforce while still giving agencies a single place in the community to collaborate on the designs.
Why the other options are wrong
- A. Uploading 2.5 GB design files to Salesforce Files moves very large files into Salesforce storage instead of linking to where they already live.
- B. Salesforce Connect exposes external records as external objects; it does not upload or store binary files.
- D. Streaming multi-gigabyte files through a Request and Reply callout runs into callout size and timeout limits.
Question 4
A business requires automating the check and updating of the phone number type classification (mobile vs. landline) for all incoming calls delivered to its phone sales agents. The following conditions exist:
1. At peak, the call center can receive up to 100,000 calls per day.
2. The phone number type classification is a service provided by an external service API.
3. Business is flexible with timing and frequency to check and update the records (throughout the night or every 6-12 hours is sufficient).
A Remote-Call-In pattern and/or Batch Synchronization (Replication via ETL:
System -> Salesforce) are determined to work with a middleware hosted on customer premise.
In order to implement these patterns and mechanisms, which component should an integration architect recommend?
Show answer and explanation
Correct answer: C. Connected App configured in Salesforce to authenticate the middleware.
Both chosen patterns, Remote Call-In and Batch Data Synchronization into Salesforce, have the on-premise middleware calling Salesforce APIs to update the phone number classification. Inbound API access must be authenticated, and a connected app is the Salesforce component that lets an external client such as middleware authenticate with OAuth 2.0, with scopes and policies controlling what it can do. Configuring a connected app for the middleware therefore provides secure, governed access for the overnight or periodic updates of up to 100,000 records a day.
Why the other options are wrong
- A. Remote Site Settings authorize outbound callouts from Salesforce and do not authenticate inbound middleware calls.
- B. Here the middleware calls out to Salesforce, so internal resources are not being exposed to inbound traffic.
- D. An API gateway for Salesforce-to-middleware requests addresses the opposite direction from Remote Call-In and ETL into Salesforce.
Question 5
Universal Containers (UC) is currently managing a custom monolithic web service that runs on an on-premise server.
This monolithic web service is responsible for Point-to-Point (P2P) integrations between:
1. Salesforce and a legacy billing application
2. Salesforce and a cloud-based Enterprise Resource Planning application
3. Salesforce and a data lake
UC has found that the tight interdependencies between systems is causing integrations to fail.
What should an architect recommend to decouple the systems and improve performance of the integrations?
Show answer and explanation
Correct answer: B. Leverage modular design by breaking up the web service into smaller pieces for a microservice architecture.
The failures come from tight interdependencies inside one monolithic web service that handles three unrelated point-to-point integrations. Breaking it into smaller, independently deployable services, one concern per integration, decouples the systems so that a problem with the billing application, the ERP or the data lake no longer brings down the others. Each microservice can also be scaled and tuned for its own workload, which improves integration performance as well as resilience.
Why the other options are wrong
- A. Optimizing the same monolith keeps the shared dependencies that cause the failures.
- C. The Bulk API only changes how data is loaded into Salesforce and leaves the coupling between systems untouched.
- D. Moving the monolith to a cloud host changes where it runs, not how tightly the integrations depend on each other.
Question 6
Northern Trail Outfitters needs to present shipping costs and estimated delivery times to its customers. Shipping services used vary by region, and have similar, but distinct service request parameters.
Which integration component capability should be used?
Show answer and explanation
Correct answer: C. Enterprise Service Bus to determine which shipping service to use, and transform requests to the necessary format.
Shipping services vary by region and each expects slightly different request parameters. An Enterprise Service Bus can apply routing logic to choose the right shipping service for the customer's region and transform the single request from Salesforce into the format that service expects, then return the cost and delivery estimate. Keeping routing and transformation in the ESB means adding or changing a shipper does not require changes in Salesforce.
Why the other options are wrong
- A. Outbound Messaging is one-way and cannot return the costs and delivery times customers need to see.
- B. An Apex REST service is an inbound endpoint, and putting routing logic in Salesforce makes each shipper change a code change.
- D. Collecting shipper-specific form data in an ESB user interface is not a function of an ESB and would fragment the customer experience.
Question 7
An integration developer is developing an HR synchronization app for a client. The app synchronizes Salesforce record data changes with an HR system that's external to Salesforce.
What should the integration architect recommend to ensure notifications are stored for up to three (3) days if data replication fails?
Show answer and explanation
Correct answer: D. Change Data Capture
The app synchronizes Salesforce record changes with an external HR system, which is exactly what Change Data Capture is built for. It publishes change events automatically whenever records are created, updated, deleted or undeleted, and the event bus retains those events for three days. If replication fails, the subscriber can use the replay ID of the last event it processed to retrieve the missed changes within that window, so no notifications are lost while the HR system recovers.
Why the other options are wrong
- A. Platform Events would need custom publishing logic for every record change instead of capturing changes automatically.
- B. Callouts are point-in-time requests with no stored notifications to recover after a failure.
- C. Generic events are retained for only 24 hours, not three days.
Question 8
A company needs to be able to send data from Salesforce to a home grown system behind a corporate firewall. The data needs to be pushed only one way and doesn't need to be sent in real time. The average volume is 2 million records per day.
What should an integration architect consider when choosing the right option in building the integration between the external system and Salesforce?
Show answer and explanation
Correct answer: B. Due to high volume of records, the external system will need to use a BULK API Rest endpoint to connect to Salesforce.
The data flows one way, does not need to be real time, and averages 2 million records a day to a system behind a corporate firewall. Rather than opening the firewall to inbound pushes from Salesforce, the external system can pull the data on a schedule. At this volume it should use the Bulk API REST endpoint, which processes large data sets asynchronously in batches and uses far fewer API calls than record-by-record requests.
Why the other options are wrong
- A. Salesforce is not limited by concurrent requests to an external REST API; the concern at this volume is choosing a bulk mechanism.
- C. Pushing 2 million records through REST callouts from Salesforce runs into callout limits and requires opening the firewall.
- D. A third-party staging tool is an option, not a requirement, because the Bulk API handles this volume directly.
Question 9
Northern Trail Outfitters is seeking to improve the performance and security of outbound integrations from Salesforce to on-premise servers.
What should the architect consider before recommending a solution?
Show answer and explanation
Correct answer: B. External gateway products in use
Improving the performance and security of callouts from Salesforce to on-premise servers depends heavily on the network path into the corporate data center. Before recommending a solution, the architect should know which external gateway products, such as API gateways, reverse proxies or secure gateways in the DMZ, are already in place, because they determine how traffic is authenticated, inspected and routed and what performance features such as caching are available. The design can then reuse that infrastructure.
Why the other options are wrong
- A. Shield Platform Encryption protects data at rest in Salesforce, not the path of outbound calls to on-premise servers.
- C. Deterministic encryption concerns filtering on encrypted fields, not callout performance or transport security.
- D. Default gateway restrictions are generic network routing settings rather than an integration design consideration.
Question 10
Northern Trail Outfitters (NTO) is planning to create a native employee-facing mobile app with the look and feel of Salesforce Lightning Experience.
The mobile app needs to integrate with NTO's Salesforce org.
Which Salesforce API should be used to implement this integration?
Show answer and explanation
Correct answer: A. User Interface API
User Interface API is built for native mobile and custom web apps that need to look and behave like Salesforce Lightning Experience. A single response returns record data together with layout information, field metadata, picklist values and theme details, and it respects the user's field-level security and sharing. The app can therefore render Lightning-style record pages with the same layouts administrators configure, without assembling metadata and data from several separate calls.
Why the other options are wrong
- B. Streaming API pushes notifications about changes and provides no layout or metadata for building a UI.
- C. Connect REST API focuses on Chatter, Experience Cloud sites and related resources, not Lightning record layouts.
- D. The REST API returns record data without the layout and display metadata needed to mirror Lightning Experience.
Question 11
Universal Containers (UC) is a leading provider of management training globally. UC requested students' course registration data generated from the Salesforce student community to be synced with the learning management system (LMS). Any update to the course registration data needs to be reflected in the LMS.
Which integration mechanism should be used to meet the requirement?
Show answer and explanation
Correct answer: C. Change Data Capture (CDC)
The LMS must stay in sync with course registration records, reflecting every create and update made in the student community. Change Data Capture publishes change events automatically whenever records of a selected object are created, updated, deleted or undeleted, including the changed field values. The LMS, or middleware in front of it, subscribes to those events and applies the changes, so no custom trigger or publishing logic is needed and every update reaches the LMS.
Why the other options are wrong
- A. Platform Events would require custom logic to publish an event on every registration change.
- B. Streaming API is the general family of push channels; Change Data Capture is the specific mechanism designed for record synchronization.
- D. Outbound Messages are sent from a workflow rule or flow and send selected fields rather than a complete record-change feed.
Question 12
A business-to-consumer (B2C) enterprise customer has the following use case that involves processing payment from an external payment gateway service in Salesforce.
1. Customer requests Customer Service Representative (CSR) for a Service upgrade.
2. Customer provides credit card details to CSR for payment.
3. CSR submits payment information in Salesforce, and it is processed in an external payment gateway.
4. CSR receives confirmation of payment.
5. CSR upgrades service for customer and confirms customer.
This use case requires the CSR to obtain confirmation of payment before upgrading the service. The integration with the payment gateway needs to be reliable and monitored for audit purposes. The payment gateway service is an external RESTful service that the B2C enterprise customer has subscribed for.
What should an integration architect recommend for this integration?
Show answer and explanation
Correct answer: B. Make a callout to the payment gateway through ESB supporting error handling and logging for audit purposes.
The CSR must wait for payment confirmation before upgrading the service, so this is a synchronous Request and Reply interaction, and the integration must be reliable and monitored for audit. Routing the callout through the ESB gives a central point for error handling, retries, transformation and logging of every request and response, which satisfies the audit requirement without building that infrastructure in Salesforce. The ESB returns the gateway's confirmation to Salesforce so the CSR can complete the upgrade.
Why the other options are wrong
- A. Platform events are fire-and-forget, so the CSR would not receive a synchronous payment confirmation.
- C. External Services simplifies invoking the API but provides no central error handling or audit monitoring.
- D. A custom Apex callout puts logging, retries and audit tracking into custom code that must be built and maintained.
Question 13
Which Web Services Description Language (WSDL) should an architect consider when creating an integration that might be used for more than one Salesforce org and different metadata?
Show answer and explanation
Correct answer: C. Partner WSDL
The Partner WSDL is loosely typed and not tied to one org's data model, so a single client application can work with many Salesforce orgs that have different metadata. It represents records generically as sObjects with fields, which is why partners and ISVs use it for integrations deployed across multiple customers. The Enterprise WSDL, by contrast, is strongly typed to the objects and fields of one specific org and must be regenerated when that org's metadata changes.
Why the other options are wrong
- A. There is no Corporate WSDL among the Salesforce API WSDLs.
- B. The Enterprise WSDL is strongly typed to a single org's metadata, so it does not suit multiple orgs.
- D. SOAP API WSDL is a general description, while the multi-org option is specifically the Partner WSDL.
Question 14
An organization needs to integrate Salesforce with an external system and is considering authentication options. The organization has already implemented SAML using a third-party identity provider (IdP) for integrations between other systems.
Which use case can leverage the existing SAML integration to connect Salesforce with other internal systems?
Show answer and explanation
Correct answer: C. Make formula fields with HYPERLINK() to external web servers more secure.
SAML is a browser-based single sign-on protocol: a user who has authenticated with the identity provider can open other applications that trust the same IdP without logging in again. When a formula field uses HYPERLINK() to send a Salesforce user to an internal web server, that server, as a SAML service provider, can authenticate the user through the existing third-party IdP. This lets Salesforce users move securely into other internal systems while reusing the SAML integration already in place.
Why the other options are wrong
- A. Apex REST callouts are system-to-system calls secured with named credentials, OAuth or certificates, not interactive SAML logins.
- B. An external Java client calling the Salesforce API authenticates with an OAuth flow through a connected app rather than the existing SAML integration.
- D. Apex SOAP callouts are also server-to-server and do not use browser-based SAML single sign-on.
Question 15
An architect recommended using Apex code to make callouts to an external system to process insurance quotes.
What should the integration architect consider to make sure this is the right option for the integration?
Show answer and explanation
Correct answer: D. The limit on long-running requests (total execution time).
Processing insurance quotes in an external system can take time, and synchronous Apex callouts that wait for a slow response hold a request open for their full duration. Salesforce limits the number of long-running synchronous requests that can run concurrently in an org, and requests that exceed that threshold are refused. The architect must therefore assess the expected response time and total execution time of each quote request to confirm that synchronous Apex callouts will not exhaust this limit under load.
Why the other options are wrong
- A. A quote request makes few callouts, so the per-transaction callout count is unlikely to be the constraint.
- B. The number of parallel callouts in a continuation applies only when a Continuation is used.
- C. The pending operations limit concerns uncommitted DML before a callout, not whether callouts are the right option.
Question 16
Universal Containers (UC) currently owns a middleware tool and has developed an API-led integration architecture with three API tiers. The first tier interfaces directly with the systems of engagement, the second tier implements business logic and aggregates data, and the third tier interfaces directly with the systems of record. Some of the systems of engagement will be a mobile application, a web application, and Salesforce.
UC has a business requirement to return data to the systems of engagement in different formats while also enforcing different security protocols.
What should an integration architect recommend to meet these requirements?
Show answer and explanation
Correct answer: C. Implement an API Gateway that all systems of engagement must interface with first.
An API gateway placed in front of the API tiers gives every system of engagement a single, managed entry point. Gateway policies can be defined per consumer, so the mobile app, the web app and Salesforce can each be held to a different security protocol, such as OAuth, client credentials or mutual TLS, and the gateway can transform responses into the format each consumer expects. Handling these cross-cutting concerns at the gateway leaves the API tiers free to focus on business logic and system access.
Why the other options are wrong
- A. The second tier holds reusable business logic and orchestration, so mixing in channel-specific formats and security would tie it to individual consumers.
- B. SAML provides browser single sign-on and does not shape the data formats returned to each system of engagement.
- D. Building separate security and formatting into each first-tier API spreads the same policies across many services instead of enforcing them centrally at a gateway.
Question 17
Northern Trail Outfitters submits orders to the manufacturing system web service. Recently, the system has experienced outages that keep service unavailable for several days.
Which solution should an architect recommend to handle errors during these types of service outages?
Show answer and explanation
Correct answer: B. Use middleware queuing and buffering to insulate Salesforce from system outages.
Outages lasting several days are longer than any retry window Salesforce features provide. Middleware with queuing and buffering accepts orders from Salesforce immediately and holds them durably until the manufacturing system comes back online, then delivers them in order and handles retries and errors centrally. Salesforce is insulated from the outage, users can keep submitting orders, and no custom retry logic needs to be built and maintained on the platform.
Why the other options are wrong
- A. Outbound Messaging retries for only about 24 hours, which is shorter than a multi-day outage.
- C. Platform events are retained for a limited window, and a custom scheduled Apex replay process is fragile compared with middleware queuing.
- D. A @future method does not return a job ID, and custom scheduled retry logic in Apex is complex to build and maintain.
Question 18
A customer of Salesforce has used Platform Events to integrate their Salesforce instance with an external third-party artificial intelligence (AI) system. The AI system provides a prediction score for each lead that is received by Salesforce. Once the prediction score is received, the lead information is saved to Platform Events for other processes. The trigger on the Platform Events has failed ever since it was rolled out to production.
What type of monitoring should the integration consultant have considered to monitor this integration?
Show answer and explanation
Correct answer: A. Set up debug logs for Platform Event triggers to monitor performance.
The integration publishes events successfully, but the trigger that processes them has failed since go-live. Platform event triggers run asynchronously as the Automated Process user, so their errors do not surface to any user session. Setting up debug logs for the platform event trigger, with a trace flag on the Automated Process entity, captures each execution, its errors and its performance, which is the monitoring needed to detect and diagnose a failing subscriber before it disrupts the downstream processes that rely on the lead data.
Why the other options are wrong
- B. Hourly publishing limits govern how many events are published, not why the subscribing trigger fails.
- C. Comparing definitions is a one-time design check, not ongoing monitoring of the integration.
- D. Lead volume says nothing about whether the platform event trigger is executing successfully.
Question 19
Service agents at Northern Trail Outfitters use Salesforce to manage cases and B2C Commerce for ordering.
Which integration solution should an architect recommend in order for the service agents to see order history from a business-to-consumer (B2C) Commerce system?
Show answer and explanation
Correct answer: C. Salesforce B2C Commerce to Service Cloud Connector
Salesforce provides the B2C Commerce to Service Cloud Connector as a prebuilt integration between the two platforms. It synchronizes customer profiles and lets service agents see order history and order details from B2C Commerce inside Service Cloud, so agents can resolve cases without switching systems. Using the supplied connector avoids building and maintaining a custom integration for a scenario Salesforce already supports out of the box.
Why the other options are wrong
- A. Calling the Commerce REST API directly means building a custom integration for something the connector already provides.
- B. The Salesforce Platform REST API exposes Salesforce data and does not retrieve order history from B2C Commerce.
- D. MuleSoft could build this integration, but it adds cost and development when a purpose-built connector exists.
Question 20
Northern Trail Outfitters (NTO) uses a custom mobile app to interact with its customers. One of the features of the app is Salesforce Chatter Feeds. NTO wants to automatically post a Chatter item to Twitter whenever the post includes the #thanksNTO hashtag.
Which API should an integration architect use to meet this requirement?
Show answer and explanation
Correct answer: A. Connect REST API
Chatter feeds, feed items and their content are exposed through the Connect REST API, which is designed for mobile apps and other clients that work with Chatter. The integration can read feed items, detect posts that include the #thanksNTO hashtag and pass their content to Twitter. Because the API returns feed data structured for display, including hashtags and mentions, it is the natural choice for a custom mobile app that already works with Chatter.
Why the other options are wrong
- B. The REST API works with sObject records and does not provide structured Chatter feed resources.
- C. Apex REST means writing custom endpoints for something the Connect REST API already provides.
- D. Streaming API pushes notifications about record or event changes and is not the API for Chatter feed content.
That was 20 of 175.
The full Salesforce Platform Integration Architect pack has all 175 questions, each with the answer, the explanation and why the other options are wrong, plus a questions-only copy for timed runs. US$39, paid once, with free monthly updates and a pass-or-your-money-back guarantee.
Not ready yet? Get the first 80 free as one PDF by email. Get the Half Pack →