Skip to main content

New to this track? Our Salesforce Certified Platform Developer exam prep is the recommended first step. See our certification path to understand where this certification fits. The Salesforce Certified Heroku Architect exam weights Heroku Enterprise most heavily (28%), see the full breakdown, study tips, and practice questions below.

Free sample questionsIndependent exam-prep resourceEditorial policyConfirm current details in the official guide

Free practice • No sign-up required

Independent study resource

Exam objectives and policies can change. Confirm current details with the certification issuer.

Check current objectives

Use the official exam guide as the source for current objectives and section weights.

Practice for learning

Use sample questions to review concepts and identify topics for further study.

Works on Any Device

Full practice experience on phone, tablet, or desktop

Free, No Sign-Up

Start practicing immediately, no account or email required

Heroku ArchitectAdvanced

Salesforce Certified Heroku Architect – Complete 2026 Guide

This exam tests how you design apps on Heroku that scale, stay secure and work with Salesforce. You choose runtimes, data services, deployment flows and integration options. The exam is 60 scored questions in 105 minutes, with a 72% pass mark, weighted towards Heroku Enterprise (28%) and Data (17%).

Salesforce Certified Heroku Architect exam quick facts
Official exam name and codeSalesforce Certified Heroku Architect (Heroku Architect)
Questions60
Time allotted105 min
Passing score72%
Exam fee$400
Retake fee$200
Prerequisite certificationsRecommended preparation: Salesforce Certified Platform Developer
Study guide releaseSummer '26

Salesforce Certified Heroku Architect Exam Weightage by Section

1. Heroku Platform (10%)

10%

2. Data (17%)

17%

3. Security (15%)

15%

4. Heroku Enterprise (28%)

28%

5. Architect Applications (15%)

15%

6. Integrations (15%)

15%

Exam Topics

HerokuScalabilityDeploymentBuild WorkflowsTeamsBest Practices

Last updated:

Who is the Heroku Architect exam for?

  • Hands-on practitioners: You build and run apps on Heroku and want to show you can design them for scale, security and uptime.
  • Solution designers: You plan solutions where Heroku apps work alongside Salesforce, using Heroku Connect, events or APIs.
  • Career builders: You are a developer moving into an architect role on projects that mix Salesforce and custom apps.

Exam Fees & Registration

Exam Fee

$400

One-time registration fee

Retake Fee

$200

If you need to retake the exam

Comparing certs? View all Salesforce exam fees in one place →

Certification Validity

Maintenance and renewal requirements vary. Check the issuer’s current credential page for requirements that apply to you.

How to Register

Check current Salesforce Certified Heroku Architect credential and registration details with the issuer.

View official credential details

Official references: Use the issuer’s current materials to confirm exam objectives, format, and fees before booking.

Official sources: Salesforce exam pricing·Independent exam-prep resource; details may change.

Exam Tips

  • 1Heroku Architecture is 30%. Know dynos, add-ons, buildpacks, and scaling strategies.
  • 2Understand scaling and performance: horizontal vs vertical scaling, caching, database optimization.
  • 3Know security and data: encryption, compliance, data residency.
  • 4Be ready for "how would you architect X on Heroku?" questions.

Prerequisites

  • •Heroku development experience
  • •Architect-level knowledge

Focus Areas

  • •Heroku Architecture
  • •Scaling and Performance
  • •Security and Data
  • •Integration

Study Strategy

Design Heroku-based solutions.

Understand dyno types, scaling strategies, and add-ons.

Practice explaining architectural decisions for cloud-native apps.

Exam Format and First-Attempt Readiness

The Salesforce Certified Heroku Architect exam has 60 questions in 105 min, passing score 72%. Most Salesforce exams test scenario-based decisions. Focus on when to use each feature, not just terms.

  • Study Heroku Enterprise and Data first, together they carry 45% of the exam.
  • Do timed question sets. Build pacing and confidence.
  • Review why wrong answers are wrong. It improves scenario reasoning.
  • Book the exam once your mock scores are steady at 85%+.
Is the Heroku Architect Exam Hard?
Heroku Architect Exam Format Explained

Heroku Architect: Key Concepts

Heroku Platform Architecture & Dynos

Heroku apps run in dynos, which are lightweight containers. Web dynos take HTTP traffic, worker dynos run background jobs, and one-off dynos run admin tasks. Dynos are stateless: the file system is wiped on restart, and dynos are restarted at least once a day. So any data you need to keep must go to a database, cache or file store such as S3. The router drops requests that take more than 30 seconds, so long work belongs on a worker with a queue. Scale horizontally by adding dynos, and vertically by choosing larger dyno types. Performance dynos support autoscaling based on response time.

Data Services: Heroku Postgres & Redis

Heroku Postgres is the main managed database. Followers are read-only copies that take reporting load off the main database. Forks are full copies used for testing or migrations. Higher plans offer high availability, with a standby that takes over if the primary fails. Heroku Key-Value Store (Redis-compatible) is used for caching, sessions and fast queues. Apache Kafka on Heroku handles high-volume event streams between services. Use connection pooling when many dynos connect to one database, so you don't run out of connections.

CI/CD, Review Apps & Release Management

Pipelines move the same build through stages, such as development, staging and production, by promoting it instead of rebuilding. Review apps create a temporary app for each pull request, so changes can be tested before merging. Heroku CI runs tests on every push. Builds use buildpacks or Docker images to turn code into a slug that dynos run. Settings live in config vars, not code, so the same build can run in every stage. Release phase commands, like database migrations, run before new dynos take traffic. Rollbacks are one command to an earlier release.

Security, Compliance & Private Spaces

The Common Runtime is shared and fine for many apps. Private Spaces give you an isolated network in a chosen region, with fixed outbound IPs, internal routing and options to connect to your own network by VPN or peering. Shield Private Spaces add stricter controls, such as extra logging and encryption, for regulated data like health records or payment data. Use Heroku Enterprise teams with SSO and fine-grained permissions to control who can deploy and change settings. Keep secrets in config vars and limit who can read them.

Architecture Patterns for Salesforce + Heroku

Heroku Connect syncs Salesforce objects with tables in Heroku Postgres, one way or both ways. It suits apps that need to read lots of Salesforce data quickly or write it back without using up API calls in the app. For events, Heroku apps can subscribe to Platform Events or Change Data Capture, or publish back to Salesforce. For direct calls, apps can use the Salesforce REST API, and Salesforce can call Heroku with Heroku AppLink or External Services. A common pattern: customer-facing apps on Heroku handle heavy public traffic, while Salesforce stays the system of record for staff.

Heroku Connect bidirectional sync (exam mental model)

Heroku Connect maps Salesforce objects to Postgres tables and polls both sides on a schedule (or via streaming), writing changes in whichever direction is configured; this lets a Heroku app read/write Salesforce data without direct API calls.

Salesforce objects sync bidirectionally through Heroku Connect to Postgres tables, which the Heroku app reads and writes directlySalesforceObjects (Account, custom)Heroku ConnectMapping + polling/streamingPostgresMapped tablesHerokuApp

How to Pass the Heroku Architect Exam

Most questions give you an app with one or two firm needs, like compliance, scale or a link to Salesforce. Use those to rule out the designs that almost work.

Heroku Infrastructure Components

Web dynos for requests, worker dynos for background jobs, one-off dynos for admin tasks. Long work goes to workers with a queue.

Heroku Connect Architecture

Use Heroku Connect when an app needs fast access to lots of Salesforce data in Postgres. Know that sync isn't instant and that two-way sync needs care with conflicts.

High Availability & Scaling

Several dynos for the app, a high-availability Postgres plan for the database, and followers for read-heavy reporting.

Security Architecture

Common Runtime for most apps, Private Spaces for network isolation and fixed IPs, Shield Private Spaces for regulated data.

Performance Optimization

Cache with Key-Value Store, pool database connections, move slow work off web dynos, and use autoscaling on performance dynos.

Is the Salesforce Certified Heroku Architect Exam Hard?

DifficultyChallenging; plan about 8-16 weeks of focused prep.

Book When ReadyScore 85%+ on three timed mocks before scheduling.

Use Practice, Not DumpsOriginal practice questions build skill without risking your credential.

Exam format: 60 questions, 105 min, 72% passing score.

Salesforce Certified Heroku Architect Practice Questions (With Explanations)

Test your knowledge with these 15 sample questions.

Question 1: A web request sometimes takes 45 seconds because it generates a large PDF. Users see timeout errors. What is the best fix?

Key takeaway: The router drops requests over 30 seconds.

Show correct answer and explanation

Correct answer: A. Queue PDF creation on a worker dyno

The router drops requests over 30 seconds. Background work on a worker dyno avoids the limit and keeps web dynos free.

  • Refreshing just starts the same slow request again.
  • A larger dyno may be faster but can still pass 30 seconds.
  • You can't turn off the router timeout.

Question 2: An app saves uploaded images to the dyno's local disk. Images vanish after a while. Why?

Key takeaway: Dynos are restarted at least daily and their disk is wiped.

Show correct answer and explanation

Correct answer: C. Dyno file systems are wiped on restart

Dynos are restarted at least daily and their disk is wiped. Files must go to a store like S3.

  • Size isn't the cause. Any file on local disk is lost.
  • The images weren't in the database.
  • Heroku doesn't target images. The whole local disk is temporary.

Question 3: A healthcare app on Heroku will store patient data and must meet strict compliance rules. Which runtime fits?

Key takeaway: Shield Private Spaces add the extra isolation, logging and encryption controls needed for regulated data such as health records.

Show correct answer and explanation

Correct answer: A. A Shield Private Space

Shield Private Spaces add the extra isolation, logging and encryption controls needed for regulated data such as health records.

  • Free or hobby dynos are not suitable for production or compliance.
  • One-off dynos are for admin tasks, not a runtime choice.
  • The Common Runtime is shared and not built for regulated data.

Question 4: Reporting queries are slowing down the main Heroku Postgres database used by the live app. What should be added?

Key takeaway: A follower is a read-only copy that stays in sync.

Show correct answer and explanation

Correct answer: C. A follower database for the reports

A follower is a read-only copy that stays in sync. Reports can run on it without slowing the main database.

  • An empty database has none of the data.
  • A fork is a one-time copy that quickly goes out of date.
  • More web dynos don't reduce database load.

Question 5: A partner's firewall only allows traffic from fixed IP addresses. Your Heroku app must call the partner's API. What fits?

Key takeaway: Private Spaces give the app fixed outbound IPs that the partner can allow.

Show correct answer and explanation

Correct answer: B. Private Spaces with stable outbound IPs

Private Spaces give the app fixed outbound IPs that the partner can allow.

  • Common Runtime IPs change, so the partner can't allow them.
  • Browser calls come from users' IPs, which are unknown.
  • Dyno size doesn't change the IP address.

Question 6: A customer portal on Heroku needs to read millions of Salesforce Account and Order records quickly, without using up Salesforce API calls on every page view. What should be used?

Key takeaway: Heroku Connect keeps a copy of Salesforce data in Postgres, so the app reads it locally with fast SQL queries.

Show correct answer and explanation

Correct answer: A. Heroku Connect syncing the records into Heroku Postgres

Heroku Connect keeps a copy of Salesforce data in Postgres, so the app reads it locally with fast SQL queries.

  • An API call per page view uses up limits and is slower.
  • Config vars are for settings, not data.
  • A weekly CSV is out of date.

Question 7: The team wants a temporary test app created for every pull request. Which feature fits?

Key takeaway: Review apps spin up a fresh app for each pull request and remove it when the pull request is closed.

Show correct answer and explanation

Correct answer: D. Review apps

Review apps spin up a fresh app for each pull request and remove it when the pull request is closed.

  • Release phase runs tasks before a release, like migrations.
  • Promotion moves an existing build between stages. It doesn't create per-PR apps.
  • One-off dynos run single commands.

Question 8: A tested build in staging should go to production without being rebuilt. What should the team use?

Key takeaway: Promotion copies the exact slug from staging to production, so what you tested is what goes live.

Show correct answer and explanation

Correct answer: D. Promote the build through a Pipeline

Promotion copies the exact slug from staging to production, so what you tested is what goes live.

  • Pushing again rebuilds the code, which might not match what was tested.
  • Copying by hand is error-prone.
  • A different buildpack produces a different build.

Question 9: Database migrations must run before new code starts taking traffic. Where should they run?

Key takeaway: Release phase commands run after the build and before new dynos start, so the database is ready first.

Show correct answer and explanation

Correct answer: C. In the release phase

Release phase commands run after the build and before new dynos start, so the database is ready first.

  • Manual steps get missed.
  • Running on the first request can cause errors for users.
  • Browsers can't run database migrations.

Question 10: A Heroku app should react within seconds when an Opportunity is closed in Salesforce. What fits?

Key takeaway: Events push the change to the app as it happens, without polling.

Show correct answer and explanation

Correct answer: B. Subscribing to Change Data Capture

Events push the change to the app as it happens, without polling.

  • Weekly exports are far too slow.
  • Hourly polling is slow and wastes API calls.
  • Email isn't an integration.

Question 11: The app stores its database password in the source code. What should be done?

Key takeaway: Config vars keep settings and secrets out of code, and each environment can have its own values.

Show correct answer and explanation

Correct answer: C. Move the password to a config var

Config vars keep settings and secrets out of code, and each environment can have its own values.

  • A comment is still in the code.
  • The secret is still in the repository, just harder to read.
  • A README is even more visible.

Question 12: Many dynos connect to one Heroku Postgres database, and the database runs out of connections at peak. What helps most?

Key takeaway: Connection pooling shares a smaller number of database connections between many app processes.

Show correct answer and explanation

Correct answer: A. Use connection pooling

Connection pooling shares a smaller number of database connections between many app processes.

  • More dynos would open even more connections.
  • Removing indexes slows queries and doesn't reduce connections.
  • Restarts cause downtime and don't fix the cause.

Question 13: The live app's database must keep running if the primary database server fails. What should be chosen?

Key takeaway: High-availability plans keep a standby that takes over automatically if the primary fails.

Show correct answer and explanation

Correct answer: B. A Postgres plan with high availability

High-availability plans keep a standby that takes over automatically if the primary fails.

  • Backups let you restore, but with downtime and some data loss.
  • A follower can be promoted by hand, but it isn't an automatic failover plan.
  • Dyno size doesn't protect the database.

Question 14: A Heroku app needs a fast cache for user sessions and frequently read data. What should be used?

Key takeaway: Key-Value Store (Redis-compatible) is built for fast caching and session storage.

Show correct answer and explanation

Correct answer: D. Heroku Key-Value Store

Key-Value Store (Redis-compatible) is built for fast caching and session storage.

  • A follower is a read copy of the database, not a cache.
  • Config vars are for settings.
  • Local disk is wiped on restart and isn't shared between dynos.

Question 15: Many services must exchange a very high volume of events reliably, and new consumers will be added over time. What fits?

Key takeaway: Kafka is built for high-volume event streams, and new consumers can read the same stream without changing producers.

Show correct answer and explanation

Correct answer: B. Apache Kafka on Heroku

Kafka is built for high-volume event streams, and new consumers can read the same stream without changing producers.

  • Logs aren't a reliable way to deliver events to other services.
  • Polling a table doesn't scale to high volume.
  • Point-to-point calls tie services together and grow messy.

Get the Full Question Bank

Most candidates book the exam after scoring 85%+ on full mocks.

If you’re planning to test this quarter, aim to complete full mocks at least 10–14 days before your exam date.

Contact us for more information on our full 60-question mock exams and a personalized study plan. Use the contact form.

Request Mock Exams & Study Plan

Salesforce Certified Heroku Architect Exam FAQs

What is covered on the Salesforce Certified Heroku Architect exam?
This page shows the section-wise exam weightage so you know exactly which topics carry the most weight. Use the exam topics and practice questions above to align your study with the official outline.
What is Heroku Architect?
Heroku Architect designs cloud-native application architectures on Heroku, focusing on scaling, performance, security, and integration with Salesforce and other systems.
What experience do I need for Heroku Architect?
You need Heroku development experience and architect-level knowledge. Understanding cloud architecture, scaling strategies, and Heroku platform capabilities is essential.
Are there free practice questions for the Salesforce Certified Heroku Architect exam?
Yes. This page includes free sample practice questions with explanations. Use them alongside the current official documentation.
How do I prepare for the Salesforce Certified Heroku Architect certification?
Use the exam tips, prerequisites, and study strategy on this page. Focus first on the highest-weighted sections, then take the sample practice questions. Schedule the exam when you consistently score well on practice tests.

Next Architect Certifications

After Heroku Architect, many people move toward the wider Salesforce architect credentials:

TL;DR: What Heroku Architect Tests

  • Heroku platform architecture: Dyno types (web, worker, one-off), process model, add-ons ecosystem (Heroku Postgres, Redis, Kafka), buildpacks, Pipelines for CI/CD, review apps, and Private Spaces for network-isolated deployments.
  • Salesforce integration patterns: Heroku Connect for bidirectional data sync with Salesforce, direct Salesforce REST/SOAP API calls from Heroku apps, Platform Events for event-driven integration, and External Objects via Salesforce Connect pointing to Heroku Postgres.
  • Security and scalability: Heroku Private Spaces for compliance (PCI, HIPAA), Shield Postgres for encryption at rest, configuring SSL/TLS for Heroku apps, horizontal scaling (dyno formation), and database connection pooling for high-throughput applications.

Which Sections Carry the Most Weight

Heroku Architecture and Design30%
Salesforce-Heroku Integration25%
Security and Compliance22%
Performance and Scalability15%

Architecture + Integration + Security = 77%. The Heroku Connect data sync pattern and Private Spaces security model are the most exam-critical topics.

Scenario Strategy: How to Approach Heroku Architect Questions

Questions describe an application or integration requirement and ask which Heroku architecture decision, integration pattern, or platform feature is appropriate. Always consider the requirement dimensions: real-time vs. batch, compliance requirements, and data volume.

  • For integration pattern questions: Heroku Connect = near-real-time bidirectional sync of Salesforce data to Postgres (best for data that Heroku apps read and write back to Salesforce). Direct API = synchronous Salesforce API calls for real-time reads/writes. Platform Events = event-driven async integration. External Objects = read Heroku Postgres data from Salesforce without syncing it.
  • For compliance questions: Heroku Private Spaces provide network isolation (no public internet exposure), a dedicated runtime environment, and compliance features for PCI DSS and HIPAA. Heroku Shield adds additional encryption and logging. When a scenario mentions regulated data (payment cards, health data), Private Spaces + Shield is the answer.
  • For scaling questions: horizontal scaling = add more dynos (handles more concurrent requests). Vertical scaling = increase dyno size (handles larger memory workloads). Worker dynos process background jobs, scale them separately from web dynos. Connection pooling (PgBouncer) is required when many dynos need database connections simultaneously.

Booking Readiness Benchmark

85%+ on 3 timed full mocks before booking

Heroku Architect is a specialist certification primarily taken by Salesforce technical architects who need to design Heroku-based solutions. Real Heroku deployment experience, building and deploying an application, configuring Heroku Connect, and managing a Private Spaces environment, is strongly recommended before booking.

3 Concepts That Fail Most Heroku Architect Candidates

These are not the hardest topics; they are the ones where candidates are most confidently wrong. Learn the distinction early.

1. Heroku Connect: Bidirectional Sync Has Write Conflict Rules

Heroku Connect synchronises data between a Heroku Postgres database and Salesforce objects. By default it is unidirectional (Salesforce to Postgres). Enabling bidirectional sync introduces write conflicts when both sides update the same record. Candidates assume bidirectional sync just works, the exam expects knowledge of the "Salesforce wins" vs "database wins" conflict resolution setting and when each applies.

2. Dyno Types: Web vs Worker vs One-Off

Web dynos handle HTTP requests and must respond within 30 seconds (otherwise Heroku returns a timeout). Worker dynos run background jobs with no time limit (used for long-running processes, queue processing). One-off dynos are for ad-hoc tasks (database migrations, scripts). Candidates run long-running jobs on Web dynos, the exam expects Worker dynos for async processing.

3. Add-ons vs Buildpacks: Services vs Build Configuration

Add-ons are third-party services attached to a Heroku app (Postgres database, Redis cache, monitoring tools); they are provisioned and billed separately. Buildpacks are scripts that transform your code into a runnable slug at build time (Node.js buildpack installs npm packages, Python buildpack installs pip packages). Candidates confuse add-ons (runtime services) with buildpacks (build-time configuration).

Questions Candidates Ask

What is the Salesforce Heroku Architect exam format?
The Salesforce Heroku Architect exam has 60 multiple-choice questions, a 105-minute time limit, a 72% passing score, and a $200 fee ($100 retake). It tests the ability to architect Heroku applications integrated with Salesforce: platform services, security, performance, scalability, and the Salesforce-Heroku integration patterns.
What are the highest-weight Heroku Architect exam sections?
Heroku Architecture and Design (30%) and Salesforce-Heroku Integration (25%) together account for 55% of the exam. Understanding Heroku dynos, add-ons, Heroku Postgres, Heroku Connect for data synchronisation, and when to use Heroku vs. Salesforce Platform for different application requirements are the most heavily tested areas.
What is Heroku Connect and why is it important for this exam?
Heroku Connect is a managed data synchronisation service that syncs Salesforce data to a Heroku Postgres database in near real-time. It is a key integration pattern for Heroku-Salesforce architectures. The exam tests when to use Heroku Connect vs. direct API integration, how to configure sync mappings, and the limitations of Heroku Connect (API limits, sync latency, supported object types).
When should an architect choose Heroku over Salesforce Platform?
Heroku is preferred when you need: open-source language flexibility (Node.js, Python, Ruby, Java), high-volume public-facing web applications, complex compute workloads that exceed Salesforce governor limits, or when mixing Salesforce and non-Salesforce data sources. Salesforce Platform is preferred for CRM-centric applications, declarative automation, and when staying within the Salesforce ecosystem is a priority.
What concepts do most Heroku Architect candidates get wrong?
The most commonly misunderstood topics for the Heroku Architect exam are: (1) Heroku Connect: Bidirectional Sync Has Write Conflict Rules; (2) Dyno Types: Web vs Worker vs One-Off; (3) Add-ons vs Buildpacks: Services vs Build Configuration. Candidates are most confidently wrong on these. Learn the distinctions early to avoid losing marks on questions you expect to get right.
Why do most Heroku Architect candidates fail questions about Heroku Connect?
Heroku Connect synchronises data between a Heroku Postgres database and Salesforce objects. By default it is unidirectional (Salesforce to Postgres). Enabling bidirectional sync introduces write conflicts when both sides update the same record. Candidates assume bidirectional sync just works, the exam expects knowledge of the "Salesforce wins" vs "database wins" conflict resolution setting and...
Why do most Heroku Architect candidates fail questions about Dyno Types?
Web dynos handle HTTP requests and must respond within 30 seconds (otherwise Heroku returns a timeout). Worker dynos run background jobs with no time limit (used for long-running processes, queue processing). One-off dynos are for ad-hoc tasks (database migrations, scripts). Candidates run long-running jobs on Web dynos, the exam expects Worker dynos for async processing.
Why do most Heroku Architect candidates fail questions about Add-ons vs Buildpacks?
Add-ons are third-party services attached to a Heroku app (Postgres database, Redis cache, monitoring tools); they are provisioned and billed separately. Buildpacks are scripts that transform your code into a runnable slug at build time (Node.js buildpack installs npm packages, Python buildpack installs pip packages). Candidates confuse add-ons (runtime services) with buildpacks (build-time co...

Related Study Resources

Start Heroku Architect Prep

After this exam, consider Application Architect or System Architect next.

Share or cite this study guide

Free to link from a forum post, study repo or article. Copy the Markdown and paste it anywhere.