The Setup

Imagine a sales dashboard:

• Source Analysis: A table of product categories (e.g., Electronics, Clothing, Books), with a field CategoryID (e.g., ELEC, CLTH, BOOK).
• Target Analysis: A sales trends view, filterable by a parameter pCategory.

Goal: Click a category (e.g., Electronics, CategoryID = ELEC) in the table and open the trends, pre-filtered to that category.

The Solution

I used a URL action in the source analysis:

https://us-east-1.quicksight.aws.amazon.com/sn/analyses/b7b1cc83-6833-4047-8625-8f2ac8db15a4#p.pCategory=<<CategoryID>>

Steps

Source Analysis:

• Opened my table visual showing categories.
• Confirmed CategoryID was a column (e.g., ELEC, CLTH).
• Selected the table > “Actions” pane > “+ Add action” > “URL action.”
• Entered the URL: https://us-east-1.quicksight.aws.amazon.com/sn/analyses/b7b1cc83-6833-4047-8625-8f2ac8db15a4#p.pCategory=<<CategoryID>>.
• Mapped <<CategoryID>> to the CategoryID field in the dropdown.
• Set activation to “Select” (left-click).
• Saved.

Target Analysis:

• Already had a parameter pCategory linked to a filter on the category field.
• Tested with a hardcoded URL: https://us-east-1.quicksight.aws.amazon.com/sn/analyses/b7b1cc83-6833-4047-8625-8f2ac8db15a4#p.pCategory=ELEC—worked perfectly.

Tested:

• Clicked a row with CategoryID = ELEC.
• New tab opened: https://us-east-1.quicksight.aws.amazon.com/sn/analyses/b7b1cc83-6833-4047-8625-8f2ac8db15a4#p.pCategory=ELEC.
• Sales trends filtered to Electronics. Repeated for CLTH and BOOK — all good.

How It Works

• Source: The table has CategoryID. Clicking a row (e.g., ELEC) tells QuickSight to grab that value.
• URL Action: <<CategoryID>> pulls the clicked row’s CategoryID and plugs it into the URL.
• Target: The pCategory parameter receives the value (e.g., ELEC) and filters the visuals.

No parameters needed in the source — just the CategoryID field.

Final Setup

Source Anlysis

• Visual: Table with CategoryID column.
• URL Action:

https://us-east-1.quicksight.aws.amazon.com/sn/analyses/b7b1cc83-6833-4047-8625-8f2ac8db15a4#p.pCategory=<<CategoryID>>

• Mapped to CategoryID.
• Activation: “Select.”

Target Analysis

• Parameter: pCategory, filtering the category field.

Tips

1. Field in Visual: Ensure CategoryID is in the table’s field wells (e.g., “Group by”).
2. Syntax: Use <<field_name>> for dataset fields — no $ unless it’s a parameter.
3. Test Hardcoded: Verify the target with a fixed value first.

Wrap-Up

QuickSight URL actions make dashboards interactive with minimal fuss. Click a category, see its trends — done. If you’re linking analyses, start with <<field_name>> and save yourself a headache. Happy building!

Questions about QuickSight? Hit me up in the comments!

— Shravya Reddy, Cloud Engineer

In AWS QuickSight, when using Athena as a data source with Lake Formation, permissions are crucial for seamless data access. The current permission model uses the credentials of the data source creator to validate permissions through Lake Formation. However, if you need to change the permissions to another user or group, or if the data source creator’s account is deleted, you may encounter permission errors. In this guide, I’ll walk you through a workaround to address this issue.

The Problem

When the data source is created by a user who is later deleted, QuickSight cannot change the data source creator to a new user. This leads to permission errors because Lake Formation continues to verify permissions based on the original creator’s credentials. Currently, QuickSight does not support updating the data source creator.

The Workaround

A practical workaround involves assigning a RoleArn to the data source and then deleting the data source creator from QuickSight. This removes the dependency on the original creator. Once the creator is deleted, Lake Formation will validate permissions through the RoleArn assigned to the data source.

Step-by-Step Solution

1. Create an IAM Role

• Create an IAM role, e.g., “quicksight_role_athena_test”.
• Select “Custom trust policy” and add the below policy

{ 

    "Version": "2012-10-17", 

    "Statement": [ 

        { 

            "Effect": "Allow", 

            "Principal": { 

                "Service": "quicksight.amazonaws.com" 

            }, 

            "Action": "sts:AssumeRole" 

        } 

    ] 

} 

• Provide Administrator permissions to this role.

2. Assign the Role to the Data Source

• Use the following CLI command to assign the role to the data source:

aws quicksight update-data-source --aws-account-id <account-id> --data-source-id "<datasource-id>" --name <datasource-name> --data-source-parameters '{"AthenaParameters":{"WorkGroup":"primary","RoleArn":"<role-arn>"}}'

• Replace the placeholders with the actual account ID, data source ID, data source name, Athena workgroup, and role ARN.
• FYI, to get the datasource id — please use the below command to do that

aws quicksight list-data-sources --aws-account-id <account-id>

3. Update Lake Formation Permissions

• Route to Lakeformation.
• Choose Tables -> Select each table at a time and choose Actions -> Grant
• Choose the IAM Role created above -> Choose “Describe”
• Choose the same table again and now in Actions -> Choose “Grant on Target”
• Choose the IAM Role created above -> Choose “Select”
• Make sure to repeat these steps on all tables. Make sure to grant SELECT and DESCRIBE permissions in Lake Formation to the IAM role “quicksight_role_athena_test.”

4. Verify

• Manually refresh the QuickSight.
• Choose datasets and verify if the custom SQL is visible and working as expected.

5. Delete a User

• Delete the user and transfer all the resources to another user

Conclusion

By following these steps, you can eliminate the dependency on the original data source creator, ensuring that QuickSight accesses data through the IAM role permissions. This approach provides a robust solution to permission errors and improves data source management in QuickSight with Athena Lake Formation.

If you found this guide helpful, please share it with your network. For more AWS and QuickSight tips, follow “CloudApplied” on Medium!

— Shravya Reddy, Cloud Engineer

We can spend a day without food and water. But, spending a day without spending money is nearly impossible. We are still fortunate that food and water are still part of our physical realm while spending money has gone to the virtual realm with the internet. So, it is good to spend some time on how our hard-earned money traverses this virtual realm from the customer’s plastic card to the provider’s metal vault.

Let’s understand the payment process through a romantic story that happened in California, USA.

Surya decides to buy a beautiful bouquet of roses for his wife Cynthia on her birthday. Since he works all day, he doesn’t have time to go to a florist and get to build a bouquet so he wants to order a flower bouquet online on Floral Mark’s website so that it gets delivered on the day of her birthday. He adds a bouquet of roses and lilies to the website’s online cart and goes to the payment page, where he enters his VISA credit card details and clicks on the “Place the Order” button. The order is successfully created with necessary details provided to the user.

Here is where the story enters the virtual world of payments. The card details entered by Surya are captured by a Payment Gateway and transfers them to Payment Processor, they initiate settlement request to the Payment Networks.

· Payment Gateway: Collects card details securely. For example, Stripe, Apple Pay, Google Pay, etc.

· Payment Processor: Bridges communication between banks and payment networks. For example, Square, PayPal, etc.

· Payment Network: Validates and authorizes transactions. For example, VISA, MasterCard, Rupay, etc.

To pay for $80 flower bouquet he ordered, Surya entered his VISA card details in Floral Mark’s Website, whose payment gateway is Stripe collects his card details. Then securely transfers it to a payment processor like Square to handle authorization of payment from Surya’s card, routes it to the VISA payment network. VISA validates the transaction, applies security rules, and helps route the payment authorization and settlement messages to all the parties involved in this transaction (Payment Processors, Sender, and Receiver’s bank). If Surya has sufficient credit in his credit card, the charge for the bouquet is debited from his credit card and credited to Floral Mark’s bank account. Else Payment Network will send a decline message to the payment processor from there Surya would have received a transaction failure message. Surya is not only a lovely husband but resourceful husband, the transaction succeeds.

All the above would have been a very linear transaction if Floral Mark was a local business, handles only one currency. Floral Mark operates in multiples countries — UK, USA, India, and Dubai, and each of the countries has got its own currency. The finance teams of Global Businesses like Floral Mark that handle multiple currencies has also have to handle multiple payment networks. For example, VISA, and MasterCard have their presence in most of the countries. However, networks like Discover and Rupay are limited to USA and India respectively. American Express, it’s a complicated, let’s not get there. All Said and done, looks like Surya and Cynthia’s story is sorted, but Floral Mark’s journey is not a bed of roses as their finance team must handle transactions from various countries.

Be it physical world or virtual world there is no problem that doesn’t have any solution. Businesses like Floral Mark needs a Payment Orchestration System.

Payment Orchestration is a smart traffic controller that knows the most efficient way for each payment to travel from the customer’s bank to the merchant’s account. It’s a centralized system that manages which payment processor (the service that routes the payment) should handle each transaction based on certain criteria, like location, payment type, currency, and fees.

Floral Mark works with multiple Payment Gateways, Payment network providers, Banks, countries, and currencies. Having a linear relationship between the customer and their bank is near to impossible. Having a solid payment orchestration helps them consolidates integrations with various payment methods into single platform. They handle routing, security, risk assessment, and compliance with various regional regulations.

Payment Orchestrators broadly consider customer’s location, card type, currency, etc. and decides which payment processor to route the transaction to. This optimizes transaction costs, minimize failures as the payment orchestrating platform aims to route the transaction to the processor that’s best for that customer’s location or payment method. Before sending the payments to the payment processors, they add additional security layer for fraud checks, 3D Authentication (OTPs to the customers), etc. From there Authorization from the customer’s bank, Settlement process (The card network coordinates with the Customer and merchant’s banks to finalize amounts, apply fees, and move funds to the merchant’s bank), and finally notifies all parties (API calls to the systems to commit the transaction. Transaction message to the customers).

Payment Orchestration platforms generally offer value added services like Analytics, Best practices, and so on.

In summary, payment orchestration is especially valuable for businesses like Floral Mark that operates with complex, high-volume, or global payment needs. It streamlines processes, optimizes costs, and enhances the customer experience for them. For smaller companies with regional presence, it is good to have than a necessity.

In the end, Cynthia smiled over her flower bouquet by her Husband making her birthday more special, and small yet significant win for the Floral Mark in their daily global sales marathon.

Thanks for Reading…

-Krishna Kuntamukkala

Throughout history, mankind has witnessed revolutionary innovations that have transformed the way we live. These innovations are often grouped into generations and referred to as the “Industrial Revolution.” Each revolution has reduced physical effort and increased productivity in everything we do. As Salesforce Co-Founder and CEO, Marc Benioff, mentioned in his book “Trailblazer,” “We’re at a crossroads, and the actions we take from this moment forward will have a defining impact on what kind of world we leave behind for future generations.”

The AI generation, which we are currently in, can be considered the 4th Industrial Revolution. AI technologies, like ChatGPT, are now capable of responding in ways that were unprecedented just a few years ago. The responsiveness of these systems is expected to improve exponentially given the significant effort being invested in AI.

What is Salesforce’s Role in this AI era:

Salesforce has proved its mettle as the world’s number one CRM used by a plethora of companies from start-ups to stalwarts as their CRM system that supports their Sales, Service, Marketing, Revenue, etc. It holds petabytes of data about mentioned business processes. Additionally, integrates with every named existing software available in the market. This level of detailed data is like a feast to any AI system. However, it is imperative that the feast is served on a platter. That is what Data Cloud does.

Data Cloud collects data across the business’s systems and harmonizes it by mapping it to respective fields/objects/records in Salesforce to make it available as a unified customer data profile. Now that we have data in place, it is important to understand how it can be used by businesses effectively. Here is where Customer 360 comes into the picture.

Customer 360 refers to all the capabilities Salesforce has to offer (Sales, Service, etc.) that allow businesses to have a holistic view/360 view of their customers. With the data in place, puts the businesses in a great spot to understand and analyze customers said and unsaid needs and cater to customers in a better way. CRM Analytics is very effective in getting this done.

CRM Analytics, previously known as Tableau Analytics allows to creation of reports with great visualizations with a touch of Prescriptive and Predictive analytics. Based on the analysis, the business can decide on how to make the Customer–business relationships better. How about the decision maker having a super-intelligent trusted assistant that not only helps in making better decisions but takes care of repetitive tasks? You got it right, here comes Salesforce’s most celebrated Einstein One platform.

Einstein One powers all the core salesforce clouds (Sales, Service, Marketing, Revenue, etc.) with Artificial Intelligence. This is a hyperforce platform built on AWS that holds services of Einstein Copilot, Prompt Builder, and Model Builder assured with Einstein Trust layer. The data from various sources is grounded in the Salesforce platform to respond to the user’s prompts. And, can do some tasks for the users. For example, sending a promotional offer to all the users in the United States.

What you can do as a Salesforce Practitioner?

If you are already using the might of Salesforce AI capabilities, good for you. If you want to play around with Salesforce’s Einstein you can use the below links to play around with Salesforce Einstein.

Prompt Builder Hands-On

• https://trailhead.salesforce.com/content/learn/projects/quick-start-prompt-builder/get-started-with-prompt-builder

Einstein Copilot Hands On

• https://trailhead.salesforce.com/content/learn/projects/quick-start-einstein-copilot/connect-your-flows-to-einstein-copilot

Data Cloud Hands On

• https://trailhead.salesforce.com/content/learn/projects/create-a-data-stream-in-data-cloud?trailmix_creator_id=dataclouddocs&trailmix_slug=data-cloud-learning-path-a

As a Salesforce practitioner, you are likely intrigued by Salesforce’s role in this 4th Industrial/AI Revolution. If you are interested in knowing more, please reach out to kloudeffect@gmail.com and follow us on YouTube (CloudApplied — YouTube) for the latest updates on the space of Salesforce and AI