PDI testing - Platform Developer I Updated: 2024

eWEEK content and product recommendations are editorially independent. We may make money when you click on links to our partners. Learn More.

Salesforce blends powerful built-in capabilities with the flexibility to create your own low-code capabilities. But as any seasoned developer will tell you, building applications is the easy part. The real challenge is ensuring they don’t break over time and having robust protections in place to allow you to continue to build – and experiment – without fear of failure.

This is why testing, especially automated testing, is so important. Testing is a key aspect of achieving digital transformation. No matter the size of the company, if teams aren’t emphasizing testing, they won’t achieve the speed and reliability of changes they need.

Interested in knowing more about how teams are managing their testing processes, Copado surveyed more than 275 Salesforce professionals, who shared details of their teams’ strengths and weaknesses on testing. Here are a few big takeaways:

• 41% of teams don’t have time to test every change before each release.
• 92% experience production issues due to inadequate testing.
• 84% still rely at least partially on manual testing.
• 51% of organizations have 25+ full-time QA resources (20% have 100+ resources).
• 95% have a higher total cost of ownership for manual testing versus commercial test automation.

Also see: Digital Transformation Guide: Definition, Types & Strategy

Common Testing Challenges

Testing needs to happen at many levels, from unit tests to cross-system UI testing. Salesforce unit tests are helpful, but the unit testing culture among Salesforce developers is far weaker than in other languages like Ruby.

Salesforce mandates 75% unit test code coverage to deploy to production, but because continuous integration wasn’t common on Salesforce until recently, most Salesforce developers don’t fully appreciate the importance of these tests. So, it’s common to see sloppy tests that meet the 75% threshold but provide little or no genuine protection.

UI testing is also tricky on Salesforce, especially on their newer UI, Lightning. Lightning uses a shadow DOM (document object model) for security and performance purposes, but that makes it hard for tools like Selenium that rely on the DOM.

The underlying DOM is also subject to change without warning, which can make tests fail even if the genuine system is working properly. This means that UI testing tools for Salesforce need to have significant intelligence built-in.

Salesforce is also frequently integrated with other systems like ServiceNow, SAP, and Oracle Financials. This means that testing tools need to be able to test all of the connected systems to ensure end-to-end functionality.

As a low-code platform, Salesforce development can move quite quickly. The underlying Salesforce platform is also updated three times a year. These faster software cycles mean more changes and updates that cause unwelcome surprises down the road if you don’t have good tests in place.

System behavior can also depend on the underlying data in the system. So test data management is important, especially when working with complex apps built on top of Salesforce like Salesforce CPQ, Veeva, and nCino.

Lastly, Salesforce is designed to provide low coders a boost. This means that ideally, testing tools should be easy for non-coders to use also.

Also see: DevOps, Low-Code and RPA:  Pros and Cons 

A Few Trends in Salesforce Testing

As Salesforce matures in scope and sophistication, organizations need to be ready for high-quality, cross-cloud customizations and verify third-party integrations across other technologies and platforms.

Teams in our survey (41%) said they don’t have enough time to sufficiently test all changes before a release. Our analysis shows that teams feel pressure to deliver features because of aggressive project timelines and that development takes longer than expected.

Testing, ultimately, gets thrown on the backburner as a result. Teams often fail to provide testing the respect it deserves, and relegate it to the end of a sprint, rather than following test-driven development or other methods of shifting quality left.

Also see: What Does 2022 Hold for Intelligent Automation

Low Code Testing in Early Stages

Low-code platforms like Salesforce will continue to dominate the digital landscape as every business migrates its infrastructure to the cloud. The speed and ease of building on these platforms is unmatched, but they require robust testing to make sure changes don’t break existing systems.

Low-code platforms remain in the early stages of testing. Based on Salesforce’s research on its professionals, the majority of teams still mainly rely on manual efforts. But the rise of DevOps and the acceleration of change means that automated testing is the next frontier for today’s digital businesses.

About the Author: 

Andrew Davis, Salesforce DevOps Specialist and Senior Director of Research and Innovation, Copado.

As we transition from one year to the next, it's a season of reflection and looking forward. As an analyst, the end of the year is a time to learn from past work, analyze its outcomes and consider its potential impact on the future.

In 2023, enterprise data management (EDT) solutions underwent significant changes due to the influx of generative AI technologies. These technologies have fundamentally altered how businesses approach data management, analysis and usage. In this post, I’ll review some of 2023’s highlights in this field.

How Different Areas Of EDT Are Evolving

Over the past year, there have been promising developments in EDT across several key areas. These include data management itself, where the focus has been on using AI to Excellerate how data is organized and accessed. The data cloud sector has also experienced growth, with more businesses adopting cloud-based solutions because of their flexibility, scalability and facility for integrating tools that handle unstructured data.

In data protection and governance, there has been a continuous effort to enhance security measures to safeguard sensitive information. Database technologies have also improved, particularly in handling and processing large data volumes more efficiently by incorporating generative AI.

Recent advancements in data integration and intelligent platforms have been geared towards better aggregating data from multiple sources, allowing for more comprehensive data analysis. The integration of AI and ML has further enhanced the capabilities of these platforms, improving data analysis interpretation and offering more profound and insightful analytical outcomes.

Full disclosure: Amazon Web Services, Cisco Systems, Cloudera, Cohesity, Commvault, Google Cloud, IBM, LogicMonitor, Microsoft, MongoDB, Oracle, Rubrik, Salesforce, Software AG, Splunk, and Veeam are clients of Moor Insights & Strategy, but this article reflects my independent viewpoint, and no one at any client company has been given editorial input on this piece.

Bringing AI To Data Management—And Vice Versa

“In a way, this AI revolution is actually a data revolution,” Salesforce cofounder and CTO Parker Harris said during his part of this year’s Dreamforce keynote, “because the AI revolution wouldn't exist without the power of all that data.” Harris's statement emphasizes the vital role of data in businesses and points to the increasing necessity for effective data management strategies in 2024.

As data becomes more central, the demand for scalable and secure EDT solutions is rising. My latest series of articles focusing on EDT began with an introductory piece outlining its fundamental aspects and implications for business operations. This was followed by a more in-depth exploration of EDT, particularly highlighting how it can benefit businesses in data utilization. These articles elaborated on the practical uses and benefits of EDT and its importance in guiding the strategies and operations of modern businesses.

As businesses continue to leverage generative AI for deeper insights, the greater accessibility of data is set to revolutionize how they manage information. This development means enterprises can now utilize data that was previously inaccessible—a move that highlights the importance of data integration for both business operations and strategic decision-making. For instance, untapped social media data could offer valuable customer sentiment insights, while neglected sensor data from manufacturing processes might reveal efficiency improvements. In both cases, not using this data equates to a missed opportunity to use an asset, similar to unsold inventory that takes up space and resources without providing any return.

Revolutionizing Data Cloud Platforms

Incorporating AI into data cloud platforms has revolutionized processing and analyzing data. These AI models can handle vast datasets more efficiently, extracting previously unattainable insights due to the limitations of traditional data analysis methods.

Over the year, my own collaborations with multiple companies suggested the range of technological progressions. As I highlighted in a few of my articles, Google notably improved its data cloud platform and focused on generative AI with projects including Gemini, Duet AI and Vertex AI, reflecting its solid commitment to AI innovation. Salesforce introduced the Einstein 1 Platform and later expanded its offerings with the Data Cloud Vector Database, providing users with access to their unstructured enterprise data, thus broadening the scope of their data intelligence. IBM also launched watsonx, a platform dedicated to AI development and data management. These moves from major tech firms reflect a trend towards advanced AI applications and more sophisticated data management solutions.

At the AWS re:Invent conference, I observed several notable launches. Amazon Q is a new AI assistant designed for business customization. Amazon DataZone was enhanced with AI features to Excellerate the handling of organizational data. The AWS Supply Chain service received updates to help with forecasting, inventory management and provider communications. Amazon Bedrock, released earlier in the year, now includes access to advanced AI models from leading AI companies. A new storage class, Amazon S3 Express One Zone, was introduced for rapid data access needs. Additionally, Amazon Redshift received upgrades to Excellerate query performance. These developments reflect AWS's focus on integrating AI and optimizing data management and storage capabilities.

Recent articles have highlighted Microsoft's role in the AI renaissance, one focusing on the launch of Copilot as covered by my colleagues at Moor Insights & Strategy, and another analyzing the competitive dynamics in the AI industry. Additionally, Microsoft has expanded its data platform capabilities by integrating AI into Fabric, a comprehensive analytics solution. This suite includes a range of services including a data lake, data engineering and data integration, all conveniently centralized in one location. In collaboration, Oracle and Microsoft have partnered to make Oracle Database available on the Azure platform, showcasing a strategic move in cloud computing and database management.

Automating Data Protection And Governance

With the growing importance of data privacy and security, AI increasingly enables the automation of data governance, compliance and cybersecurity processes, reducing the need for manual oversight and intervention. This trend comes in response to the rise in incidents of data breaches and cyberattacks. AI-driven systems have become more proficient at monitoring data usage, ensuring adherence to legal standards and identifying potential security or compliance issues. This makes them a better option than traditional manual approaches for ensuring data safety and compliance.

Security is not only about protecting data but also about ensuring it can recover quickly from any disruptions, a quality known as data resilience. This resilience has become a key part of security strategies for forward-thinking businesses. Veeam emphasized “Radical Resilience” when it rolled out a new data protection initiative focused on better products, improved service and testing, continuous releases and greater accountability. Meanwhile, Rubrik introduced its security cloud, which focuses on data protection, threat analytics, security posture and cyber recovery. Cohesity, which specializes in AI-powered data security and management, is now offering features such as immutable backup snapshots and AI-driven threat detection; in 2023, it also unveiled a top-flight CEO advisory council to influence strategic decisions. Commvault has incorporated AI into its services, offering a new product that combines its SaaS and software data protection into one platform.

LogicMonitor upgraded its platform for monitoring and observability to include support for hybrid IT infrastructures. This enhancement allows for better monitoring across an organization's diverse IT environments. Additionally, Cisco has announced its intention to acquire Splunk. This acquisition will integrate Splunk's expertise in areas such as security information and event management, ransomware tools, industrial IoT vulnerability alerting, user behavior analytics and orchestration and digital experience monitoring that includes visibility into the performance of the underlying infrastructure.

Key Changes for Database Technology

Advancements in AI and ML integration are making database technology more intuitive and efficient. Oracle Database 23c features AI Vector Search, which simplifies interactions with data by using ML to identify similar objects in datasets. Oracle also introduced the Fusion Data Intelligence Platform, which combines data, analytics, AI models and apps to provide a comprehensive view of various business aspects. The platform also employs AI/ML models to automate tasks including data categorization, anomaly detection, predictive analytics for forecasting and customer segmentation, workflow optimization and robotic process automation.

In my previous discussion about IBM's partnership with AWS, a major highlight is the integration of Amazon Relational Database Service with IBM Db2. This collaboration brings a fully managed Db2 database engine to AWS's infrastructure, offering scalability and various storage options. The partnership between AWS and IBM will likely grow as the trend of companies forming more integrated and significant ecosystems continues.

Database technology also evolved with MongoDB queryable encryption features for continuous data content concealment. MongoDB Atlas Vector Search now also integrates with Amazon Bedrock, which enables developers to deploy generative AI applications on AWS more effectively. It’s also notable that Couchbase announced Capella iQ, which integrates generative AI technologies that exploit natural language processing to automatically create trial code, data sets and even unit tests. By doing this, the tool is streamlining the development process, enabling developers to focus more on high-level tasks rather than the nitty-gritty of code writing.

Leveraging Data Integration Platforms

Generative AI technologies have also improved data integration capabilities by using historical data, analyses of trends, customer behaviors and market dynamics. This advancement is particularly influential in the finance, retail and healthcare sectors, where predictive insights are critical for strategic and operational decisions. There's been a shift towards adopting data lake house architectures, which combine the features of data lakes and data warehouses to help meet the challenges of handling large, varied data types and formats, providing both scalability and efficient management. This evolution in data architecture caters to the growing complexity and volume of data in various industries.

Integrating various data sources is crucial for many companies to enhance their business operations. Software AG has introduced Super iPaaS, an evolution of the traditional integration platform as a service (iPaaS). This advanced platform is AI-enabled and designed to integrate hybrid environments, offering expansive integration capabilities. Cloudera has also made strides with new data management features that incorporate generative AI, enabling the use of unstructured data both on-premises and in cloud environments. Its hybrid approach effectively consolidates client data for better management. Informatica's intelligent data management cloud platform integrates AI and automation tools, streamlining the process of collecting, integrating, cleaning and analyzing data from diverse sources and formats. This creates an accessible data repository that benefits business intelligence and analytics.

That’s a Wrap!

In my collaborations throughout the year with various companies, one key theme has emerged in this AI-driven era – data has become even more fundamentally important for businesses. It's clear that the success of AI heavily relies on the quality of the data it uses, and AI models are effective only when the data they process is accurate, relevant and unbiased.

For example, in applications such as CRM or supply chain optimization, outcomes are directly influenced by the data’s integrity. Instances where AI failed to meet expectations could often be traced to poor data quality, whether it was incomplete, outdated or biased. This year has highlighted the necessity of not just collecting large amounts of data but ensuring its quality and relevance. Real-world experience underscores the need for strict data governance and the implementation of systems that certain data accuracy and fairness, all of which are essential for the effective use of AI in business.

As AI technology advances and data quality improves, the use of generative AI in understanding and engaging with customers is becoming ever more prominent. Backed by good data management, this enhances the customer experience by making the customer journey more personalized and informative. It allows businesses to gain valuable insights from customer interactions, helping them continuously refine and Excellerate their offerings and customer relations. I expect this trend to grow, further emphasizing the role of AI in customer engagement and shaping business strategies. In fact, this symbiotic relationship between AI-driven personalization and customer engagement is becoming a cornerstone of not only data management strategy but modern business strategy overall, significantly impacting how companies connect with their customers.

Wrapping up, it's evident that the emphasis on data quality is critical for improving AI's performance. Data management, cloud services, data protection and governance, databases, data integration and intelligent platforms have all significantly contributed to the advancement of AI. In 2024, I expect we’ll see even more emphasis on ensuring the accuracy and relevance of data so that AI can provide dependable insights.

The sforce Developer Program for Microsoft Office includes free software tools, a marketplace for developers to sell their custom wares to Salesforce.com customers, and an online community for Office developers that includes message boards and integration examples.

The software, dubbed "sforce Toolkit for Office," is a Microsoft Office plug-in that helps developers access the sforce API directly from within Office 2000 and later applications. The kit is in beta and can be downloaded from the Salesforce.com site.

"This program builds on our vision of providing the world's most customizable CRM by allowing our customers to share and manage customer information from not just their Web browsers, but also via integration with Microsoft's Office," said Marc Benioff, Salesforce.com's chief executive, in a statement.

Salesforce.com is a set of sales and customer relationship management (CRM) Web applications that are typically accessed via a browser. The San Francisco-based company launched an Office edition of its service over a year ago that delivers data to Microsoft's suite, but Wednesday's announcement is an attempt to convince developers to build custom Office applications that link with Salesforce.com.

The announcement was made at the Microsoft Office System Developer Conference, which opened Wednesday and runs through Friday. The first-ever such conference, it's part of Microsoft's efforts to boost adoption of Office 2003 and its use as a XML-centric developer platform. Chairman Bill Gates is scheduled to speak at the meeting on Friday.

It may sound counter-intuitive to say that developers shouldn’t perform testing on the products they produce. After all, who knows a site or app better than those who created it? Aren’t developers exactly the people who should be testing software, given that they put it together and know how it’s supposed to work?

It’s this common-sense assumption that lies behind the idea that developers should assist test teams with strategies like unit testing—if they’re not taking complete responsibility for testing themselves. The latter approach—having all testing, GUI and functional, carried out by developers rather than a dedicated QA team—has been adopted by many development and digital agencies, which seems logical.

(Related: How testing works in an agile world)

It also seems prudent in terms of costs. Full-time QA teams are expensive to maintain and, because of the intermittent nature of software testing, are often underutilized. From this perspective, having developers perform testing as and when it’s needed makes more sense than employing testers who will always have too much downtime.

But as rational as all of this may sound, it’s not a good idea to do without professional software testers. The reason? Software is often more likely to be released with bugs when tested by developers, not less likely.

There are several contributing factors here, but one of the most important is that testing overburdens developers with workloads that are too heavy. A developer who has spent all day coding and is then required to put in a couple of hours testing is likely to tire quickly, lose focus and make mistakes, resulting in crucial bugs being missed.

This creates a need for post-go-live fixes, which further increase workloads and make it even more likely that exhausted developers will miss bugs on other testing projects, too. A self-defeating cycle can result, where all efforts to ensure that software works better end up making it worse.

Another contributing factor is the fact that, often, developers simply don’t have enough time to perform testing effectively. When testing has to be fitted in around development work and completed within the standard timeframe of the two or three days before a site or app goes live, it is difficult for developers to ensure good functionality by covering a broad range of Web and mobile platforms.

And broad device coverage is becoming increasingly indispensible to success online. As the mobile device market continues to expand, and as apps and websites are used on a wider array of smartphones and tablets, any piece of software that has only been tested on a handful of devices runs the risk of functioning poorly for many of its users.

Compounding these issues of time and energy is the repetitive nature of software testing itself. As a website is tested on more and more devices and browsers, the individual doing the testing can quickly develop expectations as to how that site should look on the next device or browser, and these expectations can come to replace the perception of what’s actually there, leading to bugs being missed.

For this reason, and generally speaking, a single individual should not test the same website on more than two different browsers or devices, in order to remain effective. This principle, however, can be hard to put into practice when a handful of developers need to try to test software on multiple platforms as quickly as possible.

And, in fact, developers are at a further disadvantage here precisely because they know the software they produce so well. An intimate knowledge of the functionality and design of sites and apps can lead to developers having stronger preconceived notions of what to expect when testing than professional testers, making it even more likely that obscure bugs will be overlooked.

In addition to these all of these pressures and constraints, there’s the crucial fact that software testing, as a discipline, requires a very specific set of skills, many of which are not necessarily shared with the practice of software development.

From effectively dividing up a piece of software for unit testing, to writing test scripts, deciding which aspects of a site would benefit from exploratory testing, and collating and acting on the results of successive test cycles, professional software testers structure their work in many different ways to ensure that software is exposed to the greatest amount of scrutiny possible.

Although developers may achieve a certain amount of success with a largely exploratory approach to testing their software, there can be no substitute for highly structured testing that utilizes years of professional knowledge and expertise.

In light of these drawbacks and complications, the strategy of having developers perform testing looks shortsighted and counter-productive. Not only does it often result in bugs on apps and websites escaping into the live environment and damaging user experience, it also degrades developers’ wellbeing and the quality of their work.

As well as putting exhausted developers into a self-defeating cycle of ineffective testing and post-go-live fixes, having them test software can harm their creativity and attention to detail while developing. This can pose a significant threat to the long-term vitality of development and digital agencies, which need to invest in working with professional software testers to safeguard the quality of their output.

Recognizing that testing is not a relatively unimportant, extra task that can be assigned to developers, but an essential component in the production of outstanding software, will thus enable agencies to keep their service levels high, not to mention their developers’ energy and enthusiasm, too.

Salesforce Inc. today announced the general availability of Unlimited Edition+, which it describes as a simplified technology package of the technology it uses in its Sales Cloud, Service Cloud, Einstein artificial intelligence, Data Cloud, Slack, Tableau and Industries applications.

The company also announced a marketplace for people with Salesforce skills to find jobs.

Citing its own research that found that 65% of senior IT executives say they can’t justify implementing generative AI at their companies right now for a variety of technical and reliability reasons, Salesforce said UE+ simplifies the process of buying and implementing its technology, including analytics and AI.

UE+ for Sales and Service, which was announced at the company’s Dreamforce conference in September, includes everything in the Sales Cloud Unlimited edition plus Data Cloud, Einstein credits, Sales Enablement, Sales Planning, Maps, Revenue Intelligence, Slack and Slack Sales Elevate. Sellers can use data such as opportunity history, open cases and latest prospect contacts to customize pitches to prospects.

Service Cloud UE+ includes everything in the Service Cloud Unlimited edition plus Data Cloud, Einstein Credits, Service Cloud Voice, Digital Engagement, Feedback Management, Self-Service, Slack and customer relationship management analytics. It can be used to resolve cases more quickly and allow agents to find answers using generative AI.

UE+ for Industries is a new package that combines Unlimited+ for Sales and Service with industry-specific data models and features aimed at the financial services, healthcare and manufacturing industries.

Financial Services Cloud UE+ for Sales and Service helps companies in those industries connect all of their customer data on one platform with AI used for personalization.

Health Cloud UE+ for Service enables healthcare companies to Excellerate contact center response times and offer digital healthcare services with real-time collaboration and comprehensive views of patients, providers and partners.

Manufacturing Cloud UE+ for Sales embeds AI capabilities across the sales cycle, enabling manufacturers to get a high-level view of how their companies are performing against negotiated sales agreements and creating AI-generated summaries to determine where to prioritize resources.

The Trailblazer Career Marketplace is an extension of Salesforce’s Trailhead online learning platform where job-seekers can connect with prospective employers. Candidates can create a unified view of their relevant Salesforce experience on their profile, including work history, projects, Tested skills and credentials. They can also add an introductory video and look for job opportunities using a filtered search tool.  Job-seekers have the option of masking personal data like their full name and profile picture to reduce the risk of bias in the hiring process.

Employers, who must be Salesforce partners, can search for candidates by product-specific experience, certifications, years of experience and other factors. They can also publish video clips and custom profiles of their companies that list open jobs.

Citing International Data Corp. research, Salesforce says it will create 11.6 million net new jobs by 2028.

Photo: Salesforce

Salesforce recently added another lot of AI functionality to its Einstein 1 Platform. Among other things, vector databases are now better supported. In addition, the capabilities of the generative AI assistant Einstein Copilot have also been expanded.

With all the new AI-enabled functionality Salesforce is now bringing to its renamed Einstein 1 Platform, teams of users will soon be able to plug AI into their workflows more efficiently.

In this regard, it does not matter whether they build their AI-based application for the Salesforce platform or use the generative Einstein Copilot assistant to retrieve required data for their (development) projects.

For this, the new functionality should combine the elements from Salesforce Data Cloud and Einstein Copilot to the various CRM applications and Einstein Copilot Studio for building AI-based applications.

Enhanced vector database support

New features include enhanced support for vector databases in Salesforce Data Cloud. This now converts unstructured data into a workable vector format. This data is combined with structured information that should make generative AI tooling and analytics available in workflows for various Salesforce CRM solutions.

This, in turn, should also provide better insights and analytics via prompts.

Ultimately, this vector database support makes it easier for developers to use their combined unstructured and structured data to train various LLMs. The company promises that fine-tuning of these models will be unnecessary from now on.

Update Einstein AI Copilot

The update for the generative Einstein AI Copilot assistant also offers this enhanced vector database functionality. This technology is intended to support the new AI feature Einstein Copilot Search in the generative assistant.

Einstein Copilot Search helps users query all unstructured and structured business data. The exact results of these search queries are then brought up directly in users’ various workflows.

Finally, links are provided to the underlying source material so users can be confident that the data on which the answers are based is reliable.

Availability

The generative AI assistant Einstein AI Copilot will not become generally available until February 2024, according to Salesforce. At the same time, the company will also test vector database support and enhanced AI Search.

Also read: Salesforce needs AI from AWS to strengthen its offerings