Microsoft ships two major release waves per year for Dynamics 365, plus a continuous stream of Proactive Quality Updates and hotfixes in between. For most enterprise teams, the real problem is not that any single update is hard to test – it is that testing never stops, and tooling built for a slower era keeps breaking under the pace.
Two things changed in 2026. Microsoft set a firm end-of-support date for RSAT (15 May 2027), and Wave 1 introduced autonomous AI agents inside Business Central and F&O – software that independently posts invoices, creates orders, and assigns GL accounts without a human in the loop.
This guide covers the leading AI-assisted D365 testing tools: what sets them apart, what each does best, and how to choose the right one for your environment.
In this article:
- Why traditional D365 automation keeps failing
- RSAT end of support – what it means in practice
- The new challenge: validating AI agents
- Review of 8 tools: XPLUS, ACCELQ, Tosca, TestComplete, Elevaite365, RSAT, EasyRepro, Playwright/Selenium
- Comparison tables and evaluation framework
- Free RSAT migration path: XPLUS Freemium Testing
Why Traditional D365 Automation Keeps Failing
The common complaint is not a lack of automation – it is the maintenance cost of the automation teams already have. Scripts written against a specific UI state are tightly coupled to it. A relabelled field or a moved button from a release wave causes the test to fail – not because the process broke, but because the script described pixels instead of intent.
The result is predictable: teams quietly narrow regression to a handful of “critical” scenarios, accept the risk on the rest, and still firefight after every wave. AI-assisted tools attack the problem at the root – through self-healing (tests adapt to UI changes instead of breaking), intent-based authoring (tests created from recordings or plain language), and codeless ownership (functional users run tests without developers).
RSAT: End of Support on 15 May 2027
Per Microsoft’s own documentation, RSAT is marked for deprecation and will not be supported after 15 May 2027 – no technical support, bug fixes, updates, or new features. You can keep running it, but every issue becomes yours alone to solve.
For most F&O teams, this confirms what they had suspected for years. RSAT had not been a major area of Microsoft investment for some time, and it carried a real ceiling:
- Heavy front-loaded setup (environments, Azure DevOps test plans, authentication)
- Brittle recordings tied to the UI as it existed at recording time
- Maintenance that compounds with the size of your library
- F&O-only scope – no coverage for Customer Engagement or integration testing
Business Central, worth noting, never had an RSAT equivalent — its AL Test Framework serves a different, developer-oriented purpose — so BC teams start from a completely different place.
The timeline matters: teams with hundreds of processes and years of recorded tests cannot migrate in a weekend. Mapping the library, converting coverage, and validating against a live wave all compete with go-lives and period closes. Start in 2026 and you arrive prepared; wait until late 2026 and you will migrate in the middle of everything else.
The New Challenge: Validating AI Agents
Wave 1 2026 introduced AI agents acting autonomously inside D365 – matching invoices, creating orders, applying pricing rules, posting to the ledger. Testing these is a different category of work.
The old question was: “did the workflow complete?” The new one is: “did the agent make the correct decision?” An agent can run a workflow flawlessly and still assign the wrong GL account or apply outdated pricing. Validating that against outcomes auditors can defend is not something traditional UI automation – or the AL Test Framework – was built for.
The honest state of the market: most tools handle the deterministic layer well, and agent validation is still maturing. If you are deploying Copilot or Agent365 features, plan an agent-validation strategy separately from process regression — they are not the same problem.
Dynamics 365 Testing Tools Compared
Quick comparison table below, followed by a detailed review of each tool:
1. Testing by XPLUS (Executive Automats 365)
Best for: full F&O and Business Central regression, RSAT migration, teams that want a Dynamics 365-focused tool
Testing by XPLUS (also known as Executive Automats 365, listed among Microsoft’s regression tooling options on Microsoft Learn) is a codeless, autonomous regression solution for D365 F&O and Business Central. This deep Dynamics 365 focus is a deliberate design choice – the team knows Dynamics inside out and stays on top of what One Version and PQU updates commonly affect, and pairs the software with an implementation team rather than handing over a tool with a maintenance burden attached.
Key capabilities:
- Codeless recording – functional users own testing without involving developers
- Full regression suites running in hours across all modules and customisations
- Self-healing that repairs a script automatically when a UI element changes and a step would otherwise fail
- AI Optimizer that analyses a script against its goal and adjusts it to keep working — adding missing steps, removing redundant ones, and updating test data
- Performance testing to check how key processes hold up under load
- Automatic documentation feeding directly into audit and compliance workflows
- Native Azure DevOps and Power BI integration
- Business Central support – one of very few tools on the market to cover BC
Where XPLUS pulls ahead: the deep Dynamics 365 focus means the team stays on top of what each One Version update tends to affect, and self-healing plus the AI Optimizer keep post-wave script maintenance to a minimum – real economics, not a marketing promise.
2. ACCELQ
Best for: enterprise teams requiring cross-platform coverage (D365 + Salesforce + others) where D365 is one of several systems under test
ACCELQ is a codeless enterprise platform with AI self-healing and pre-built Dynamics 365 assets (the Dynamics Universe). Forrester Wave 2025 named it a Leader in Autonomous Testing Platforms. The tool covers all D365 modules — Sales, F&O, Customer Service, Supply Chain, Field Service and CI/CD integration spans multiple platforms (Jenkins, Bamboo, Azure, GitLab, CircleCI).
Limitation: ACCELQ is a multi-platform tool for ERP and CRM broadly. For teams focused exclusively on D365, that means greater licensing complexity and no D365-specific tracking of update behaviour across release waves.
3. Tricentis Tosca
Best for: large enterprises with risk-based testing requirements, strong SAP ecosystems, and a need for automatic process documentation
Tosca is an enterprise-grade platform built around model-based test design. When a UI change would otherwise break 200 scripts, updating one module updates all test cases that reference it – which is real maintenance savings at scale. Its strongest competitive advantage is SAP integration, not D365 specifically.
Vision AI is Tosca’s self-healing feature, using image recognition to locate UI elements. Important caveat: Vision AI requires a separate licence on top of the base Tosca platform. Users on G2 and TrustRadius consistently flag high licensing costs and significant implementation complexity as the main drawbacks.
Limitation: higher total cost and implementation complexity. For teams that do not need SAP coverage or cross-platform ERP testing, Tosca brings many capabilities that may never be used.
4. TestComplete (SmartBear)
Best for: teams testing both web and desktop applications, where D365 coexists with legacy software or visually complex apps
TestComplete supports D365 through keyword-driven (codeless) testing as well as scripted tests in JavaScript, Python, or VBScript. It integrates with Azure Test Plans and Azure Pipelines natively. In June 2026, SmartBear announced a new Vision AI capability that unifies property-based, OCR, and visual recognition into a single engine – extending coverage to complex graphical UIs that previously required manual testing.
Limitation: not D365-specific, which means more configuration overhead and no automatic tracking of release-wave behaviour. User reviews on Gartner and PeerSpot note that self-healing is not fully reliable, and parallel execution can be unstable. Post-wave maintenance is still required.
5. Elevaite365
Best for: F&O teams seeking a D365-focused tool with built-in RSAT script migration and a process documentation module
Elevaite365 is a D365-focused platform offering end-to-end automation, process documentation, and — new in 2026 – Test Agent: a tool that converts Copilot actions into structured, repeatable test runs with audit-ready logs. It offers an RSAT script converter and an uncapped pricing model featuring unlimited test scripts (up to 50 per integration licence), unlimited executions, and unlimited users.
Where it stands out: the Test Agent for Copilot action validation is a meaningful step in a direction the market is only beginning to explore. The uncapped pricing model is attractive for smaller teams.
Where XPLUS pulls ahead: XPLUS covers Business Central, which Elevaite365 does not currently support. For organisations with a mixed F&O + BC portfolio, this is a critical difference.
6. RSAT (Regression Suite Automation Tool)
Best for: F&O UAT compliance and only that. Free with F&O licensing.
RSAT is built by Microsoft, required for F&O UAT before production deployment, and free with Finance and Operations licensing. For F&O UAT specifically, it is the right tool and costs nothing extra. End-of-support deadline: 15 May 2027.
Limitation: F&O only, no AI self-healing (every wave requires manual updates), CI/CD limited to Azure Pipelines, no CRM module support.
7. EasyRepro
Best for: developer-led teams testing CRM modules (Sales, Customer Service, Field Service) who need full code control
EasyRepro is a Microsoft-built framework for automating UAT across D365 Customer Engagement modules. It is maintained by Microsoft and updated with each D365 CE release, ensuring native compatibility. Requires C# and Selenium knowledge.
Limitation: CRM only, no codeless option, requires dedicated developer resources and ongoing maintenance.
8. Playwright / Selenium
Best for: developer-led teams with full script control and a budget for ongoing maintenance
Playwright and Selenium are open-source frameworks for web automation. Playwright offers better performance and cross-browser reliability than Selenium, but both require manual management of D365’s dynamic element IDs and regular updates after each release wave. Neither framework self-heals on its own, though AI-assisted tooling can be layered on to repair broken scripts.
Limitation: no D365-specific handling, no built-in self-healing (AI add-ons can help), high maintenance cost. Appropriate only for teams with strong developer resources dedicated to testing.
Choosing an RSAT Replacement
For teams migrating from RSAT before the May 2027 deadline, the key questions are: do you test only F&O or also Business Central, how large is your current test library, and do you have the resources to maintain scripts after every wave.
Evaluation Framework: What to Check Before You Choose
Evaluate tools against your actual situation, not a feature checklist:
- Module coverage – your modules and customisations, not just out-of-the-box flows
- Release-wave alignment – does it self-heal, or will you rewrite after every wave
- Maintenance model – who owns broken tests – this is the biggest hidden cost
- CI/CD integration – native Azure DevOps vs. REST API requiring custom scripts
- Cross-module handoffs – the most expensive failures happen at the seams between modules
- Business Central – does the tool cover BC if it is part of your portfolio
- AI agent validation – if Copilot or agents are on your roadmap, plan that strategy separately
- Audit defensibility – traceable logs for SOX, GDPR, and compliance requirements
Summary
D365 testing in 2026 has split into two distinct jobs: keeping deterministic processes stable through an accelerating release cadence, and validating the decisions of autonomous AI agents that traditional tools were never built to check.
The first is well served by purpose-built, self-healing automation and with RSAT ending support in May 2027, now is the right moment to evaluate alternatives. The second is an emerging discipline to plan for separately, before agents are quietly making financial decisions in production.
Testing by XPLUS stands out as the only tool on the market combining full F&O and Business Central coverage, a deep Dynamics 365 focus with self-healing and an AI Optimizer, codeless ownership for functional teams, and native Azure DevOps + Power BI integration. For teams facing the RSAT deadline or planning to extend coverage to Business Central – a demo against your own environment is a better starting point than a generic walkthrough.
