Wednesday, 13 August 2025
Introduction
Global law firms depend upon complex, distributed IT environments where the performance of vital applications directly impacts the productivity of fee earners and the overall efficiency of legal service delivery. As well as productivity loss and a poor user experience, any downtime quickly becomes expensive. Core platforms include Document Management Systems (e.g. iManage), Practice Management Systems (e.g. Elite 3E, Aderant, SAP Fulcrum), Case and Workflow Management platforms (e.g. Intapp Case Management), and the Microsoft Office 365 suite; particularly Outlook. These applications form the operational backbone of modern firms. Users are everyone in the firm, which can number in the thousands. These applications are subjected to constant change – patches, new versions, operating system updates, security patches or infrastructure changes. Test automation can enable regression testing and allow BAU services to be regularly checked as changes are made, but performance is more difficult to accurately measure, without standing up expensive, integrated test environments that may not match production infrastructure.
The business-critical applications most firms use are configured with custom plug-ins, Office templates and workflow automations tailored to each firm’s processes. Outlook is a prime example, commonly extended with multiple add-ins to support time recording, email filing, document search / retrieval and saving, as well as matter enquiry. One law firm we worked with identified that some users were loading over 30 Outlook plugins, which together can impact performance, even if many are inactive most of the time. While such customisations improve productivity, they introduce performance risk—especially when workflows exist across multiple integrated applications, which are likely to be a mix of cloud, hybrid and on-premise implementations.
Increasingly, the effectiveness of a legal IT estate is not defined by individual applications in isolation, but by the strength and performance of the entire suite of integrated applications. The flow of data between systems is essential to ensuring matter visibility, compliance, time capture, and billing information. A slowdown or failure in one component often impacts multiple touchpoints in the user journey. Integrations are increasingly handled by specialist platforms, which helps standardise connections; allowing integrations to be brokered and standardised by a central platform; for example Mulesoft.
There are additional complexities and variables in Infrastructure across the sector that can affect end user experience and application performance. Larger firms tend to roll out a controlled, standardised laptop build (most commonly Windows but also Mac) across their estate, while others adopt a Bring Your Own Device (BYOD) approach, accessing applications via Virtual Desktop Infrastructure (VDI). Citrix remains widely used to centralise application delivery, adding complexity to performance diagnostics via session virtualisation. Meanwhile, an increasing number of firms have deployed Zero Trust Network Access (ZTNA) solutions—such as Zscaler, Fortinet, or Cisco Duo—to support secure, policy-based access, both remotely and from the office. These technologies, while essential for cyber resilience, can introduce latency through traffic inspection, identity enforcement, and inline security processing; all contributing to the performance overhead.
The complexities explained so far are not the end of the story; there are other aspects to consider that affect end user performance. These can include:
- WAN performance, routing policies, and regional latency driven by distance
- VPN or ZTNA enforcement, including packet inspection overheads
- Internet quality both at home and in the office
- Office infrastructure, including outdated local networking equipment
- Hybrid hosting models, where services span on-premise and cloud environments
Despite these complexities, fee earners rightly expect consistent and responsive access to applications—wherever they are working. IT and architecture teams are often left responding to performance complaints without the visibility or metrics needed to diagnose root causes or confirm system health.
This white paper discusses the difference between central performance testing and checking user experience at remote locations, arguing there is a case for both—balancing structured performance testing with targeted monitoring—aligned to business criticality, user impact, and the operational realities of the global legal sector.
A Smarter Approach to Performance Testing
At its core, performance testing is about assessing the capacity of the infrastructure and application stack to handle expected or peak workloads while maintaining acceptable response times. It is a technical evaluation of how well systems behave under pressure—not a measure of subjective user experience at a particular location. While user response times can be affected by many variables, these may be best addressed through application performance monitoring (APM) tools, rather than solely via formal performance testing tools.
Despite advances in tooling, global-scale performance testing remains challenging for several reasons:
- While enterprise tools such as LoadRunner, NeoLoad, and JMeter support distributed agents, many firms do not maintain globally accessible or mirrored non-production environments that are of a similar specification to production, with all the integrated applications live systems have.
- Local test execution often depends on available infrastructure and personnel, which can vary significantly across regions.
- Achieving realistic concurrency levels—for example simulating 500 users from London, 100 from Sydney, and 250 from New York—is logistically complex and costly to coordinate.
- Test data management, user provisioning, and scheduling across time zones all add operational overhead.
As a result, regular full-scale performance tests are rarely sustainable for all applications and geographies. But also, are they worth it? To achieve optimal results from performance testing, a large number of concurrent users / transactions will be needed, which could affect the stability of global systems, as well as interfere with production traffic on corporate networks.
We recommend a tiered approach that balances formal testing with continuous monitoring, aligned to the criticality of each application and its role in the firm’s operations.
Strategic testing should be aligned to major change events, such as the rollout of a new practice management system, a significant infrastructure migration, or the deployment of high-impact plug-ins to Outlook or DMS platforms. These events justify the coordination required to execute full-scale performance tests with meaningful load levels and distributed test agents.
Tactical testing and monitoring can then be used for business-as-usual (BAU) assurance, providing lightweight, continuous visibility into application responsiveness, error rates, and usage patterns. This can be achieved using APM tools like Dynatrace, AppDynamics and New Relic) or end-user telemetry platforms such as Systrack, often complemented by RUM data to highlight regional variances in experience.
Monitoring can turn subjective complaints into quantifiable trends—highlighting degradation over time, flagging location-specific slowdowns, and enabling proactive incident management. Dashboards can be built to track key metrics for core systems across regions, making it easier to prioritise remediation and justify investment.
Ultimately, a smarter performance assurance model should:
- Align testing effort with risk, business impact, and user reach
- Reserve formal testing for high-value, high-change scenarios
- Introduce monitoring to provide continuous insight across the broader application landscape
This hybrid approach ensures that performance assurance becomes an ongoing capability, not a one-off activity—and can support the consistent delivery of quality application services to users across the firm.
A Combined Testing Model
To measure and regularly check application performance in a complex, distributed legal environment, what often works best is a combination of both approaches - structured and formalised performance testing on applications when warranted (defined by the organisational Test Strategy) and regular checks of user experience at global locations. These can be supplemented by regular questionnaires to key users at global locations to survey them on performance, especially where issues have been reported in the past. There is no one size fits all approach; a tailored approach based on the criticality of applications and how they map to business processes and key users (especially fee earners) is usually the way to go.
We generally recommend all organisations that do testing operate under the guide wires of an agreed and executive level approved Test Strategy. Shameless plug here; if your firm doesn’t have a Test Strategy, we can help! The Test Strategy is a topic wider than this white paper, but briefly, the performance element would usually categorise applications by business criticality and include a matrix of possible changes to each category, as explained further below.
A Risk-Based Testing Framework
Not all systems have the same level of risk and hence demand the same level of scrutiny. The strategy should categorise applications based on:
- Business criticality – How essential is the application to fee earners or client delivery?
- User volume and distribution – How many users access it concurrently, and from how many regions?
- Fee earner dependency – Does poor performance directly impact revenue-generating activity?
- Change frequency and technical complexity – How often is the system updated, and how integrated is it with other applications?
We have seen applications categorised in a range of different conventions, typically including Bronze, Silver, Gold but also a level above - Platinum or Crown Jewel applications.
Using these criteria, firms can determine:
- Which applications should undergo regular formal performance testing (e.g. Outlook, Document Management Systems, Practice Management System, Time Recording / Billing)
- Which should be monitored in production using lightweight, continuous telemetry tools
- Which can be assessed on an ad hoc basis when change occurs or issues are reported
This tiered approach ensures effort is focused where the risk and impact are highest, and avoids overburdening technical teams with unnecessary test cycles.
Tooling Strategy
Performance assurance should incorporate two complementary methods:
Centralised Performance Testing
Full-scale, pre-production testing using globally distributed injectors or agents. Suitable for major rollouts, infrastructure changes, or complex application upgrades.
Tools we have had success with are:
- LoadRunner – enterprise-grade protocol support and scalability for desktop and Citrix
- JMeter – cost-effective, flexible open source tool for browser based and API driven applications
Application Performance Monitoring (APM)
Production-based insight into system behaviour under real user load. Enables early detection of performance degradation and regional anomalies.
Recommended tools include:
- Dynatrace, AppDynamics, New Relic – transaction tracing, root cause analytics
- Lakeside Systrack – endpoint and user-level telemetry
- Real User Monitoring (RUM) tools – capture actual end-user response times by location
APM tools, once installed and setup, naturally fall within the remit of the Operations team within an organisation - so that real time insights can be obtained, together with regular automated reports on user experience.
Together, these approaches allow firms to balance deep, event-driven testing with lightweight, ongoing observability—tailored to the needs of each application and user group.
Application Performance Monitoring
Less well understood than performance testing, this section aims to provide a little more context on how firms might use APM to keep track of the experience of their Global userbase. In this section we will explore two approaches to APM - Synthetic Monitoring and Real User Monitoring (RUM).
Synthetic Monitoring
Much like performance testing, Synthetic Monitoring involves running scripted, simulated user journeys from strategic locations on a scheduled basis. These synthetic transactions imitate key workflows such as using Outlook to retrieve a document from iManage and sending an email, opening a matter in Intapp, or filing a document to iManage - and allow firms to measure how applications perform in a controlled and repeatable way, independent of real user activity. This is different from traditional performance testing; these processes are run in the production environment and not at high volumes of users. Although they can be synchronised and run together from various locations across a firm’s estate - it’s obviously not advisable to be running large volumes of simulated traffic across production infrastructure on which the firm depends. However, these tests can be very useful in establishing performance baselines, comparing performance across a range of geographic locations and generally acting as an early warning system - often detecting degradation before users experience or report it. The results are automatically collated into a central portal with highly configurable dashboards and automated alarms / alerts for instances where performance is identified as being below pre-set thresholds.
Key benefits:
- Proactive and predictable – issues can be identified before they affect users
- Easy to target specific locations or endpoints (e.g. regional offices, VPN exit points)
- Scripted workflows mean consistency across test runs
Considerations:
- Requires ongoing maintenance of scripts as applications evolve
- Simulated traffic does not account for variations in real user behaviour or device types
- Not suitable for uncovering usage-specific problems (e.g. user paths, plug-in conflicts)
Real User Monitoring (RUM)
RUM tools passively capture actual user interactions with web- or desktop-based applications, collecting telemetry such as page load times, API response delays, JavaScript errors, device types, and network conditions. This gives an unfiltered view of how applications perform in real-world conditions—across all regions, configurations, and user profiles.
RUM is particularly valuable for identifying issues specific to user segments, devices, or office locations and troubleshooting performance complaints that are difficult to reproduce. As there are no scripts needed in the RUM approach, it is easier to deploy and enables Testing and Operations teams to quickly understand a broader cross section of usage across different locations, platforms, operating systems and browsers.
Key benefits:
- Grounded in real user behaviour – no assumptions or scripts required
- Helps correlate user experience issues with environment variables (e.g. region, bandwidth)
- Offers long-term performance trend visibility
Considerations:
- Reactive rather than proactive – only highlights issues once they are experienced
- High data volumes may require filtering and tuning to generate actionable insights
- Less control over which transactions are measured and potentially more output to sift through, although this can be configured.
It is also important to note that many APM platforms offer both synthetic and RUM capabilities in a single suite. Examples include Dynatrace, AppDynamics, New Relic and DataDog.
Conclusion
Application performance testing is a vital component of testing and assurance for global law firms, particularly where fee earners rely on a core set of tightly integrated business applications, across Global locations and there are a wide range of factors to take into consideration which can affect end user experience.
Traditional approaches to performance testing, while still valuable, are often too resource-intensive to apply consistently across all systems and locations on a regular, BAU basis. A smarter, risk-based strategy combines targeted, formal performance tests during high-impact change events with continuous monitoring of production systems using tools that offer synthetic and real user visibility.
By adopting a hybrid performance testing and monitoring model, firms can focus testing effort where it delivers the greatest value, while relying on tools to monitor and report on performance in real time, allowing them to identify and resolve issues before they affect end users.
Prolifics Testing have been working closely with some of the World’s largest Global Law firms for over 10 years. Our experienced team can provide structured advice or targeted performance testing services, both tactically on projects and more strategically across organisations.
If you’d like to discuss any of the ideas in this white paper, or explore how we could support your firm, please get in touch for an initial, no obligation consultation.
Jonathan Binks - Head of Delivery
Prolifics Testing UK
