I'm looking for:
AI & Automation Services

Automate workflows, integrate systems, and unlock AI-driven efficiency.

Recently viewed
All ServicesBook a CallContact

Performance Test Engineering London

Apache JMeter, k6, Gatling and Locust simulate realistic user loads, measure p95 latency and Apdex scores, and identify bottlenecks before production release. London software teams reducing outage risk, proving SLA compliance and validating scalability ahead of high-traffic events gain the most value.

Fixed-price projects start from £6,000. ISTQB-aligned load, stress and soak test design combined with AWS CloudWatch and Grafana monitoring surfaces capacity constraints early.

  • Fixed-price projects from £6,000
  • Load, stress, spike and soak testing
  • Prioritised remediation in the findings report
Call to Discuss Your Project
07442 569900
WhatsApp Us
+44 7442 569900

Benefits and fit

What should performance testing prove before you release?

Engineering leads, release managers and platform architects at London software businesses gain most when planned releases, high-traffic events or SLA obligations require documented performance evidence.

Bottlenecks found before launch

JMeter and Grafana load runs identify database query bottlenecks, connection pool limits and API timeout thresholds before production traffic exposes them to real users.

Documented SLA evidence

ISTQB-aligned performance test reports with p95 latency, Apdex scores and throughput data provide documented SLA compliance evidence for clients, auditors and stakeholders.

Automated performance gates

k6 tests wired into Jenkins or GitHub Actions pipelines block releases that miss latency or error rate thresholds, preventing performance regressions from reaching staging.

Scalability validated before events

Gatling spike tests and AWS CloudWatch auto-scaling validation confirm that your platform handles planned traffic surges without degradation or error rate increases.

Before you commit

Is performance testing the right starting point?

A good fit

A planned release, a high-traffic event or an SLA obligation needs documented performance evidence, and you want bottlenecks identified and attributed before promotion to production.

A smaller change may be enough

If you need direction before tooling, a performance testing strategy and SLA framework directs investment at the highest-risk areas of your platform.

What we deliver

What does a performance testing engagement deliver?

Five areas of load, stress and scalability work. Expand an area to see what your team receives.

JMeter load and stress testing

Engineering teams get Apache JMeter test plans that simulate realistic user volumes, ramp profiles and transaction mixes for web, API and microservice layers.

Distributed JMeter execution scales to high concurrent user counts.

Results integrate with Grafana and New Relic for real-time monitoring and post-test bottleneck analysis.

k6 performance tests in CI/CD pipelines

DevOps and release teams get k6 performance test scripts integrated into Jenkins, GitHub Actions or CircleCI pipelines with pass/fail thresholds based on p95 latency, error rate and throughput.

Automated performance gates block releases that degrade application speed before staging promotion, building a performance regression baseline across every release.

Gatling spike and scalability testing

Platform teams get Gatling Scala simulation scripts that model sudden traffic bursts, concurrent session spikes and auto-scaling trigger points.

AWS CloudWatch metrics and Apdex scoring confirm that cloud infrastructure responds correctly to planned high-traffic events before launch day.

Soak and endurance testing

Operations and platform teams get soak test runs of four to twelve hours that reveal memory leaks, connection pool exhaustion and thread contention under sustained concurrent load.

Datadog and New Relic dashboards track resource trends throughout the soak window, and findings reports prioritise remediation before production deployment.

Performance findings reports and remediation guidance

Engineering and product teams get ISTQB-aligned performance test reports naming each bottleneck, its root cause and a prioritised remediation recommendation.

Grafana chart exports, p95 latency distributions and Apdex trend data give stakeholders the evidence needed for capacity planning, infrastructure spend decisions and SLA negotiations.

Test types

Which type of test answers your question?

Softomate designs load, stress, soak and spike tests for each engagement based on your traffic profile, release risk and SLA requirements. ISTQB performance testing standards govern scenario design across all test types.

Load and stress testing

Load testing measures application behaviour at expected peak user volumes to confirm SLA targets are met under normal conditions. Stress testing pushes the application beyond design capacity to identify the failure point and recovery behaviour.

Soak testing

Soak testing runs sustained load for extended periods, typically four to twelve hours, to surface memory leaks, connection pool exhaustion and degradation over time.

Spike testing

Spike testing simulates sudden traffic bursts to check auto-scaling and queue behaviour.

Our process

How is a performance test delivered?

Softomate maps traffic profiles, designs ISTQB-aligned test scenarios, builds JMeter or k6 test plans and executes runs with real-time monitoring in short delivery phases. Engineering leads, platform architects, DevOps contacts and product managers stay involved from discovery through results review.

  1. Discover

    Map traffic profiles, SLA targets and risk areas

    Colleagues mapping a workflow with notes on a wall
    Library photography, not client work.

    Traffic profiles, SLA targets, infrastructure topology and performance risk areas are mapped in discovery sessions with engineering leads, platform architects and product managers.

    What you receiveA load model brief, monitoring tooling inventory and environment access plan for JMeter, k6 or Gatling test execution before scope is approved.

    How we do this

    Softomate connects test execution to AWS CloudWatch, New Relic, Datadog and Grafana monitoring so bottlenecks are identified and attributed before promotion to production.

  2. Plan

    Agree scenarios, thresholds and ramp profiles

    Reviewing charts and project information on a tablet
    Library photography, not client work.

    Test scenario types, acceptance thresholds, ramp profiles and monitoring coverage are agreed with platform owners and release managers during planning.

    What you receiveA performance test strategy, Apdex target definitions and environment preparation checklist before scripting and execution begin.

    How we do this

    Throughput targets, Apdex thresholds and error rate budgets are defined as acceptance criteria before scripting begins.

  3. Design

    Design scripts, load models and dashboards

    Working with source code at a desktop computer
    Library photography, not client work.

    Test scripts, load models and monitoring dashboards are designed with platform engineers and DevOps contacts.

    What you receiveApproved JMeter or k6 script structures, Grafana dashboard configuration, New Relic alert rules and AWS CloudWatch metric selection before execution begins.

    How we do this

    Tool selection depends on your technology stack, CI/CD environment and existing observability infrastructure. A short technical review confirms the right combination before test design and scripting begins.

  4. Build and Integrate

    Build scripts and wire in the pipeline

    Colleagues reviewing software on two monitors
    Library photography, not client work.

    Working test scripts, distributed execution configuration and CI/CD pipeline integrations are built in short sprints with engineering contacts and environment owners.

    What you receiveValidated JMeter plans, k6 pipeline scripts, Grafana dashboard exports and Datadog or New Relic alert configurations.

    How we do this

    Docker containers isolate test execution so pipeline performance stays predictable.

  5. Launch and Optimise

    Run the tests and report the findings

    A team attending a presentation in a meeting room
    Library photography, not client work.

    Live test execution, real-time monitoring review and findings reporting happen with engineering and product stakeholders after environment sign-off.

    What you receiveA full performance findings report, Grafana chart exports, prioritised remediation recommendations and a re-test plan for confirmed bottlenecks.

    How we do this

    Teams typically act on two to four high-priority findings per engagement before re-running confirmation tests.

Relevant results

Results from similar performance projects

Case study outcomes from Softomate performance test engineering engagements. Client names anonymised per NDA.

Case study · UK financial services firm · 50-million-row transaction database

How we cut a 4-second finance query to 80ms on a 50-million-row database

A UK financial services firm's internal portal had a critical client-position query that took 4.2 seconds against a 50M-row transaction database, blocking adviser workflow.

Read the case study

4.2 seconds to 80msclient-position query against a 50M-row transaction database

6 advisersinternal portal users

JMeter · London fintech lender

Database bottleneck found before PSD2 compliance launch

A London fintech lender found a database connection pool bottleneck causing p95 latency of 8.2 seconds under 500 concurrent users during JMeter load testing, four weeks before their PSD2-compliant payment API launched.

Connection pool sizing was corrected, reducing p95 latency to 340 milliseconds. New Relic monitoring and ISTQB test reports documented the improvement for FCA audit purposes.

8.2 seconds to 340msp95 latency under 500 concurrent users

k6 and GitHub Actions · UK SaaS platform · thirty engineers

Pipeline gates reduced performance incidents by 62 per cent

A UK SaaS platform with thirty engineers reduced performance-related production incidents by sixty-two per cent within twelve weeks after k6 pipeline gates launched in GitHub Actions.

Automated p95 latency and error rate thresholds blocked four high-risk releases from reaching staging. Grafana dashboards gave engineering leads real-time visibility across all builds in the release cycle.

62 per centfewer performance-related production incidents within twelve weeks

4 releaseshigh-risk releases blocked from staging

Gatling · UK e-commerce retailer

Black Friday traffic surge handled without degradation

A UK e-commerce retailer validated Black Friday spike capacity using Gatling simulations modelling five times peak daily traffic. AWS auto-scaling configuration was adjusted after Apdex scores fell below threshold at three times normal load.

The live event processed the highest transaction volume in the company's history with zero degradation incidents and an Apdex score of 0.96.

Apdex 0.96during the live event

Zerodegradation incidents

Locust soak test · NHS-contracted healthcare platform

Memory leak resolved before NHS contract renewal

An NHS-contracted healthcare platform discovered a memory leak causing 34 per cent memory growth per hour during an eight-hour Locust soak test. The leak was attributed to a third-party session library and patched before contract renewal review.

Datadog dashboards documented stable resource consumption across a subsequent ten-hour soak test, providing evidence for the contract renewal submission.

34 per centmemory growth per hour found in an eight-hour soak test

Ten hoursstable follow-up soak test

Pricing and running costs

How much does performance testing cost?

Fixed-price projects start from £6,000. Softomate provides fixed project pricing after a structured discovery call. We run a capacity planning session before quoting; the fixed price follows the session, not a rough market average.

Fixed-price projectsFrom £6,000

Fixed-price performance test engineering projects start from £6,000.

Before quotingCapacity planning session

Scope depends on your target user concurrency, peak traffic patterns and SLA obligations. We run a capacity planning session before quoting; fixed price follows the session, not a rough market average.

What shapes the cost

What affects the price
Scope, number of user journeys, environment complexity and monitoring integration all affect the final cost. Distributed JMeter or Gatling test plans covering multiple services cost more because scripting and data preparation take longer.
Pipeline gates
CI/CD pipeline integration adds scope for teams wanting automated performance gates on every release.

Softomate provides fixed project pricing after a structured discovery call.

Discuss your requirements →

Why Softomate and aftercare

Why choose Softomate for performance testing?

Tests built around your SLA targets, findings you can act on and coverage beyond the application layer.

Tests built around your SLA targets

We build load test scenarios against your specific SLA targets (response time, error rate, throughput) rather than arbitrary benchmarks.

k6 is our default for modern microservices; JMeter for legacy Java stacks; Gatling for Scala environments. All three available.

Softomate selects from JMeter, k6, Gatling and Locust based on your technology stack, CI/CD environment and the scale of concurrent load your platform must handle.

Findings you can act on

A performance report that says "the API failed at 500 concurrent users" is not useful without the fix. Softomate performance engagements include profiling (New Relic, Datadog or AWS X-Ray depending on your stack) that traces the exact bottleneck to a code path or infrastructure constraint.

Findings reports name each bottleneck, its root cause and a prioritised fix recommendation, so engineering teams act on the highest-impact issues first after test runs.

AWS CloudWatch, New Relic, Datadog and Grafana give infrastructure, application and service-level visibility during every test run, not just load generator output.

Infrastructure and security coverage

Application code is rarely the only bottleneck. Our performance engineers review your AWS, Azure and GCP cloud infrastructure configuration (auto-scaling policies, RDS connection pools, CDN caching rules, load balancer timeout settings) alongside the application layer in every engagement.

Dynamic Application Security Testing (DAST) is run alongside load tests because production-scale traffic often exposes security weaknesses not visible at low volume.

OWASP ZAP scans are included in every Softomate performance engagement as standard, at no additional cost.

FAQs

Questions before you commission performance testing

Further guidance

Explore the detail behind your performance project

More on benefits, technical features, scoping, sector examples and client feedback. Open the areas relevant to your team.

More benefits and technical features

Memory leak detection via soak testing

Extended soak test runs reveal memory leaks, connection pool exhaustion and thread contention that only surface under sustained load, protecting long-running production deployments.

Real-time Grafana monitoring

Grafana dashboards displaying Apdex scores, p95 latency and error rates during test execution give engineering leads immediate visibility into application health under load.

Distributed JMeter execution

Distributed Apache JMeter test runs scale to thousands of concurrent virtual users across multiple load generator nodes without test environment constraints.

Grafana real-time dashboards

Grafana dashboards display p95 latency, Apdex scores, throughput and error rates in real time during every load test run for immediate engineering visibility.

AWS CloudWatch infrastructure monitoring

AWS CloudWatch metrics track CPU, memory, database connections and auto-scaling events in parallel with load generator output during every performance test.

k6 CI/CD performance gates

k6 threshold rules enforce p95 latency and error rate budgets on every pipeline run, blocking deployments that introduce performance regressions automatically.

Apdex score reporting

Apdex scores quantify user satisfaction under load by classifying responses as satisfied, tolerating or frustrated, giving stakeholders a single readability metric.

ISTQB-aligned test design

Risk-based load model design, scenario coverage planning and results interpretation follow ISTQB performance testing standards across every engagement.

How we scope and report

Fixed price after a capacity planning and SLA requirements session

Performance testing scope depends on your target user concurrency, peak traffic patterns and SLA obligations. We run a capacity planning session before quoting. Fixed price follows the session, not a rough market average.

ISTQB-aligned test design

Risk-based scenario selection, Apdex threshold definition and exit criteria align every performance test engagement with ISTQB standards before execution begins.

Outage prevention and cost savings

Identifying bottlenecks before production avoids outage costs and prevents over-provisioning of cloud infrastructure.

CI/CD performance regression gates

k6 threshold-based pipeline gates prevent performance regressions from reaching staging, building a release-by-release performance baseline that teams can reference over time.

Performance testing use cases across UK software sectors

Performance test engineering deployments use JMeter, k6, Gatling and Locust to validate application speed, scalability and resilience before high-traffic events, product launches and SLA renewal cycles. The approach suits platform engineers, DevOps leads and release managers across London financial services, e-commerce, SaaS and regulated software markets. Softomate clients typically identify two to four high-priority bottlenecks per engagement that would otherwise have caused production incidents.

JMeter load testing for fintech payment platforms

JMeter distributed load tests simulate peak transaction volumes on payment APIs, authentication flows and balance enquiry endpoints. p95 latency and error rate results provide FCA-auditable SLA compliance evidence. Softomate fintech clients typically identify three to five database connection bottlenecks per engagement before production release.

k6 pipeline gates for SaaS release management

k6 performance tests integrated into GitHub Actions pipelines block SaaS releases where API response times exceed agreed p95 thresholds. Automated performance baselines build across every release branch. Softomate SaaS clients typically reduce performance-related production incidents by sixty per cent within three months of pipeline gate deployment.

Gatling spike tests for e-commerce traffic events

Gatling spike simulations model Black Friday, seasonal sale and marketing campaign traffic surges to confirm AWS auto-scaling triggers at the right thresholds. Apdex scores and CloudWatch metrics validate infrastructure capacity before the event window. Softomate e-commerce clients avoid basket abandonment caused by performance degradation during high-demand periods.

Soak testing for healthcare platform stability

Eight-hour soak tests on NHS-contracted and regulated healthcare platforms reveal memory leaks, session expiry failures and thread pool exhaustion under sustained concurrent load. Datadog dashboards track resource trends throughout. Softomate clients use soak test findings to prevent degradation incidents in platforms running twenty-four hours a day without scheduled downtime.

More questions about standards and monitoring
Does Softomate follow ISTQB standards for performance test engineering?

Yes. Softomate applies ISTQB performance testing principles across test scenario design, load model definition, monitoring coverage and results interpretation. Risk-based scenario selection prioritises the user journeys and API endpoints most likely to bottleneck under production load. Throughput targets, Apdex thresholds and error rate budgets are defined as acceptance criteria before scripting begins. Test completion criteria, exit conditions and defect severity classifications align with your release and acceptance process. Performance test reports include Grafana charts, p95 latency distributions and remediation priorities suitable for engineering, product and executive review.

What monitoring tools do you use during performance test runs?

Softomate uses AWS CloudWatch for cloud infrastructure metrics, New Relic and Datadog for application performance monitoring and distributed tracing, and Grafana for real-time dashboard visualisation during test runs. k6 Cloud and JMeter reports provide load generator metrics including virtual user ramp, throughput and error distribution. Apdex scores, p95 and p99 latency percentiles and response time histograms are captured for every test scenario. Monitoring configuration is agreed during discovery so coverage spans load generator, application server, database and external service layers throughout the test window.

What clients say

"From understanding requirements and offering solutions, to implementing functionality accurately and very promptly, your business is in great hands."

R. Nagpal - Google review

"They have this attitude which says 'Nothing is too much trouble.' This is what makes them so approachable."

P. Mathur - Google review

Pair performance testing with wider quality coverage

Teams needing wider quality coverage can pair performance testing with our automation test engineering services, complete testing services, testing strategy consultancy and vulnerability assessment and penetration testing services.

Vulnerability assessment and penetration testing

Combine performance validation with VAPT security testing to confirm your platform meets both SLA and UK GDPR security requirements before production launch.

Work with us

Let's talk about your performance testing

Software teams seeking load test evidence, SLA compliance and scalability validation. JMeter, k6, Gatling and ISTQB-aligned test design can identify bottlenecks before launch, prevent outages and prove capacity for high-traffic events.

Book a discovery call WhatsApp us
Deen Dayal Yadav, founder of Softomate Solutions

Deen Dayal Yadav

Online

Hi there 👋

How can I help you?

Chat on WhatsApp