On-demand Webinar

Mastering Performance Testing: Best Practices

45 min Performance 2 speakers
Webinar
SJSandra John
Performance Engineering Lead, ZS Associates
DBDeep Barot
Founder, ContextQA (host)

Performance testing is really about user satisfaction — a slow app sends users straight to a competitor. This session explains how to embed performance work across the whole lifecycle, the major test types, the metrics that matter, tooling, baselines, and the shift-left future of performance engineering.

What you'll learn

Walk away knowing how to apply it

Why performance bugs are expensive and how shifting left prevents them
Set a meaningful baseline grounded in scalability and future usage
Tell load, volume, stress, spike, and endurance testing apart
Track the KPIs that matter: response time, throughput, CPU/memory
Choose performance tools and know where to start
Inside this session

What the conversation covers

Defining performance as user satisfaction (the payment-timeout example)

Embedding performance from requirements and design, not just at the end

Setting baselines around scalability and 10–20x future users

Who sets the baseline: people with visibility into future usage

Test types: load, volume, stress, spike, and endurance — and when to use each

Key metrics and an internal per-API response-time threshold

Tooling and record-modify-reuse scripting with realistic think-time

Key takeaways

The ideas worth remembering

Performance is about user satisfaction — slow apps lose users in real time

Embed performance at every stage; the dedicated test should be a validation step

Baselines come from future-usage visibility, not a fixed day-one user count

Different goals need different (and combined) test types

By the time you reach performance testing, it should just be a validation activity — a lot needs to happen before.
— Sandra John
Speakers

Who you'll hear from

SJ

Sandra John

Performance Engineering Lead, ZS Associates

DB

Deep Barot

Founder, ContextQA (host)

See ContextQA in action

Go from watching to doing — spin up an AI agent and watch it test, self-heal, and report for you.