The holiday shopping season is one of the most stressful periods for operators of retail and ecommerce businesses, as the seasonal surge of holiday shoppers can put massive amounts of stress and strain on even the most well-architected websites. Here’s a recent example from 2021: The Office Depot website suffered an outage during Cyber Monday that knocked the online shop offline for hours, impacting the ability of customers to place orders online.
Whether you’re preparing for “Black Friday”—the Friday immediately after the Thanksgiving Holiday in the U.S.—or prepping for “Cyber Monday” on the Monday following Thanksgiving, one of the best ways to stay ahead of any issues your ecommerce site might experience with a massive influx of traffic is to have an active monitoring solution so you find out about broken user flows before your customers do.
Synthetic Monitoring is Critical. But Does it Have to Be Hard?
If you’re unfamiliar with the phrase, synthetic monitoring is an approach to monitoring applications and websites that simulates the actions of a real user. Synthetic monitoring has real value because it mirrors the exact actions that a visitor may take when visiting your website, say browsing an online shop, adding items to a shopping cart, and then checking out. Being able to simulate that user flow—at speed and at scale—as part of a monitoring regimen can help you quickly and reliably identify website issues before a user encounters them.
That continuous reliability monitoring of online shop functionality is a must for any ecommerce website, but it doesn’t have to be difficult. Checkly makes it easy to create synthetic monitoring checks of your critical user flows—like customers adding items to a shopping cart, checking out, and collecting a payment—that can be configured to alert you immediately about any issues.
In this post, we’ll outline a checklist of items that will help you when thinking about a monitoring regimen for your ecommerce site. In addition, we also include a video tutorial later in this post that shows you exactly how to create an ecommerce site monitoring check using Playwright and Checkly.
Checkly Ecommerce Monitoring Checklist
- Start with low-hanging fruit: Create monitoring checks for key user flows, like creating a new account, adding an item to a shopping cart, and checking out. (Check out our Checkly Guide to End-to-End Monitoring for more information about monitoring.)
Our recommendation: Use Playwright Codegen to record scripts with low effort.
- Get solid alerting in place: Set up the appropriate channels so that you are alerted as soon as something goes wrong, without waiting for your users to stumble upon the underlying issue.
Our recommendation: For your vital user flows, hook into your PagerDuty/OpsGenie setup to leverage escalations.
- Leverage your site history: A good place to start is to review when your site failed in the past. What problems and issues have you historically seen? It’s good to monitor recurring cases specifically, as your customers might not forgive you if you make the same exact mistake the second time around.
Our recommendation: Set up monitoring checks for functionality that has broken as soon as your incident is over.
- Monitor your API endpoints: Don’t forget that your API endpoints might also need to be monitored. Why? Monitoring your API endpoints might be complementary to the key user flows mentioned above, or the API endpoint might be involved in delivering the functionality of the key flow. It also might deliver an additional, disconnected functionality. For example: Some merchant platforms might be offering most of their services via API.
Our recommendation: Import your existing Swagger/OpenAPI spec or curl commands to roll out API checks in seconds.
- Monitor 3rd party APIs: Monitor your 3rd party APIs to make sure you know when there's a problem happening upstream—without having to wait for confirmation from the provider.
Our recommendation: Set up monitoring checks on all the 3rd party APIs you’d like to keep tabs on.
- Enable high frequency monitoring: When you use a low granularity / frequency of your monitoring, that can translate into downtimes going undetected, and recurring issues might be harder to spot when trying to establish patterns. Read our dedicated blog post on high-frequency checks for more information.
Our recommendation: Monitor your key end-to-end flows once every 1-5 minutes. For crucial APIs we recommend running checks every minute or less.
- Mind your test hygiene: For monitoring checkouts, test account hygiene is vitally important. Make sure you keep your test account clean, even though you’ll be using it to add items to the shopping cart and simulate real orders.
Our recommendation: Use snippets and setup and teardown scripts.
- Don’t stop at production: While it’s obvious you should have an active monitoring regimen for your live website, it is also just as necessary—and will save you and your team lots of work—to implement quality gates in your deployment pipeline and test code on pre-production as well.
Our recommendation: Kick off your existing monitoring checks on demand against pre-prod via GitHub, Vercel or command line triggers.
“Black Friday is next week already and teams should be preparing for load testing, auto-scaling, checks and so much more. Using a platform like Checkly is very helpful for ensuring performance in a sustainable way and detecting tipping points as fast as possible.” —Andreas Lehr, Senior Information Technology Architect at Schwarz
Ecommerce Monitoring Example Using Playwright and Checkly
Now that we’ve reviewed a quick checklist of items for you to consider with your monitoring setup, here’s a short video that shows the creation of a sample monitoring checkout using Playwright and Checkly. We’ll show a Playwright check being created, the check being replayed, and then the check being created and run on Checkly.
More specifically, here’s an outline of the steps shown in the video. Note: If you’d like to follow along with the steps here, make sure that you have Playwright installed first.
- Use Playwright to run Playwright codegen (Codegen docs).
- Record a sample user flow on your webshop. In the video example—using the demo Vercel online store—we simulate adding an item to a shopping cart, checking out, filling out customer data, and choosing a shipping provider.
- Re-run the check with Playwright using one command.
- Schedule the user flow check using Checkly for automated synthetic monitoring.
There you have it! In this post we’ve outlined an ecommerce monitoring checklist and then walked you step-by-step through creating an online shop user flow monitoring check using both Playwright and Checkly.
Have any questions or comments about the article, or have some of your own ecommerce monitoring tips to share? Reach out to Giovanni Rago (@rag0g) and Jeff James (@jeffjames3) on Twitter and we’d be happy to continue the conversation.