Preparing your SP-API Application for Peak Shopping Events 📊🛍️🚀

[ritikacd]

Every year, Amazon’s major sales bring exciting opportunities for shoppers on Amazon.com. This leads to a big rush of shoppers eagerly waiting to make a purchase. Is your Amazon Selling Partner API (SP-API) application ready to handle the peak shopping events?

Authored by: @ritikacd Solutions Architect, Selling Partner Developer Services

This tutorial provides insights and guidance to make sure your SP-API application is ready to manage high traffic during peak shopping seasons.

Inspect your Application for bottlenecks 🔍

Every peak event is a learning opportunity to evaluate your application’s functionality, robustness, and performance. Analyze bottlenecks and failure rates after each major shopping events like Prime Day or Black Friday instead of waiting until the last minute. Monitor APIs and endpoints that negatively impacted business, and address issues well before the next peak.

To proactively identify and resolve issues, ensure proper metrics and alerting for your application. Evaluate metrics against anticipated seasonal traffic and set monitoring alerts for surges. Load testing gives insights to address bottlenecks ahead of shopping events, allowing time to handle potential issues internally.

Known potential pitfalls 💡

When preparing your application for peak traffic, keep in mind the rate limits on API resource usage. Exceeding these limits can lead to:

• ❌ Job failures
• ⏳ Delays
• 💸 Inefficiencies and costs

To ensure your application isn’t wasting resources, implement these strategies to optimize your application’s workload:

STRATEGY 1: Implement a rate limiter

Implementing the rate limiter ensures consistent calling pattern over time, preventing traffic spikes and minimizing bandwidth wastage.

• 🔑 Key Points:
  • SP-API operates with usage plans and rate limits.
  • Usage limits for API operations are specific to each seller and region.
  • For developers with multiple seller authorizations, limit applies to each seller account.
• During peak season, expect high traffic for:
  • 💰 Pricing API
  • 🛒 Orders API

Why? Sellers adjust item prices in response to:
🏷️ Competitive offers
📈 High order volume

💡 Important Understanding:

• Orders API and Product Pricing v2022-05-01 API operate under a Dynamic Usage plan:

  • This effectively means that rate-limits are automatically adjusted based on the current and historical business needs of that seller.
  • But dynamic usage plans are to right-size those limits over time, meaning rates might change.
  • The default rate limits published in the API Reference documentation can be used to plan your application.

• Monitor your limits:

  • Check x-amzn-RateLimit-Limit response header for current rate limits and build a monitor; tracking remaining requests against the rate limit.
  • Refer to the Usage Plans and Rate Limits documentation and video on architecting your application using the Rate Limiting on the Selling Partner API to learn more.

STRATEGY 2: Reduce number of API requests

As a best practice, remove redundant API requests from your application, especially during peak traffic. This ensures your application only requests data essential for the business operations and meeting customer needs.

1. Build an Event-Driven Architecture using Notifications API: Event-Driven Architecture is an asynchronous design pattern where a service triggers an event to signal a change in its state. Other services listen to these events and perform actions based on certain conditions.
🚀 Key Notification Types:

• ORDER_CHANGE for real-time order updates. Ensuring that polling is only necessary for specific buyer data. Refer to this tutorial to learn how to use this notification
• ANY_OFFER_CHANGED and PRICING_HEALTH updates on the price changes or whenever seller’s offers become ineligible to be the Featured Offer (Buy Box offer) because of an uncompetitive price.
• REPORT_PROCESSING_FINISHED notifies when requested report status changes to DONE, CANCELLED, or FATAL.
• LISTINGS_ITEM_STATUS_CHANGE notifies on listing status changes, including buyable transitions and listing creation or deletion for any SKU.
• LISTINGS_ITEM_ISSUES_CHANGE provides near real-time updates on listing issues, including enforcement actions and their impact, helping you optimize workflows and retrieving issue details via Listings APIs based on impact.

2. Bulk Data Download Strategies using Reports API: Where applicable, download data in bulk using a single API request to optimize your data retrieval process.
🚀 Key Report Types:

• Order reports to drive the fulfillment process for MFN orders. You can get all the unshipped orders that become eligible for shipping in the specific date range.
• Order Tracking reports for comprehensive order tracking. The report contains all the orders regardless of the shipping status.
• Inventory Reports used to get information about seller’s product listings, you can choose different reports based on your specific use cases.
• Rapid Retail Analytics provide almost real time reporting, enabling Vendors to identify peak shopping hours and optimize campaigns by featuring high-performing, in-stock ASINs.

Tip

🚀 Best Practices - Schedule Reports ⏰

• Set up periodic auto-generation of reports
• Reports are returned when new data becomes available
  🏆 Benefit: Eliminates the need for direct API calls to request new data

3. 📦 Bulk uploads using Feeds API: To carry out bulk tasks in a single API request like creating products or updating orders information. Use Feeds API to enqueue updates that are not business-critical.

4. 🔄 Smart Workload Management: Optimize performance during high traffic periods by:

• Pause data flows that don't require real-time updates
• Reschedule to run in batches when your application experiences a lighter workload period

By implementing these strategies, you can significantly improve your application's performance, reduce unnecessary API calls, and ensure smooth operations even during peak usage times. 🏆💯

STRATEGY 3: Implement Retry techniques

When your application receives RequestLimitExceeded response, it’s a signal that your API calls are being throttled due to exceeding the rate limit. To mitigate this issue and enhance your application’s reliability, implement these proactive retry techniques:

📈 Exponential Backoff: This technique spreads out your API requests by introducing progressively longer waits between retries, improving flow control.

• 🎯 Start with a short initial wait time (e.g., 1 second)
• 2️⃣ Double the wait time for each subsequent retry
• 🔝 Set a maximum wait time to prevent excessive delays
• 🔀 Add a small random factor to prevent synchronized retries from multiple clients

🏆 Benefits:

• 🛡️ Minimizes impact on your application's workload
• 🔋 Enhances overall system reliability
• 🌊 Smooths out traffic spikes
• 🤝 Demonstrates good API citizenship

Remember, implementing smart retry logic not only helps your application recover from temporary failures but also contributes to a healthier, more efficient API ecosystem for all users. 🌟🔧

Check for large scale issues affecting Amazon API Services 🚨

You took all measures upfront so your application handles peak traffic well. And then the big day comes and things sail out smoothly. Just in case you see problems at your end, there are a couple of channels to pursue:

For issues or service updates regarding SP-API:

• Troubleshooting tips and common errors SP-API Errors FAQs
• Check the Selling Partner API - Health Dashboard
• Posts on GitHub and SP-API Announcements

In most cases, if there is no large-scale issue reported on these sites, then the issue may be isolated to your own application.

Contact Developer Support to receive assistance 📞👩‍💻🆘

If you need further assistance with the issue you are experiencing, reach out to the SP-API developer support team via the Contact Us form. Provide comprehensive issue details, including impact metrics and business relevance, to expedite support.

Let us know what do you think! 👇 React or reply down below to share your feedback. Happy shopping!

• ❤️ Like what you see?

• 👍 Excited to try it out?

• 🎉 Want to see more of this?

Keywords: SellingPartnerAPI, AmazonSPAPI, Amazon, PeakSeason, DeveloperTips