Loading

Sale Event Preparation checklist / Assessment for a B2C Commerce Storefront

Udgivelsesdato: Sep 22, 2025
Beskrivelse
Environment: Production

This checklist is designed to help prepare for a successful sale event. The checklist divides items into weekly tasks beginning two weeks prior to the sale event. It includes tasks leading up to the sale and during the sale, and ends with post-sale tasks to measure success. Please also see Get Ready For Your Peak Season - Best Practices and Troubleshooting for B2C Commerce.
Løsning

For any last minute sale event, the information below is still valid. Please also check the real-time performance metrics on CCAC Reports & Dashboard to assess resource utilization per minute for indicators.

For Flash sales/hype sale/product launch sales, please refer to B2C Commerce - Hype/Flash Sales Checklist.
 

2 Weeks Before Sale Event

1) Gather all required sale event information.

Please provide as much detail as possible:
  • Maximum requests per day:
  • Maximum requests per hour: 
  • Maximum orders per day: 
  • Maximum orders per hour: 
  • Date of last similar sale, sale type, and peak-hour orders: 
  • Estimated multiplier vs. typical hour on realm (ex. "2X normal hourly orders"): 
  • Expected traffic distribution (e.g. Peak traffic of for e.g. 50% spike in visits/orders in first 10 minutes OR specific hour of the day): 
  • If a hype/product launch sale, list of specific SKUs (master, variation, etc.): 
 
The information above can be gathered with your Business Analytics team, or from B2C Commerce platform's Analytics if you are basing the expected sale event traffic off of a previous sale event, OR if you are trending based on latest traffic.
 
Below are the steps to gather this information from a previous sale event:
  • If you do not have access to Reports and Dashboards, please review the following instructions.
  • Requests over a certain time period broken down by pipeline, controller, OCAPI, or SCAPI can be found via CCAC Reports & Dashboard's Technical dashboard.
  • Order information is available via the CCAC R&D Sales dashboard.
 

2) Check if you need to open a Sale Event Preparation ticket.  

If any of the conditions below from Informing B2C Commerce Support about a Sales Event are met, please open a ticket with Salesforce B2C Commerce Support team:
  • At a minimum, one thousand (1000) orders are expected in the peak hour of the sale event..
  • Sale Event maximum for orders, requests, or visitors is 2x (200%) or more than a typical hour on the realm.
  • The event is expected to have a different type or traffic pattern than previous events.
 

3) Check if you use a stacked CDN or other third party service, like Remote Proxy, in front of the B2C Commerce Platform’s embedded CDN (eCDN).

For detailed setup instructions, please review Configure an External CDN or Third-Party Proxy.



4) Check potential bot traffic patterns.

If your storefront is impacted by malicious bot traffic, especially during sale events, please consider bot mitigation options, like configuring best practices for bots, captcha, rate limiting, or bot mitigation solutions as outlined here and here.
 
The Top Robots report in the Reports & Dashboards Traffic dashboard provides you with an overview of the number of unique bots (including only known bots), and the total number of requests from bots.

5) Check Pipeline/Controller Performance.
Please check the performance of your key pipelines/controllers that are most commonly used on the storefront via the Technical dashboard of Reports and Dashboards e.g. Home, Search, Product, etc. pipelines and any additional pipelines. Our recommendation is to keep main pipeline/controller response times under 500 ms.

Please also ensure caching is applied to as permitted by your use-case so that number of requests to application tier are reduced which helps during high traffic scenarios. If your site is constantly hovering over 500 ms for an extended period, we recommend initiating a development effort/sprint to review pipeline/controllers code to find areas that can be improved.

It may also help to do a load testing to ensure site performs under sale event equivalent traffic.

You may find this article useful in improving caching.


6) Check quota alerts.


7) Check the Service Framework.
Ensure all third party services (including third party services used by background jobs) have timeouts configured properly. Recommendations are 1-3 seconds for storefront requests, and 5-10 seconds maximum for payment providers.

Configure and use circuit breaker and rate limiting.

Identify business critical third party services, like key payment providers:
  • Make them aware of your sale (expected traffic and peak hours)
  • Review and document your third party services escalation procedures
  • For critical services that cannot be disabled, implement a business strategy in the case where the third party service becomes unavailable.
  • For non-business critical third party services, understand how they can be disabled.

 
1 Week Before Sale Event
1) Check replication schedule.
Check replication schedules if they conflict with sale event peak traffic time. If a replication is needed prior to the hype-event, make sure to schedule the (code or data) replication at least four (4) hours prior to the start of the hype-event. Please review the replication best practices.
2) Check if there are any platform code update planned. For more information, see the B2C Commerce Deployment Schedule.
Due to the sale event, if you need to postpone or opt-out of a scheduled platform deployment, please create a support request at least 72 hours prior to the sale event.
 
3) Check for any sale event promotion/e-mail campaign.
If you are planning to run an online marketing promotion, please review Campaign Management Best Practices.

If you are planning to send personalized emails to your customers in preparation for holiday promotions and sales, it's important to protect your cache hit rate to maintain your site's performance:

 

4) Check for code freeze window.
Please consider applying code freeze at least few days before Sale event start date to avoid any untested changes to disrupt sale event traffic.
 
5) Ensure there is no manual cache validation.
As noted in replication best practices, please avoid any manual cache invalidation OR content changes that could impact the site preparedness for peak traffic.
 
6) Check job schedules.
  • Devote all resources to storefront requests processing during the peak of your traffic. Remember all three instances (Development, Staging, Production) in your PIG environment share the same database sub-system
  • Ensure no heavy/intensive jobs are scheduled to run during your high traffic:
  • Disable/postpone execution of non-business critical jobs
  • Run your heavy jobs ahead of the traffic (replication, index rebuilds - use incremental rebuilds, site backups/exports, etc.)
  • Monitor your job history to identify heavy/long running jobs
  • Build search indexes on your Staging instance and use replication to update indexes on Production. On Production, use incremental indexing.
  • Do not run page cache clears during high traffic times, or set up restricted cache partitions in advance in case of a business emergency.

 
During the Sale Event
1) Check service status for any delays in 3rd party response times
You can view the live data on 3rd party response times for past 15 minutes by navigating to Administration >  Operations >  Service Status in Business Manager. This window will provide useful data like success rate, error rate and unavailability rate in terms of number of calls per second. Based on this narrow down to any specific 3rd part call experiencing issues with sale event traffic.
 
2) Review the real-time data in Reports and Dashboards as outlined in the Real-Time dashboard of Reports and Dashboards.

 
Post-Sale Event
  • Gather information like actual orders per hour/day, visitors per hour/day vis-a-vis projected. Use the same method used for gathering the data discussed in the initial section of the document.
  • Verify the accuracy and saving this data over a period of time would help trending and better projections.
 
Vidensartikelnummer

000391437

 
Indlæser
Salesforce Help | Article