Muhammad Ikram Let's talk

Black Friday Speed Is Built Weeks Before the Traffic

On sale day, thousands of shoppers ask for the same product pages. A cache answers them from memory instead of the database.

That is caching in one line: fetch once, serve the copy many times. A hit is fast. A miss goes to the database and stores a copy.

The risk is stale data. Give every copy an expiry, clear it when a price changes, and read stock live at payment.

Redis is the common tool for this, an in-memory data store. Redis 8 ships under RSALv2, SSPLv1 or AGPLv3.

Who owns cache rules on your team?

First comment (optional): Redis docs on eviction and cache hit rate: https://redis.io/docs/latest/develop/reference/eviction/

01 / 08Caching

Caching, for business leaders

Stay fast on sale day without a bigger database

What a cache stores, what a hit and a miss mean, and why stale prices are the real risk.

Shopper>Cache>Database

02 / 08Caching

What a cache actually is

A cache keeps a copy of data your site reads often in fast memory. The app checks the copy first, so it does not ask the slower database every time.

Fetch once. Serve the copy many times.

03 / 08Caching

Hit or miss: one request

1

App checks the cache

Is a copy of this product page there?

2

Hit: answer from memory

The database never hears about it.

3

Miss: go to the database

Fetch it, store a copy with an expiry, then answer.

Every hit is a question your database never has to answer.

04 / 08Caching

What to cache, and what not to

DataCache it?How
Product pagesYesExpire in minutes
Search resultsYesShort expiry
SessionsYesExpire when idle
Sale pricesCarefulClear on every change
Stock at paymentNoAlways read live

05 / 08Caching

Redis, in one slide

An in-memory data store that redis.io says millions of developers use as a cache, database, streaming engine and message broker.

2009open source project since
3Redis 8 licence optionsRSALv2, SSPLv1 or AGPLv3

06 / 08Caching

Why it matters at peak

No cache

  • Every visit hits the database
  • Pages slow as traffic climbs
  • Scaling means a bigger database

With a cache

  • Repeat reads come from memory
  • The database handles misses and orders
  • Spikes land on the cache first

07 / 08Caching

Control the stale-data risk

+

Give every copy an expiry

Redis deletes the key when its time runs out
+

Clear the copy when a price changes

Do not wait for the expiry on sale day
+

Read stock live at payment

A cached "in stock" can oversell
+

Track the hit rate

Hits / (hits + misses), from Redis INFO

08 / 08Caching

Your turn

Who decides when your cached prices expire on sale day?

Comment

Tell us how you handle it

Share

With your tech or ecommerce lead

Save

For your Black Friday prep

Sources: redis.io docs (Get started, Key eviction, EXPIRE) · redis.io/legal/licenses · redis.io blog, 1 May 2025. Checked Sep 2026.

More insights

Keep reading

All posts
Carousel

E-commerce & Ops 2 min read

Your Search Box Is the Highest-Intent Funnel You Own

A shopper who types a query already knows what they want. Zero results, a missed typo or a slow page sends that intent elsewhere. Deloitte linked a 0.1s faster mobile site to 8.4% more retail conversions. Audit the search log first.

Carousel

E-commerce & Ops 2 min read

Choose a PIM by Who Runs It, Not by the Demo

A common trap: picking a PIM from the slickest demo. The costlier question is who runs it. Self-hosting means the team owns upgrades; SaaS means the vendor roadmap does. Get that wrong and every channel launch slows down.