Science & Tech · Big Tech & Internet · 2 hrs ago
HyperLogLog estimates unique visitors using far less memory
Counting page views is easy, but counting unique visitors is harder because one person may open a page more than once.
A precise count requires a website to keep and check a list of visitor IDs, which can grow very large across many pages and ads.
HyperLogLog estimates the count instead of tracking every ID, using much less memory.
Its central idea is compared with flipping coins and recording the longest run of tails before heads appears.
Because long runs are rare, that record can give a rough idea of how many rounds took place; the algorithm applies a similar idea to visitors.
Google Analytics uses an augmented version of HyperLogLog to estimate visitor metrics, and the method is also used in genomics and cybersecurity.
The trade-off is that the result is an estimate rather than an exact count.
HyperLogLog estimates the number of unique visitors while using far less memory than storing a full list of visitor IDs.
It gives close estimates rather than exact counts.
Google Analytics uses an augmented version of the algorithm to estimate visitor metrics.
The article says Google Analytics tracks traffic on nearly half of all websites.
HyperLogLog is also used in fields including genomics and cybersecurity.
- Who
- Websites and services use HyperLogLog to estimate unique visitors.
- What
- The algorithm estimates unique counts using a fraction of the memory required by full lists of IDs.
- When
- The article was published on November 9, 2026.
- Where
- It is used in Google Analytics and in fields including genomics and cybersecurity.
- Why
- It helps estimate how many distinct people saw content without storing a full roster of IDs.
This story does not have two clearly opposing sides.
This story does not have a timeline yet.
- Publication date
- 2026-11-09
- Google Analytics reach
- Nearly half of all websites
- Coin-flip example
- 10 tails in a row occurs in one out of every 1,024 rounds
- Memory use
- A fraction of the memory required by full lists of IDs
- Other fields
- Genomics and cybersecurity
What people are saying
It settles for close estimates instead of exact counts, and in exchange, it uses only a fraction of the memory required by full lists of IDs.
Google Analytics, which tracks traffic on nearly half of all websites, relies on an augmented version of it to estimate visitor metrics.
A streak of 10 tails turns up only about once in every 1,000 rounds.