How Real-Time Data is Changing Business Optimization - Mixpanel
Blog Post

How Real-Time Data is Changing Business Optimization

Last edited: Jan 23, 2021
Suhail Doshi

This post was originally written by Jeremy Richardson on Mashable

For a while, a big limitation of online optimization tools was their lack of
real-time reporting. Google Analytics, the most popular analytics service out there,
can easily take a full day before displaying your data. This was acceptable
back when the web was static, but as websites become more and more dynamic,
the rate at which we analyze and iterate based on our collected data has
dramatically increased.

There are many industries where optimizing in real-time can have a large
impact on overall business performance. Unfortunately, not all companies are
aware of the potential value in tracking information in real-time. LetÂ’s take
a look at a few areas where real-time data processing is already making a big

Content Websites

huffpo image

One area where we are seeing real-time analytics improve content companies is
in article headlines. For most of us, an article headline is all we use to
decide whether or not to read on, so having a good one is definitely
important. The Huffington Post is ahead of the game
. They use analytics to run A/B split
on their important articles –
in real-time. The Huffington Post initially shows 2 headlines for the same
story, after 5 minutes of testing they discard the less popular one.

Aside from A/B testing article headlines, real-time analytics can play a large
role in organizing and prioritizing entire stories as they break. If
publishers know the popularity of an article based on the number of reads,
comments, tweets, or Facebook “Likes” in real-time, then they can make
informed decisions about how to optimize their pages with the best content. As
media companies continue to concentrate more of their business online, real-
time data analysis will likely play a large role in shaping how key decisions
are made in every aspect of the company.

Game Companies

Social gaming is a huge industry that can, and does, benefit from analyzing
their data in real-time. ItÂ’s a very cut-throat business: Whenever a new
strategy is successful, itÂ’s immediately copied by everyone else. Analytics
are necessary to keep an edge on the competition. Specifically, real-time
analytics are important because social games often have a short lifetime, and
the makers want to get as much value out of a single customer as possible. If
they canÂ’t iterate on a game for a whole day, and the average user only plays
for 7 days, then thatÂ’s a large opportunity that is lost due to a lack of

A large gaming company that IÂ’’ve spoken with sees well over 30 million MAUÂ’s
and they analyze and make changes to their games every hour, every day. Real-
time analytics gives them the ability to maximize their iteration and get the
most out of each game they make. ItÂ’s a large part of why theyÂ’re so

Live Streaming

In live streaming, a flash crowd is when a large group of people shows up and
then disappears in a relatively short time span. Imagine how many people tuned
in when Landon Donovan scored that epic goal to advance USA in the World
, for example. The amount
of data here would have been huge if every network packet from every user was
captured, and was likely too much information for most companies to capture,
record and analyze in any timely and useful way. The huge amount of crowd
behaviors would need to be observed in real-time.

An engineer at live streaming company Veetle told me
they often need to reserve and allocate additional resources when a crowd is
larger than they expect. For simple cases, like the season finale of Mad Men,
there are simple allocation heuristics they use that automatically make up for
a larger than expected audience. However, for extreme cases smart allocation
is a difficult problem.

With something like the Donovan goal, they would have had to have a person
observing the flux of people streaming the game, and manually decide what
resources to properly allocate – something they would only be able to do with
real-time analytics.

Instant Gratification Makes us Happier

woopra image

Having data available in real-time simply makes you happier: The ability to
set up and start tracking immediately gives you the peace of mind in knowing
that everything is working smoothly. And, if something does end up being
wrong, no time will have been wasted waiting and hoping that the data will
eventually show up in a report somewhere. The joy that real-time brings to
customers is evident: Analytics company Woopra flat
out asked their customers why real-time mattered.

The responses Woopra received focused on the niche information that real-time
data provided them. For example, one user wrote: “Just had a potential new
customer on the phone, Woopra instantly shows me the 38 pages on our website
they read before calling.” Another user found real-time data helped better
filter the content he was producing, while for others the benefit was in
identifying trends. “Real-time stats are great to understand upcoming trends.
Helps to track user interest during different hours etc,” said one user in

[Full Disclosure: Woopra is a competitor to the author’s company.]

Your Business in Real-Time

Being able to analyze your data in real-time has the potential to add a lot of
value to your business, whether you run a hugely popular website like The
Huffington Post or a small business. However, taking advantage of real-time
analytics requires commitment – you have to be ready to act on the data as
fast as you receive it. ItÂ’s possible to automate some of the work, but you
usually need a human decision maker. This can be difficult to get correct but
it can make your advantage even bigger once you succeed.

Get the latest from Mixpanel
This field is required.