According to the Web Analytics Association (2008), Web
analytics can be defined as, “…the measurement, collection, analysis and
reporting of Internet data for the purposes of understanding and optimizing Web
usage.” Web analytics provide specific insights that are beneficial for a
company such as, how many consumers visit a website, what do they do when they
get there, and are they finding what they need while on the website. The
information provided by web analytics is particularly important for e-commerce
sites. According to Kissmetrics (2016), understanding trends and audience behaviors
allows businesses to see what marketing channels are no longer worth the time
and money and will also help businesses decide how to shift their strategy
before any permanent damage is done to their bottom line.
There are many benefits to analyzing web metrics for
e-commerce sites no matter the size of the business. One could argue that
smaller businesses should invest more in analyzing web analytics as resources
may be more limiting than they would be for larger businesses. However, I want
to talk about one of the largest e-commerce sites – Amazon.
Amazon
Amazon, an international e-commerce company offering online retail,
computing services, consumer electronics, digital content as well as other
local services such as daily deals and groceries, is the leading e-retailer in
the United States (Statista, 2016). In 2015, Amazon had more than 107 billion
dollars in net sales with the majority of revenue generated through the sale of
electronics and other products, followed by media and other activities
(Statista, 2016).
According to Smart Data Collective (2014), Amazon primarily
used the user-to-user collaborative filtering algorithm to provide customers
with recommendations based off of other customers’ purchasing behaviors, which
Amazon still uses today, to an extent. To put it simply, user-to-user filtering
means that Amazon would find users with similar purchasing behaviors and then suggest
items that they have bought but the current consumer has not. This type of
algorithm includes the, “Customers who bought this item, also bought this one…”
section at the bottom of the item page. Although this sort of algorithm was
useful when Amazon was just a book retailer, Amazon has since grown
exponentially in size since 1995 when the e-commerce site was founded.
According to the same article by Smart Data Collective,
Amazon currently uses an item-to-item collaborative filtering algorithm, which takes
into account a users’ previous purchases, what they have put into their shopping
cart or wish list, items they have previously reviewed or rated, as well as
what other similar users have bought (2014). All of this information allows
Amazon create a highly customized experience for consumers, which helps keep
the customers coming back to Amazon for future purchases.
How Amazon Uses Web Analytics
Although Amazon has done a fantastic job at creating a
customized experience for consumers, the company has also done a fantastic job
in terms of search engine optimization (SEO). Although most people may
immediately think of Google when discussing SEO, many forget that Amazon also
is a search engine and is actually quite powerful. Unlink Google, the majority of
consumers that visit Amazon are looking to buy something, not just find
information.
According to Hubspot (2015), 44% of people go directly to
Amazon to start their product searches, compared to 34% who use search engines
like Google, Bing, and Yahoo to search for products. According to SEMrush
(2015), Amazon’s page rankings and SEO tactics are very similar to Google’s,
including headers, text ration, keywords, etc.; however, there are some
important differences since there is less control over a product page than
there is for a website or blog. SEMrush (2015) suggests utilizing the page
title, subtitle, product description, editorial review, product picture file
name, and Amazon’s “special” keywords to boost the page ranking of a product.
The special keywords are for Amazon’s internal use to know where to display
products; Amazon will ask for these keywords when a product is uploaded
(SEMrush, 2015).
Conclusion
It’s not a surprise that Amazon is a powerful e-commerce
website. Consumers are already in the mindset to buy something when visiting
the website and Amazon has over 200 million customer accounts (DZone, 2016), 80
million of which are Amazon Prime members. Amazon is utilizing big data on the
200 million customer accounts by hosting their one billion gigabyte of data on
more than 1,400,000 servers to increase sales through predictive analytics, according
to Ann of DZone (2016). Amazon has created what seems like an unstoppable
e-commerce site, but there are a few concerns that may bother some consumers. Unlike
Google and other sites, Amazon is not as transparent with their data collection
and algorithms. Amazon also seems more involved with my search history and
online behavior than other sites that I visit, at least in my personal opinion and
experience with Amazon.
Today I browsed Amazon for a new desk, later today I
found Amazon ads in the middle of my Facebook feed for the exact desks I had
browsed through earlier in the day. Although this can be helpful, there is a
point where it does become a little too intrusive. However, as a marketer, I do
understand that Amazon is hoping to get me back to their website to complete
the sale, but the advertisement almost lead me to a physical store because I
thought it was a little too much targeting. Then again, I have since looked on
Amazon at those desks again, just not through the Facebook advertisement. Maybe
the data segmenting is working on me and I don’t even know it. However,
according to Ad Exchanger, people are now more aware than ever that advertisers
are targeting them and they’re openly talking about it (2014), so when does
target marketing become too intrusive or even unethical? Amazon should be
cautious of how much they target their consumers or it may end up having a
reverse effect like it almost did on
me.
References
Ann. (2016). How Amazon uses its own
cloud to process vast, multidimensional datasets. DZone. Retrieved from https://dzone.com/articles/big-data-analytics-delivering-business-value-at-am
Butlion, J. (2015). An introduction
to analytics for ecommerce websites. Kissmetrics.
Retrieved from https://blog.kissmetrics.com/intro-to-ecommerce-analytics/
Chesson, D. (2015). Amazon SEO:
Because Amazon is a search engine too. SEMrush.
Retrieved from https://www.semrush.com/blog/amazon-seo-tactics-because-amazon-is-a-search-engine-too/
Hubspot. (2015). The ultimate list of
marketing statistics. Hubspot.
Retrieved from http://www.hubspot.com/marketing-statistics
Marr, B. (2014). Amazon: Using big
data analytics to read your mind. Smart
Data Collective. Retrieved from http://www.smartdatacollective.com/bernardmarr/182796/amazon-using-big-data-analytics-read-your-mind
Statista (2016). Statistics and facts
about Amazon. Statista. Retrieved
from https://www.statista.com/topics/846/amazon/
Swanson, S. (2014). Is there such a
thing as too much targeting? Ad
Exchanger. Retrieved from https://adexchanger.com/data-driven-thinking/is-there-such-a-thing-as-too-much-targeting/
Web Analytics Association. (2008).
Web analytics definitions. Retrieved from http://www.digitalanalyticsassociation.org/Files/PDF_standards/WebAnalyticsDefinitions.pdf
![About [span]me[/span]](https://pbs.twimg.com/profile_images/789160793518387200/4ckig3yW.jpg)
