E-Commerce Analytics

Prof Anindita Paul

Information Systems
IIM Kozhikode
Learning Outcomes
• Metrics for measuring Online Marketing
• Web Analytics and key metrics to track the
success of websites
• Hands on Web Analytics
Web Analytics
 Web analytics is the method of monitoring how users interact with your
online presence. It looks at their behaviours and their activities and provides
you reports which help you to understand how people consumed your
content.
 Most web analytics platforms require you to embed a tracking code
 Best use of Web analytics is having questions beforehand to ask. Without
questions WA may not be as effective.
 Some questions that are suited for web analytics could include:
What keywords are people using?
How do people find the site?
Where do people go next when they’re on the site?
What are the biggest entry points?
What sort of content performs the best?
What content isn’t performing as well?
If you have email marketing in place, how many people are reading the emails?
What links in the emails are performing the best?
Are there any trends or patterns in the most popular links that you should be aware of?
What can you improve as a result of these questions?
 Web Analytics (Contd.)
 Used as KPI to understand how you are doing against your objectives.
 Metrics used are CTR, Number of views, Bounce rate etc.
 Popular Web Analytics tools include: Google analytics, Adobe Analytics,
Crazyegg, Clicktale.

Audience Data Audience Behaviour Campaign Data

▪ number of visits, number of ▪ common landing pages ▪ which campaigns drove
unique visitors ▪ common exit page the most traffic
▪ new vs. returning visitor ▪ frequently visited pages ▪ which websites referred
ratio ▪ length of time spent per the most traffic
▪ what country they are from visit ▪ which keyword searches
▪ what browser or device they ▪ number of pages per resulted in a visit
are on (desktop vs. mobile) visit ▪ campaign medium
▪ bounce rate breakdown, such as email
vs. social media
Online Marketing specific metrics
• Audience size/market share metrics
▪ Impressions, click-through rate (CTR), page views, viewability rate,
stickiness, loyalty, reach, recency
• Conversion to customer metrics
▪ Acquisition rate
(percentage of visitors who register or visit product pages indicating interest in
the product)
▪ Conversion rate (measures the percentage of visitors who actually purchase
something),
▪ Browse-to-buy ratio (the ratio of items purchased to product views),
▪ Cart conversion rate (ratio of actual orders to “Add to cart” clicks),
▪ Abandonment rate (measures the percentage of shoppers who begin a
shopping cart form but then fail to complete the form and leave the website.)
• Video ad metrics
▪ View time, completion rate, skip rate
• E-mail campaign metrics
▪ Open rate, delivery rate, click-through rate, bounce-back rate
3 Ways to Get More Out of Your Web
Analytics by Filip Matous
• What are the points being made by the Author?
3 Ways to Get More Out of Your Web
Analytics
1. Focus on metrics that really matter to 7. Break the metrics into macro (main
your business goals objective) and micro (assist)

2. Communicating those metrics in useful ▪ Macro - Lead metrics: contact page

ways used to send email; your phone listed
is called; target company visits your
3. Dig deeper into trends; System 2 site
thinking ▪ Micro Lead metrics: Downloads;
videos watched
4. Avoid aggregate data
8. Good metrics represented as ratios as
5. Figure out causation you are not looking at isolated numbers
6. Have core metrics determined (a
9. Finally communicate and use emotional
handful of those) and continue to focus
intelligence to get your team feel safe to
on those month over month, year over
learn and improve the metrics
year
Exercise – Analysing Website
1. List out Ikea’s brand message (e.g., convenience).
2. How effectively does Ikea’s website/app reflect its brand
promise? (What web/app features commensurate with 1 above)
3. What would be some questions that Ikea should ask about its
website? (mostly related to 2 above)
4. What will be some commensurate Web Analytics
features/metrics that Ikea should examine? (that should also
answer the questions asked in 3 above)
5. What can you say about UX and Business goals (relation
between these) from the exercise above?
Web Analytics and the Online
Purchasing Process
An Online Consumer Purchasing Model
How Well Does Online Advertising
Work?
• Use R O I to measure ad campaign
• Difficulty of cross-platform attribution
• Highest click-through rates: Search engine ads,
permission e-mail campaigns
• Online channels compare favorably with traditional
• Most powerful marketing campaigns use multiple
channels, including online, catalog, T V, radio,
newspapers, stores
Online Marketing Communications:
Typical Click-through Rates
The Costs of Online Advertising
• Pricing models
 Barter, cost per thousand (C P M), cost per click (C P C), cost
per action (C P A), hybrid, sponsorship
(IAB: 62% used performance based pricing, 35% used CPM
based model, and 3.3% hybrid)
• Measuring issues
 Correlating online marketing to online or offline sales
• In general, online marketing is more expensive on
CPM basis, but more efficient in producing sales
• Effective cost-per-thousand (e C P M) – Measures RoI
from an Ad by dividing the total earnings by total
number of impressions in thousands
Web Analytics - Google
• Visit the link :
https://support. google.com/analytics/answer/6367342?hl=en#zippy=%2Cin-
this-article

• Clickon “Access Demo Account” and follow

instructions
• You
need to visit the Web Analytics page for “Google
Merchandise Store”
Web Analytics Dashboard
Some metrics to track the success of
a website
1. Website traffic
 Unique visits tells how well you attract new visitors, while
repeat visits tells how well you retain visitors.
 Now, in an ideal world, unique visits and repeat visits will
grow simultaneously, giving a healthy increase in total
visits.
 What if total visits growth is driven entirely by repeat visits?
 In such a scenario, the metrics are telling that you are successfully
building a loyal following (awesome!), but you aren’t doing enough to
attract new visitors.
 You should be adopting new strategies to improve your discoverability,
which will bring new eyes to the website.
Some metrics to track the success of
a website (contd.)
2. Traffic sources
• Organic Search: traffic coming via the search engines
• Referral: traffic from another website
• Direct: traffic typing your domain into the
browser/clicking through emails or bookmarks
• Social: traffic from social media

• Which source of traffic is best?

Some metrics to track the success of
a website (contd.)
3. Bounce rate
• The bounce rate metric, displayed as a percentage, tells how many
visitors leave the website immediately after arriving—Google defines
these as “single-page sessions.” The lower the bounce rate, the more
visitors there are sticking around to enjoy the website, and
(hopefully) converting
• Bounce rates do differ based on the type of website—blogs will differ
from landing pages which will differ from eCommerce stores. You
also have to apply some logic to the situation.
• It doesn’t tell why they are leaving—
 Common reasons for a high bounce rate include: slow load times,
broken websites, bad first impression (poor website aesthetics), and
badly targeted keywords.
Some metrics to track the success of
a website (contd.)
4. Top pages
• The best performing pages in terms of traffic volume—Analytics
displays the number of pageviews, and how those pageviews look as
a percentage of total pageviews across the entire website.
• Knowing what pages receive the most traffic is hugely important
because it gives the real-world data showing what the audience
responds to. Experimenting with different types of content helps in
analyzing what’s working, and the business can produce more of the
material the readers like.
• The number of social shares per page is also as an indicator of a
strong article.
Some metrics to track the success of
a website (contd.)
5. Conversion rate
• Conversion rate is another crude top-level metric, but it’s arguably
the most important metric of all as it can have a significant impact
on the site’s profitability—if you can increase your conversion rate
from 1% to 2%, the profits double
• Total conversion numbers are important, but it’s the
conversion rate that tells how well you encourage your traffic to
perform a desired action.
• Conversion rate: Conversions/ Unique Visitors
• An eCommerce store might have three conversion goals:
▪ A sale (most important!)
▪ A subscriber to an email list
▪ A social share (least important, but still valuable)
Some metrics to track the success of
a website (contd.)
6. Conversion by traffic source
• Hypothetical example website with the following
performance metrics – How can the following be
interpreted?
1. Organic search traffic: 4% conversion rate
2. Referral traffic: 10% conversion rate
3. Direct traffic: 12% conversion rate
4. Social traffic: 1% conversion rate
Some metrics to track the success of
a website (contd.)
• Direct traffic is the most valuable traffic source for this
website; is usually driven by loyal visitors, so with this in
mind, this website should take steps to encourage loyalty
and repeat visits.
• Social traffic is the least effective, so the website should
assess their SM strategy.
• Referral traffic converts 150% better than search engine
traffic. This is an indicator that the strategy for
generating referral traffic is good (website is being linked
to from the right places) but the search engine strategy is
less effective (website targeting the wrong keywords).
Some metrics to track the success of
a website (contd.)
7. Customer’s lifetime value
• The lifetime value metric addresses this by factoring the
customer’s future purchases into the equation.
▪ If membership costs INR 1000 per month, and the average
customer remains a member for 12 months, then each customer
is worth, on average, INR 12000 to that business.
• Help to allocate resources more effectively
▪ —if they use paid advertising, they could afford to spend more
to attract customers based on the INR 12000 lifetime value
rather than the INR 1000 transaction value.
▪ This higher lifetime value means they can spend more on
advertising and promotion, will attract more customers, and
will allow their business to grow more quickly
Possible process to be followed
1. Visit the Google Merchandise Store Website. What
possible business objectives of GMS (including brand
message)? How is the website fulfilling the business
objective?
2. What specific behaviours of the users on the website?
(that help in understanding if the business
objectives are being met)
3. What web metrics can help understand the above?
4. What insights garnered from analytics?
Web Analytics - Google
• Visit the link :
https://support. google.com/analytics/answer/6367342?hl=en#zippy=%2Cin-
this-article
▪Click on “Access Demo Account”
• Assess the metrics for each stage of the Conversion
funnel
• Interpret data
• Discuss 3 findings with screenshots
Group Exercise
• Interpretations of Google Analytics for the Google merchandise
website
• Examples of questions that could be asked:
▪ Which channel is responsible for acquiring the most new users
across your different platforms?
▪ How many total unique users does the business have?
▪ How many conversions have occurred in the last week?
▪ Identify and compare the performance of different traffic
sources (organic, paid, direct, referral).
▪ Evaluate the effectiveness of social media and email marketing
efforts.