Outro

Glossary

Here are some of the words you may encounter both in the book and when you want to search for more information.

A/B testing – how you experiment with showing alternative versions of something to find out which performs best on real users.
Above the fold and the fold – the part of a web page that is visible before the user starts scrolling. The term originates from newspapers being folded and the most important material needing to be above the fold, so it was visible when you picked up the paper.
API – abbreviation for Application Programming Interface and in web analytics it concerns which tools support automation or integration. For example, through the API that Google offers for Analytics account holders, you can integrate Analytics with other digital services.
Bounce rate – refers to the proportion of visits with only one page view, after which the visitor leaves the website.
Visit – should not be confused with visitor. A visit is when a person visits the website. A visitor can make several visits during a time interval and generate multiple page views.
Visitor – indicates a unique visitor to the extent possible, or the number of unique visitors you have had.
CMS – abbreviation for Content Management System, the information system that often constitutes the website itself. Usually contains a public part that is the public website, and a logged-in part where editors work with the website's content.
Cookies – the small text files placed on a user's computer or device so that we as website publishers can keep some track. For example, it is common to use cookies to keep track of what a user has placed in their shopping cart, how far they have progressed through a multi-step process, etc.
Direct traffic – visitors who type the address directly into the browser or arrive via a bookmark.
Click-through rate – also CTR, abbreviation for Click Through Rate, refers to the proportion who choose to click on a particular link or button they have on screen.
Graceful degradation – a design approach where you provide an optimal solution for those with the most modern equipment. Through various hacks and error handling, you make it work okay for everyone else too. See also progressive enhancement.
Event tracking – involves tracking events that are not necessarily a page view. This can include a button being clicked, a document being downloaded, an external link being clicked, etc.
Referring websites – websites that have a link to your website.
Inbound link – a specific link on another website that refers to a page on your website.
Content delivery network – often CDN, abbreviation for Content Delivery Networks. Having certain parts of your website's material at a different address, possibly with a different provider. This is to offer better performance, higher speed or to offload your own web server.
Conversion – when you have transformed an interested party on the website into a customer, or a similar "refinement" of a relationship with someone you are interested in dealing with.
Correlation – a statistical term that indicates the strength and direction of a relationship between two or more variables. Correlation is often expressed with a so-called correlation coefficient.
Landing page – extra important subpages that can be seen as start pages for certain topics or purposes of the website. Often where you drive traffic through marketing and campaigns.
Logging – keeping a digital logbook of everything that happens in a system, app or similar. Almost all software logs and the content can be interesting for gaining insight into a system's usage.
Baseline measurement – initial measurement, or evaluation, of something in order to later be able to determine the effect of a change.
New visits – indication of how many of the visitors have not previously visited the website.
Key performance indicator – often called KPI, abbreviation for Key Performance Indicator, puts a number on the usefulness or efficiency of something so it can be compared over time. Key performance indicators can for example be customer satisfaction, efficiency in performing a certain task, profitability for a product group, and more.
Organic search results – search results that are not paid for, meaning you have earned your position according to a search engine's algorithms. The opposite is paid search results, which usually go by the term SEM or search marketing.
Outbound link – links that point to other websites than the one the user is currently on. Sometimes called external links as they refer to someone else's website.
Progressive disclosure – a design principle that only shows the options needed to complete something, making it easier for the user to focus.
Progressive enhancement – focuses on offering a good basic function for all users, basing design decisions on a lowest common denominator. See also graceful degradation.
Referrer – the website that refers traffic to yours. For example, if a user comes from www.gp.se, GP is the referrer for the user they sent your way.
Responsive images – since screens can have different sizes and capabilities, there is the concept of adapting images so they perform well on many different screens. One solution is to send different sized images so as not to weigh down connections more than necessary, another is to choose pixel density based on whether the screen is high-resolution. Yet another is to offer different crops depending on what the image contains so its message comes through.
ROAS – abbreviation for Return On Advertising Spending, a comparison between the expenditure on advertising and the revenue attributable to the same advertising. To arrive at ROAS, you divide the return from an ad by the cost of the ad. If the value is below 1, you have spent more money on the ad than you got back.
ROI – abbreviation for Return on Investment, concerns the return obtained as a result of an investment. It is a measure of the impact an investment has had. A high ROI means you have gained a lot from the investment. A low ROI means the investment has not covered its own costs.
Segmentation – means that you have made a selection among users in order to evaluate statistics. A segment can be to only look at those who connected with a mobile device, another to select those who started their visit on a particular landing page, a third is to make a geographical delineation of where the users were located.
Page load time – the time in seconds it takes for a visitor to load all material in a single page view.
Pages/visit – a measure of the average number of pages viewed during a user's session on the website.
Page view – the number of times a page has been shown to a user of a website.
Tracking code – the code that runs in the user's browser to record their visit and activity. Often abbreviated to GATC (Google Analytics Tracking Code) when it comes to Google's Analytics tool.
Search engine traffic and organic traffic – the visitors you attract thanks to your presence on search engines.
Time to first byte – often abbreviated to TTFB, refers to the time it takes for the server to respond or begin sending a file to the user. It is a measure of how fast a server's response or reaction time is. Sometimes also called latency.
Perceived performance – the semi-subjective feeling a user has of performance. There are tricks that improve perceived performance but do not actually affect measured "technical" performance. The experience is of course most important.
URL – abbreviation for Uniform Resource Locator, is what is commonly called a web address, or simply address. It looks something like this: https://website.se/subfolder/filename.html?querystring1=hello&querystring2=world#internallink
Virtual page views – when a page is loaded on the user's screen. A "regular" page view is when a page with its own address is loaded in the browser. Since not all perceived pages are a regular page view, you can fake it and register a change on a web page as a virtual page view. A regular page view requires the page to reload, a virtual page view you can define yourself.
Site search – the search function on a website that searches for material within that specific website.
Web standards – an umbrella term for the good open practices that exist for the web. Among many other things, that you should follow standards, that content should be accessible, etc.
Web performance optimisation – sometimes abbreviated to WPO, concerns the methods and tasks aimed at making a website faster in its service.

References

University of Gothenburg reported for discrimination:
sverigesradio.se/sida/artikel.aspx?programid=104&artikel=6101962

Web Service Awards survey "How are Sweden's websites doing in 2015":
blogg.webserviceaward.com/2015/06/10/vad-ar-poangen-med-webbplatsen/

Google trims Google+ during 2015:
developers.google.com/web/showcase/case-study/googleplus

Swedish Web Guidelines – guideline 97:
webbriktlinjer.se/r/97-lat-bakatknappen-fungera/

Facebook on their 2G Tuesdays to understand slow connections:
code.facebook.com/posts/1556407321275493/building-for-emerging-markets-the-story-behind-2g-tuesdays/

Gov.uk on their content budget:
insidegovuk.blog.gov.uk/2014/08/04/sentence-length-why-25-words-is-our-limit/?mc_cid=780729def0&mc_eid=7a6d736a2e

CSS-tricks on the challenges with custom fonts:
css-tricks.com/fout-foit-foft/

Matomo – free and open source solution for collecting website statistics:
matomo.org

Facebook error takes down countless major websites:
buzzfeed.com/jwherrman/facebook-glitch-takes-down-countless-major-websites#.ftkPJ3w1zn

On stopping spam bots from ruining your data quality:
moz.com/blog/how-to-stop-spam-bots-from-ruining-your-analytics-referral-data

References on web performance

Google: 2% slower = 2% fewer searches per user – assets.en.oreilly.com/1/event/29/Keynote%20Presentation%202.pdf
Tagman: For each second of load time, you lose 6.7% conversion – info.ensighten.com/rs/ensighten/images/just-one-second-delay-in-page-load-can-cause-7-percent-loss-in-customer-conversions.pdf">info.ensighten.com/rs/ensighten/images/just-one-second-delay-in-page-load-can-cause-7-percent-loss-in-customer-conversions.pdf
Yahoo: 400 milliseconds faster = 9% more traffic – slideshare.net/stoyan/dont-make-me-wait-or-building-highperformance-web-applications
AOL: Faster pages = more page views – assets.en.oreilly.com/1/event/29/The%20Secret%20Weapons%20of%20the%20AOL%20Optimization%20Team%20Presentation.pdf
Amazon: 0.1 seconds faster = 1% more revenue – radar.oreilly.com/2008/08/radar-theme-web-ops.html
Gomez: 75% of users in web shops would rather leave the website for a competitor than patiently wait – mcrinc.com/Documents/Newsletters/201110_why_web_performance_matters.pdf
Shopzilla: 5 seconds faster = 25% more page views, 7 to 12% more revenue – conferences.oreilly.com/velocity/velocity2009/public/schedule/detail/7709
Aberdeen Group's report from November 2008: "The performance of web applications", 1 second slower = 11% fewer page views, 7% fewer conversions – v1.aberdeen.com/launch/report/SmartBite/10179-SB-bigdata-performance-management.asp
Etsy: 160 KB of images increased the bounce rate by 12% – radar.oreilly.com/2014/01/web-performance-is-user-experience.html
Firefox: By reducing the average page load time by 2.2 seconds, their download conversions increased by 15.4% – blog.mozilla.org/metrics/2010/04/05/firefox-page-load-speed-%E2%80%93-part-ii/
Walmart: Drastically reduced conversion rate the slower the user's experience, clear correlation between page load time and conversion – webperformancetoday.com/2012/02/28/4-awesome-slides-showing-how-page-speed-correlates-to-business-metrics-at-walmart-com/
Google includes web performance in its ranking algorithm – googlewebmastercentral.blogspot.com/2010/04/using-site-speed-in-web-search-ranking.html

Academic research on performance, response times, etc.

1968: "Response Time in Man-Computer Conversational Transactions" (PDF), Robert B. Miller, Fall Joint Computer Conference 1968
1991: "Response Times: The 3 Important Limits," Jakob Nielsen, Nielsen Norman Group
2004: "A Study on Tolerable Waiting Time: How long Are Web Users Willing to Wait?" (PDF), Fiona Fui-Hoon Nah, Behaviour and Information Technology, 2004
2005: "Interaction in 4-Second Bursts: The Fragmented Nature of Attentional Resources in Mobile HCI" (PDF), Antti Oulasvirta, Sakari Tamminen, Virpi Roto, and Jaana Kuorelahti, Interruptions in Human Computer Interaction
2006: "Quantifying Interactive User Experience on Thin Clients" (PDF), Niraj Tolia, David G. Andersen, and M. Satyanarayanan, The Internet Suspend/Resume Project, Carnegie Mellon
2012: "Characterizing Web Use on Smartphones" (PDF), Chad C. Tossell, Philip Kortum, Ahmad Rahmati, Clayton Shepard, Lin Zhong, Conference on Human Factors in Computing Systems 2012
2011: "Playing With Tactile Feedback Latency in Touchscreen Interaction: Two Approaches" (PDF), Topi Kaaresoja, Eve Hoggan, Emilia Anttila, Human-Computer Interaction: INTERACT 2011

Images

Image 15: Seesaw, by harmish khambhaita, licence CC-BY-SA 2.0
Image 19: Confidence interval, by Audriusa on Wikimedia, licence GFDL
Image 64: Simonw on Twitter about FOUT, twitter.com/simonw/status/732136252086984705

Book and reading recommendations

Just Enough Research, Erika Hall, ISBN 1937557103 tba.nu/wajust
Don't Make Me Think Revisited, Steve Krug, ISBN 0321965515 tba.nu/wasteve
Successful Analytics, Brian Clifton, ISBN 1910591009 tba.nu/wabrian
Time is Money: The Business Value of Web Performance, ISBN 9781491914090 tba.nu/watammy
Search Analytics for Your Site, Louis Rosenfeld, ISBN 1933820209 tba.nu/walouis
Google Analytics Integrations, Daniel Waisberg, ISBN 1119053064 tba.nu/wadaniel
Why: A guide to finding and using causes, Samantha Kleinberg, ISBN 1-4919-4961-9 tba.nu/wasamantha
Statistics Done Wrong, Alex Reinhart, ISBN 1593276206 tba.nu/waalex
Spurious Correlations, Tyler Vigen, ISBN 9780316339438 tba.nu/watyler

Of course, check out my blog too – webbstrategiforalla.se – it has quite a bit about web analytics :)

Acknowledgements

My colleague Lars Lindsköld who constantly turns and twists every question into new angles. Morgan Skogh who helpfully answers everything I lump together as "mathematics", such as probability calculations. Colleagues Per Norin and Jonas Hägglund from Västtrafik, for all input on statistics and campaign analysis. Sven-Erik Järnberg who is a helpful devil's advocate in discussions and experiences that broaden perspectives. Sebastian Danielsson who in addition to showing how wisely one can work in the public sector also forces me to nuance my writing. The world's best Anders Lövkvist and Filip Dalväg, friends to rely on when you urgently need to bounce something over lunch to make progress in writing.

All my wonderful proofreaders who with a critical eye made the book better. Pernilla Thor has done a fantastic job, thank you! I have also received fine contributions from Leif Arndorff, Marianne Berggren and Fulton Yancy.

Kristian Norling and the publisher Intranätverk who publish this book and are a great support during the writing.

Someone I all too easily forget to thank is my dad, Christer Österberg. He has had the thankless task of questioning my generalisations ever since I was in nappies, forced me to think new thoughts and see things from angles other than the most obvious.

But the biggest thanks goes to my dear wife, Anna Johansson. Who with an angel's patience has busied herself with her own projects while I worked on the book, but has always been on hand for questions about everything under the sun. There is no better company, and by the way, the book's design is hers!

Footnotes

If a link breaks, please report it to the author. We will then update the link as soon as we can. Thank you in advance.