Have taken a data table and turned it into a, “Bumps Chart”. Like the data table, the Bumps Chart does not show magnitude; it shows just, “first”, “second”, “third”, etc.
Not visible is the metric being charted, which turned out to be the rate per item-byte.
#Bumpschart
Regarding yesterday’s CotP, an astute reader asked why there was such a focus on the illusory, central-tendency, “Arithmetic Mean Average” but not on the, “Standard Deviation”. Thought that was a good callout since will generally want to look at both types (a central-tendency type and a dispersion type). As such, Chart Four has been altered (the previous CotP remains unaltered).
Seeing data in its different information forms and states.
Web performance data sets five and six are actual; the other are illustrative.
I’ve taken the two-dimension CDF, “Hockey Stick” chart and flattened it into a one-dimension strip. The other aspect regarding the X axes is they are normalized so we may compare variables of different type and magnitude.
We then take this example visual and put into a report (download here). If you want this report generated for your home page, feel free to contact me @LvasiLiou , or using the contact info in slide 2 of the report.
….
Data from the top 100’ish Internet Retailer default Home Page loads. Is there correlation between the #Requests and Webpage Response times?
Special note, the X axis in the two charts are not normalized.
Download excel chart data https://drive.google.com/file/d/0B9n5Sarv4oonUkNDUFE3R1AtUFU/view?usp=sharing
…
Use a multi-width column (or bar) chart to show change between two data (e.g. performance of a page load with an empty browser cache versus performance of a page load with a primed browser cache).
Download Excel file here.
Credit and thanks to Jon Peltier for writing how to do this in Excel.
…
Hello! This Webtortoise post was written 2014-JUN-30 at 09:01 PM ET.
#- Use a Slopegraph to display, “change between two data”. For example, use a Slopegraph to display the change in Performance between a, “page load with an empty browser cache” versus a, “page load with a primed browser cache”.
#- In his [recommend you read this] document, Stephen Few talks about using a Slopegraph to display change between two points in time. So, special callout am using the Slopegraph to display change between two non-time points (in this case, “Browser Cache: Empty” versus, “Browser Cache: Primed”.
#- When using a Slopegraph to show Rank, put the “Fastest” time at the Top (more on this below).
#- Usual Webtortoise disclaimers apply. Whether it’s Document Complete versus Render Start, or IE versus Chrome, or Synthetic versus RUM or “X” versus “Y”… it’s the underlying chart/graph principles that are important.
Hello everyone. In this WebTortoise Story, am going to use a Slopegraph to visualize the rank in Page Load Times RE: an Empty Browser Cache versus a Primed Browser Cache. Traditional data tables or bar graphs have always been used for this, but as Webtortoise is about the many different ways to visualize web performance data, let us begin.
First, consider a general data table (not unlike the one below). It is showing the “Empty Cache” Average Document Complete Time (in ms) for some of the Internet’s top retail sites:
Second, consider a similar general data table showing the “Primed Cache” average Document Complete Time (in ms) for those same sites.
Pause….
Now consider such questions as:
— What is the rank of sites when considering an empty browser cache versus when considering a primed browser cache?
— Does the ordinal rank change? And if so, by how much?
Now… place the tables side-by-side, draw lines between them and voila! you have your Slopegraph:
There was much debate about whether or not the “fastest” site should be at the “top” or at the “bottom”. Since were making a natural progression from a basic data table (which normally has the “fastest” at the top), decided to keep the “fastest” at the top when constructing the Slopegraph.
One of the powerful capability assets of the Slopegraph is you get to visualize both, “the actual rank” and, “the ordinal rank”. This is exceptionally powerful because when you have *just* the data table, it’s tougher to do things like see clusters, compare magnitudes of change or see rank change(s).
In other words, the ordinal rank says, “So and so is first, such and such is second and so on”. But it does not say, “By how much is such and such second”? That’s why the visual is recommended because it conveys some of this certain information with much less exerted brainpower.
Download Excel: https://drive.google.com/file/d/0B9n5Sarv4oonTVd3dm01YWd1dlk/edit?usp=sharing
LinkedIn: http://www.linkedin.com/in/leovasiliou
Twitter: @LvasiLiou
#Analytics #CatchpointUser #ChartsAndDimensions #ChartsAndGraphs #Performance #SiteSpeed #WebPerformance #Webtortoise #WebPerf #WPO #DataVis
#ExcelSlopeGraph #CompetitiveBenchmark #CacheIsKing
Using a Slopegraph to show change. In this case, are showing the Page Load Times (in milliseconds) of an, “Empty Browser Cache” versus a, “Primed Browser Cache”. Note the faster times are toward the top of the Y axis.
Hello! This Webtortoise post was written 2014-APR-01 01:41 PM ET.
Hello, Everyone! Am pleased to announce there is finally a Web Performance Panacea Chart we can all use in every data analysis situation! The power of the Web Performance Panacea Chart can obviously be seen by just glancing at this example:
A few folks have stated their interest in having a “lights out” version of the same Web Performance Panacea Chart. And so, after much work and development, I am pleased to offer THIS example:
So, to give an idea of when to use the Web Performance Panacea Chart, think of some of these use cases:
– I’d like to have a chart showing me when there is a problem on my site ;
– I’d like to have a chart showing the outside weather ; or
– I’d like to have a chart showing me the mood of my significant other .
As you can, the Web Performance Panacea Chart can show all of these! What’s even more powerful, is the previous charts can be combined into a a super Web Performance Panacea Chart as shown below:
LinkedIn: http://www.linkedin.com/in/leovasiliou
Twitter: @LvasiLiou
#AprilFools
Web Tortoise 