Thu

Oct 1
2009

Jesse Robbins

More on how web performance impacts revenue...

by Jesse Robbins | @jesserobbins | comments: 10

At Velocity this year Microsoft, Google and Shopzilla each presented data on how web performance directly impacts revenue.

Their data showed that slow sites get fewer search queries per user, less revenue per visitor, fewer clicks, fewer searches, and lower search engine rankings. They found that in some cases even after site performance was improved users continued to interact as if it was slow. Bad experiences have a lasting influence on customer behavior.

What about smaller websites that aren't yet at this scale?

Alistair Croll and Sean Power, the authors of the new book Complete Web Monitoring, have continued this research for sites at smaller scale.

They used a Strangeloop Networks web acceleration appliance to optimize half the sessions to a smaller production website, tagging optimized and unoptimized visitors so they could be analyzed in Google Analytics. The Strangeloop device applies many of Steve Souders' performance rules to an existing site automatically (a kind of "Steve-in-a-Box" ;-).

The results of their analysis show how significant a reduction in page latency can be. In addition to reducing bounce rates, and increasing pages per visit & time on site, they found a 16.07% increase in conversion rates and a 5.50% increase in average order value.

conversion-rate-and-order-value.png

Check out the full post on the Watching Websites blog.


tags: alistair croll, book related, operations, performance, velocity, velocityconf, watching websites, web monitoring | comments: 10
submit:

 
Previous  |  Next

0 TrackBacks

TrackBack URL for this entry: http://blogs.oreilly.com/cgi-bin/mt/mt-t.cgi/9826

Comments: 10

manuel [2009-10-01 08:42 AM]

good info

Atle Iversen [2009-10-01 09:37 AM]

Hmmm... very interesting !

The impact was bigger than I would've imagined - time to test our website again.

Also, thank you for the link to Steve Souders' performance rules - very useful !

BWI [2009-10-01 03:25 PM]

Always new it was a big deal, but seeing figures like that it's time to review and reduce....

Don [2009-10-02 12:00 AM]

What are the units of the horizontal axis in the top graph (Impact of Page...)?

denparser [2009-10-02 10:12 AM]

it's true.. and I want it too.

Greg Horby [2009-10-02 02:46 PM]

This changes the game- I keep telling my boss we should do this but I don't have any numbers to validate it- thank you for making my business case real.

RLM [2009-10-02 06:43 PM]

Why should this be surprising? Historically, users have always hated lags between a behavior and an expected result. Typists hate lags bewteen print response and key press, Software users hate slow reponse between a widget selection and subsrquent screen display. My wife hates it when the browser won't open when she wants to check her email... And the beat goes on....

Daniel Brockman [2009-10-04 01:25 PM]

RLM - As you point out, it's not surprising, but as Greg Hornby suggests, those who control budgets in corporate heirarchies manage by measurements and won't commit funds without them. I recall once having a conversation with a CIO in which I advocated user-friendly inhouse applications, and he responded that there's little evidence of demonstrable value in user-friendliness. Public websites and the people who want them to run faster will benefit greatly from these findings. Next (I hope): Intranets. [ I see a note on the Post-A-Comment page: " (please be patient, comments may take awhile to post)" :) ]

Akishige Iwata [2009-10-04 08:24 PM]

Be careful. No causal relationship has been found/proved. The data just says there is a proportional relationship between the response speed of a website and the customer response to the site. That happens maybe because:
- The website is famous. The website provides useful, attractive information or products.
- Therefore, customers like the site. They give more queries, click more links within the site, drop more money, get more satisfied, etc.
- Therefore, the company who owns the website earns more money.
- Therefore, the company can spend more budget in the hosting server, the software development, and the network.

If the scenario above is true, the explanation the author has becomes false.

The point is, we need more investigation into whether the site response does have impact on the site revenue.

mansoor [2009-12-04 02:27 AM]

nice info!!

Post A Comment:

 (please be patient, comments may take awhile to post)





STAY CONNECTED

Subscribe to Radar Subscribe to Radar
Follow Radar on Twitter Follow Radar on Twitter

RECOMMENDED FOR YOU

  1. O'Reilly Tools of Change for Publishing Conference, February 22 - 24, 2010, New York, NY
  2. Where 2.0 Conference, March 30 - April 1, 2010, San Jose, CA
  3. O'Reilly MySQL Conference & Expo, April 12 - 15, 2010, Santa Clara, CA
  4. Web 2.0 Expo, May 3 - 6, 2010, San Francisco, CA
  5. Gov 2.0 Expo, May 25 - 27, 2010, Washington, DC
  6. $249.00
    Twitter and the Micro-Messaging Revolution, OReilly Radar Report

RECENT COMMENTS

MOST ACTIVE

RADAR TOPICS