Kissmetrics Blog

A blog about analytics, marketing and testing

Built to optimize growth. Track, analyze and engage to get more customers.

How Loading Time Affects Your Bottom Line

Page loading time is obviously an important part of any website’s user experience. And many times we’ll let it slide to accommodate better aesthetic design, new nifty functionality or to add more content to web pages. Unfortunately, website visitors tend to care more about speed than all the bells and whistles we want to add to our websites. Additionally, page loading time is becoming a more important factor when it comes to search engine rankings.

Next week we’ll post our complementary article: “Speed Is A Killer – Why Decreasing Page Load Time Can Drastically Increase Conversions” to elaborate more on this topic. For now, please enjoy and share the infographic below.

How Loading Time Affects Your Bottom Line

View an enlarged version of this Infographic »

Click here to download a .pdf version of this infographic.


Facts and Stats to Tweet:

  • 73% of mobile internet users say that they’ve encountered a website that was too slow to load. »tweet«
  • 51% of mobile internet users say that they’ve encountered a website that crashed, froze, or received an error. »tweet«
  • 38% of mobile internet users say that they’ve encountered a website that wasn’t available. »tweet«
  • 47% of consumers expect a web page to load in 2 seconds or less. »tweet«
  • 40% of people abandon a website that takes more than 3 seconds to load. »tweet«
  • A 1 second delay in page response can result in a 7% reduction in conversions. »tweet«
  • If an e-commerce site is making $100,000 per day, a 1 second page delay could potentially cost you $2.5 million in lost sales every year. »tweet«

About The Author: Sean Work is the Minister of Propaganda at Kissmetrics. Follow him on twitter (@seanvwork) or on .

  1. The current tool set for evaluating these types of correlations are extremely limited in my experience. We have been closely working with gomez, akamai, and other vendors in the past and no one has a solution here in my opinion. A good way to monitor web application performance and correlate conversion rates is what we would love to see. I am curious about where your numbers come from, I see your references and I have been asking both these companies for a way to measure our site for months.

    • No one has gotten back to you?

    • Absolutely true Orion. There are tools which help one to focus on conversions, improving user experience like heatmaps etc., I still think that not focusing on “page load time” and doing all the other stuff to impress the user makes no sense. Interactive graphics developed using javascript, jquery or flash will be of no use if one cant make the site load faster to the user. Hence as Orion said a tool to monitor web application performance and correlate conversion rates is a big void yet to be filled.

    • Pinged you with the solution to this problem, exactly why we where created to diagnose, solve and prove page load times across all devices.

    • How about performance testing the applications by simulating real world network environments? This would allow the developers to optimize code for disparate networks prior to going in production.

  2. The irony is that the infographic loaded too slowly on my iPhone, and I abandoned it. Read it later on the computer instead.

    • Thanks for the feedback Jonathan, we optimizied the files however that will always be an issue with infographics, they have no set standards to follow. I think we’ll need to look into how to cater better for mobile web users.

    • Lol, yeah, mobile is definitely an avenue people must begin to explorer more and more.

  3. Billy Hoffman Apr 29, 2011 at 6:44 am


    This is because solutions from vendors like Gomez, Keynote, etc only tell you how slow or fast your pages are. They don’t identify the what’s making them slow or things you can do to make them faster.

    For that you want to couple the insight from KISSMetrics, Google analytics with a tool to find specific perfrmance problems. Take a look at free tools like YSlow, PageSpeed, WebPageTest, as well as the free scanning service offered by my start-up Zoompf. Really helpful to find ways to improve your landing pages or pages with high drop-off rates.

    Good Luck

  4. I was very interested in the topic, but could not be bothered to look through the infographic for interesting bits. Would have much more preferred a well written blog post.

    • what a retarded comment! infographics are proven to be more engaging, hence their popularity…

  5. Thanks, a very useful infograph.

    “A 1 second delay in page response can result in a 7% reduction in conversions.” – this is a powerful statement!

  6. “40% of people abandon a website that takes more than 3 seconds to load.” But the graph shows around 20%. Which is it?

    • Tim, the 40% is for website visitors inside a shopping cart or in the middle of a shopping experience. The 20% is for general web surfing.

  7. 58.3% of statistics are made up – That’s how this infographic makes me feel. Just tossing out blanket statements like “A 1 second delay in page response can result in a 7% reduction in conversions.” without any data or details as to which page, etc. to back it up, makes me very skeptical of this whole set of information.

  8. I wish this was a little bit more clear, like do people abandon a partially loaded page after 3 seconds? Or do they leave because the wheel is turning but there’s nothing but a blank screen.

    I hate slow sites as much as anyone else, but I find I’m only ever inclined to leave when it takes too long and all I see is white. If it’s showing part of the site, I’ll usually hit refresh or wait, because I’m really not in that big of a hurry.

  9. Nice, but I’m a bit confused by the “Mobile vs Desktop” figures.

    31% thought their phone would be “a bit slower”, while 25% said they thought their phone would be “almost as fast”.

    To my mind those two phrases have the same meaning, so in effect we’re left with 56% of people expecting slightly longer loading times on their phones?

    • Even thought the responses are similar, they are still giving two answers. Just like when you do a survey and you get answers like very satisfied and somewhat of satisfied. They are similar, but different.

  10. Yaah, really Page loading time is obviously an important part of any website’s user experience. And many times we’ll let it slide to accommodate better aesthetic design, new nifty functionality or to add more content to web pages.This is one of the great post to increase our knowledge and decrese our problems.

  11. We read all point, Psychology experiments can range from simple to complex, but there are some basic terms and concepts that all students of psychology should understand. and important one is data time effect. Time effect of a data set refers to a time interval in … decomposition, the time effect of the new data will change..

  12. Not sure this page passes the 2 second rule but great infographic all the same.

    The 7% loss every second is a Forrester stat. We published a study a while back that found traffic leaves a page at 1% as second.

  13. That’s why I love Kissmetrics:
    – easy navigation
    – friendly interface
    – fast loading
    – (this should be #1) great content

    Haven’t tried to read the blogposts with infographics from my iphone, yet, but they work great on any of my laptops/pc’s

  14. Another aspect is that page load time may affect Google rankings isn’t it?

    For ordinary sites many people says that load time affects search engine ranking Most of my pages for all of my sites load in less than 1.5 seconds and I rank very well. I would aim for the 1,5 mark as set forth in Google Webmaster tools page speed graph.

    Yet for more i am not sure. But it may depend of niche for sites that have a lot of pictures Google may expect slower loading time.

    Also i agree for buyer traffic these people who buys usually have fast internet connection and they expect for load times to be very fast.

  15. This is nteresting…Was Wthe percent of users with slow Internet connection (which I guess is decreasing) and mobile devices was taken into account? There must be a certain boundary when optimizing loading speed is not cost-effective yet.

  16. Today almost 10% of searches are made through mobile devices (in USA). SO it is becoming more and more important to be sure to use gzip for the text data compression as well as smaller images on mobile version of your site. Also sites shouldn’t be “to big” because even better phones such as Android have quite slow processor when comparing to desktop so sites with too many elements are going to have large load time. Also wireless connection is slower worse, and it already has huge ping.

  17. Anyone care to guess the page speed of this page? Chrome dev tools shows load time is ~6 seconds, generating 269 HTTP requests.

    A really slow page that tells you about how important page speed is pretty much the definition of irony.

  18. Lesser the loading time lesser the bounce rate. So one must really watch out for the loading site of the website.

  19. Too bad cloudfront doesn’t support gzipping. I’d like to see KM come to our site gzipped.

  20. “If an e-commerce site is making $100,000 per day, a 1 second page delay could potentially cost you $2.5 million in lost sales every year.”

    I am guessing this was a direct mashup of abandonment rates versus conversion rates. Recognizing “potentially” is used, the likely scenario would not be hitting that ceiling. Patrons who are predisposed to buying are also predisposed to being more patient (their desire outweighs the hurdle). The potential customer who would most be affected by a slow load time tends to also be the customer who abandons due to lack of interest in the product (the window shopper).

    Load time affects shopping cart attrition, I won’t argue, but the relationship of the two statistics is much more complex than a mashup of the two numbers.

    It follows the same philosophy of security, where employing a little security only dissuades people who really don’t have interest in breaking through your security anyway. In the end, your number of break-ins is relatively unchanged.

  21. Clinton Skakun Nov 01, 2012 at 12:28 pm

    I’m almost wondering if having graphics on a mobile blog is even necessary. Obviously for a post like this it’s pretty important as the graphic makes up most of the post content. However, most images in blog posts don’t contribute that much to the content. I’m sure no one’s going to leave because there’s no image.

  22. Sandeep Kumar Feb 08, 2013 at 7:30 pm

    This is really interesting data…with the increment of more and more internet users, This ought to be considered the most considerable facts for website owners. btw is there any standard, I mean how much second load time is the best. I have dtp load time of around 3.06 seconds. Is this good or bad?

  23. Casey Dennison Mar 10, 2013 at 5:14 pm

    As mentioned in the infographic, mobile users expect their browsing experience to be close to what they would expect from a laptop/desktop.

    I clicked on the infographic (using my phone) and it took way to long to load up – leaving me no choice but to abandon the page.

    Perfect example!

  24. Over the last year Google and their applications (like YouTube) have made a consistent effort to reward businesses with fast load times. I think now it’s something like 3 seconds on your site, and less than 2 minutes on YouTube to make an impression on SERPS and visiting users. Technical SEO is going to be huge in 2015!

  25. Of course loading time is important to avoid bounce – but so is navigation and appearance.
    Too many sites (apparently) re-designed for mobile Users suck for desktop Users.
    Principally unclear navigation and too much scrolling.

    Recently re-designed sites often strip out the visual clues that distinguish an active link from a passive header.
    Details are no longer there such as borders between contents items, shaded Tabs, different fonts for active text versus passive, shading to indicate depth where panel/window hierarchy is helpful, rounded buttons.
    Without such visual clues, Contents now look little more than a text list.
    Is that row of plain text a menu bar?
    A monotone rectangle: Is it a header or a button?

    Surely the extra seconds the User needs to discover with the pointer what is active is just as discouraging a extra load time?

    Aesthetics is a fancy word for taste.
    For some minimalism is cool. For others lack of detail says “lazy”, “can’t be bothered”.
    Is it surprising that some Users find returning to flat, empty or monotone site unappealing?

    A successful business once said. “We do all the work so you don’t have to.”

    So why make the User scroll? (especially if the amount of scrolling is either unpredictyable or inconsistent from page to page)
    Why hide so much info and so many links below the fold?
    How much value to the User is there in an initial view of a HUGE photo or panel with LARGE but uninformative text?

    What is so wrong with maximising (within reason) the amount of info in the first view?
    Surely something will whet the appetite?
    The human eye and brain is capable of assimilating a large amount of info. at a glance.
    Minimalist sites with large photos and panels look dumbed-down, treating vistors as less than children.
    If asked to scroll down many views to see just crumbs of info, surely the User might think “I’ll go somewhere where I can get the same info in just 1 or 2 pages.

    Compared to an extra second load time, sites with the (lack of) features described seem way more likely to suffer “bounce”.

  26. Hi Sean,

    A very helpful infographic.

    Site speed is one of few very important considerations of any website. A faster web site always means a better experience for the reader or visitor.

    Whereas a slow site can grow the bounce rate out of a poor user experience. Which means drop in page views & most important, huge loss in revenue. The worst impacted are the e-commerce sites. Thanks for sharing the analysis.

  27. Great article. I can’t believe I haven’t found this earlier. I always thought the “load time story” was pushed by people who want to sell expensive hosting and website upgrades. This article makes a very convincing case. Time to do some work on my website :-)

    • the data is seriously out of date (years to be exact), especially the mobile user data. Users are far more demanding these days.

  28. The info graphic states that most people would wait 6-10 seconds for a page to load. The percentage who responded to this is 30%. This actually means that most people (70%) would wait a length of time that is outside the range of 6-10 seconds.

    The information is great, but the claims are misleading.

    • Hey Mb,
      What the infographic means is that the biggest *group* of people (30%) 6-10 seconds. 30% is the highest percentage, and that range (6-10 seconds) is what most people said they’d be willing to wait. So, yes, other people said they’d wait 11-15 seconds, or 20+ seconds, etc. But most people said they’d wait 6-10 seconds than any other range.

      I can, however, see how you may find this misleading.

  29. Absolutely right. Page loading affects business at greater extent. Nowadays, mobile users are more than desktop users, and they doesn’t have patience to wait for the page to load. Once they find page is slow, they will close the site and move to competitor sites. Yourself you are making way for competitor to grow business. They may remember the site, and even next time too they may ignore site thinking its slow.

    This is one of the major factor to be resolved for business owners. Make sure to optimize site to have page load less than 3 seconds. Else your business would be at great risk of losing customers.

  30. Hi guys

    Another great piece of content! I always dig your content, and love to keep following your journey! I actually just wrote an article (5000 words+) on LinkedIn, and included this great article about speed on websites.

    You can find it here:

    Hopefully one day I will have a client using kissmetrics, and then I will let you know what I think! :)

    Best regards
    Micky Weis

Comments are closed.

← Previous ArticleNext Article →