If website performance is important to you, then you’ll know just how vital it is to the success of your business’s website. To AlertBot, web performance is everything. This topic is of great interest to us, as we live and breathe web performance on a daily basis. It got us thinking – we all love a good head-to-head, mano-a-mano rivalry: Tyson vs Holyfield. The Hatfields vs The McCoys. The Jets vs The Sharks. Prego vs Ragu. Luke vs Vader. So we thought, what if we tracked the performance of two websites within a certain genre and pit them against each other. Who has the better website performance? Who will come out on top?
Every Fall, Apple releases a new iPhone like clockwork. But Apple isn’t the only game in town. With Apple celebrating the recent release of the iPhone 7, Samsung has their Galaxy S7 (which released in March). So we decided it was fitting to have Apple.com go toe-to-toe with Samsung.com. The results were not unexpected. (Well… most of the results.)
When you have companies as serious about their products and innovation as these two, you’d expect their websites to perform impeccably. And, honestly, they did.
We tracked the sites and examined three weeks in September – the 1st through the 22nd – to see how these sites performed. During this timeframe, we tested the websites around the clock from 17 different locations across the United States using AlertBot’s TrueBrowser Monitoring. The tests were performed by loading their homepages inside real Firefox browsers and giving them a maximum of 7 seconds to render and become fully interactive. Anything beyond 7 seconds (which is well above the average expected page load time) was considered a failure. After compiling all the data, this is what we found:
When we examine the reliability of a website, we’re looking for failure events – like when pages don’t fully load or go down completely – and try to identify the cause of the failure. Some common causes are slow third-party code used on pages, incomplete page content, actual web server failures, etc.
For Samsung, their website experienced no failure events during our test period, and achieved 100% uptime. This is definitely above the norm for website performance, but not unexpected for a company like Samsung. We would have loved to find some juicy failure-generated data to talk about, but Samsung’s website was as clean as a whistle on this front. (Samsung Score 10/10)
Similarly, Apple.com experienced no failure events and achieved 100% uptime. While I’d expect nothing less from a juggernaut like Apple, it’s still impressive when you consider other retailers that experience frequent website issues. (Apple Score 10/10)
When we evaluate a website’s speed, we’re looking at the time it takes the site’s homepage to render and load to the point of being fully interactive. We run these tests inside real Firefox web browsers using AlertBot’s TrueBrowser ™ monitoring.
While evaluating the speed of the websites specifically, Samsung.com’s fastest day was Friday, Sept. 2nd, with its slowest day being Saturday, Sept. 3rd. On average, the site’s homepage took around 1.7 seconds to load. That’s not bad at all! Some recent studies have shown that the expected load time for sites in ecommerce to be 2 seconds or less, so Samsung definitely fits the bill here. Some online studies have determined that if an ecommerce site is making $100,000 per day in sales, just a 1-second page delay could potentially cost the company $2.5 million in lost sales per year. (Ouch!) On its slowest day (Sept 3rd), Samsung.com saw some load times in the range of over 7 seconds at times during the day. (Samsung Score 9/10)
While evaluating Apple.com’s speed, its fastest day was also a Friday, on Sept. 9th, with its slowest day being a Friday, Sept. 2 (coincidentally, the same day Samsung experienced its fastest load time), in which the site took 10 seconds to load at times (due to a slow page file error). However, on average, the site’s homepage took around 1.3 seconds to load. It’s a hair faster than Samsung’s, but they’re close to each other. (Apple Score 9/10)
One major mistake a lot of websites make is utilizing large graphic file sizes or third party code on their home page, and it’s things like that that can really bog down a website’s speed. It’s not surprising that both Apple and Samsung avoid this mistake. While both of them display large, beautiful images on their front page, they optimize their file sizes well.
When we looked at Samsung.com’s performance at various locations around the United States, we found that the site consistently took longer to load in Texas, with its slowest time occurring in Washington, DC, but was the fastest in Florida, North Carolina and Georgia. Samsung.com had just a handful of minor site hiccups during this three-week period, but only at specific locations. For example, AlertBot registered 5 instances of slower load times: once in New York, twice in Florida, once in Washington DC and once in Washington state. Still, it managed to perform more than adequately at these locations overall. It wouldn’t be uncommon for websites to experience significant trouble in certain areas of the country on a regular basis, but we expect only the best from Samsung. (Samsung Score 9/10)
For usability, we select a common task that a typical user might want to perform on sites like these. Then, using hands-on testing, we perform the same task on each website while timing how long it takes to complete and how many mouse clicks it takes to get the job done. This time, we decided to approach each site with the intention of purchasing their latest phone. We timed how long it would take from the point of entering the URL into the browser on through to getting the phone into the online shopping cart.
From the point of typing in “Apple.com” and clicking through their site from the phone product pages all the way to the shopping cart, it took 45 seconds (and 7 clicks of the mouse) for us to add a SIM-free 256GB “jet black” iPhone 7 to the online “shopping bag.” (There’s an additional click, however, to view the cart when you’re done adding the phone to it.)
From typing “Samsung.com” into our browser and clicking through to add a Samsung Galaxy S7 Edge 32GB “unlocked” phone into our shopping cart and viewing the virtual bag, it took a shocking 1 minute and 30 seconds (in 5 mouse clicks)! We used Google Chrome as our browser for both websites and the Samsung site froze up twice during the process (in fact, we accidentally added TWO of the same phone to our cart because we were trying to click through to the cart and it was unresponsive). Just to be fair, we tried it again, and it hung up yet again during the ordering process, but this time it was a little under a minute to get to the shopping bag. All of this happened on Chrome’s latest version, too. We know web browsers can be super fickle, though, so we decided to try it a third time, this time with Mozilla Firefox, and it took 20 seconds to get the same phone into the shopping cart. On Apple’s site, for the iPhone, there are a lot more choices – from storage space to phone color – to choose from, so it makes sense as to why that process might take longer. But it is rather alarming that Samsung’s site experienced THAT much trouble while just trying to add their phone to the shopping cart.
Just to compare via Firefox, then, we re-performed the timed test for Apple.com. One could argue that re-tests don’t account for newfound familiarity with either site, but it took 25 second to add the same iPhone 7 to the shopping cart. While that’s a few seconds slower than Samsung, we also didn’t experience any problems on either browser with Apple’s site.
All things considered, here are the Usability scores:
(Samsung Score 7/10) (Apple Score 9/10)
The performance of both sites were very, very good and quite close to one another. Apple’s site just barely edged out Samsung’s on speed and geographic performance, while both matched each other on reliability. Despite their slight differences, they both performed at the top of their game in online performance. However, after factoring in our usability testing, where Apple’s site performed much more consistently, the winner for the very first AlertBot Showdown is clear: