Website Performance Survey Conclusions

Content is King… but it still needs to be delivered correctly if it wants to rule the Kingdom.

How a sample of UK medical/device websites perform when assessed against critical criteria.

Introduction

Anyone who’s embarked on the development of a new or upgraded website will probably have encountered the “techie”who does the actual site build. And he’ll no doubt have muttered something along the lines that the site will only be as good as its content. And of course, for those of us who want our sites to be some gloriously interactive sell-piece, online brochure and a window into our products for a hungry audience, we understand that perfectly well. Using the old analogy, the site builder builds the rooms and you decorate them.

These days though, it’s also about helping your audience to find their way to your newly decorated show home, and provide a sufficiently alluring experience that they hang around long enough for you to retain their custom.

Online, it’s also about Search Engine Optimisation (SEO), an acronym that conjures up images of a million pop-up ads for companies promising to get you to the top of the rankings. It also seems to furrow the brow of expert and non-expert alike, the rules to get you up the rankings being somewhat cryptic and rather dynamic. The tales of people ramming their pages with keywords in the hope of leaping up search lists are now ancient history as it’s all got rather more sophisticated. Those Google folk aren’t stupid and can see straight through the old ruses, but at heart they need to ensure that the most appropriate sites find their way to the top of the heap. This means the more your site ticks the boxes that a good site should tick, the better your chances.

We’ve written recently about a few of Google’s hot buttons, and frankly it all gets a bit geeky. That said, in essence what is clearly the case is that we can look at a site and audit it for likely appeal (ergo performance) pretty quickly. How quickly does it load on a desktop? How quickly does it load on a mobile? Is it responsive (ie does it change format to adapt to mobile devices or does it just shrink or break up or lose large chunks? Is it secure (ie encrypted… look for the padlock)?… a sign that you’re looking after your website visitors and again favoured by Google.

Then there’s what it looks like and how easy is it to use? You could say these are intangibles, but to us, very experienced in medical devices as we are, our assessment is that old-fashioned sites looking really dated as design cues and website fashions have evolved. Too many dark colours, unreadable fonts, giant graphics and clunky menus are major no no’s.

Choice of Content Management System(CMS), the underpinning for the site, is significant, many companies making the decision to use WordPress. This system offers so much to the developer and hosting experience that it’s a must in the book of most developers today.

So how does the Medical Device industry fare? We took the view that surveying a fixed number of randomly selected medical companies and assessing them using criteria that were as objective as possible, might yield some interesting snapshot that in itself might just characterise the industry. If the pilot study does its job, it might just provide a leg-up for UK medtech by highlighting a few addressable gaps.

Methods: Study outline

We conducted a survey of 21 companies, chosen randomly from an industry-body members list, all of which shall remain anonymous for the purposes of this report.

Each company was assessed against a set of criteria that reflected their likely attractiveness to search engines and potential visitors. Those criteria were:

Desktop speed, Mobile Speed, User Experience, Content Management Software(CMS), Responsiveness, Use of SSL, Attractiveness (0-10), Ease of use (0-10).

Desktop speed, Mobile Speed and User Experience were scored using Google site developer tools which operate on a traffic light basis as follows: Green Tick = no significant issues found. Yellow Exclamation Point = consider fixing this if it is not a lot of work. Red Exclamation Point = Fixing this would have a measurable impact on page performance. Threshold scores between these bands are at 65% and 85% respectively.

Choice of CMS was noted where evident.

Responsiveness was assessed by visualising sites on iphone and ipad mini.

Encryption (SSL) was confirmed by use of https:// and the padlock sign.

Attractiveness and Ease of Use were assessed independently by two people, based on a scoring system of 0-10 where 10 represents perfection and 0 represents a totally unattractive or unusable offering. For attractiveness, criteria that were assessed included layout, visual appeal, consistent use of style, uniformity of font and appropriate font size and colour against background. Ease-of-use was more about content and navigation, and included commentary about accessibility and currency of news items as well as internal links and design of contact forms. All recorded scores were an average of two-person scoring.

Notes were made of the most commonly identified negative site features.

Results

The 21 companies assessed encompassed manufacturers, distributors, and dedicated healthcare business consultants. Company size, while not documented, ranged from multinationals to UK-only SMEs.

Table 1: Scores for 21 companies, as assessed using Google’s Developer tool

CompanyDesktop SpeedMobile SpeedUser Experience
186%67%99%
277%56%96%
387%72%60%
483%69%100%
583%63%99%
663%52%99%
772%58%99%
865%57%99%
980%65%67%
1051%47%99%
1159%51%99%
1275%59%67%
1377%56%97%
1428%70%85%
1589%76%64%
1654%35%99%
1775%64%90%
1847%37%64%
1967%64%98%
2051%65%45%
2184%66%65%
  • Average desktop speed was assessed as 69%, which is marginally above the Google threshold, placing the group as a whole in the middle category, where fixes would be desirable. Only 14.3% (3/21) of assessed sites were classed as green ticks with no action required, whereas 33% (7/21) were red, so in need of urgent attention according to Google’s scoring system.
  • Mobile speeds were significantly worse, with no green ticks and 62% (13/21) scored as potentially benefiting from attention in terms of page performance.
  • When it came to formal assessment of user experience, according to Google criteria the scoring was significantly higher, with an average of 85%. Only 3 companies were red, with 15 requiring no attention at all.
  • CMS: 43% (9/21) of the medical device sites assessed used WordPress, clearly the most popular platform, followed by Windows (4/21). The remainder were either undefined(3) or adopted a variety of other solutions(5).
  • Responsiveness, in other words the way in which a site is designed to change in order to adapt to fit different devices, was a feature of 15 of 21 sites, one of which was only partially effective with menus that worked on all devices, but images and paragraphs that were foreshortened or cut as screen size decreased. This means that approaching a third of the random sample of medical device-oriented websites assessed were not adapted to use on mobile phones or tablet devices.
  • Security: Only 3 sites used encryption.
  • Attractiveness: Scores varied from 3.5/10 to 9/10, the most common reasons for low marks including cluttered layout and use of what were considered old-fashioned design concepts. The best sites were clean, free from visual glitches and with well communicated messaging about what the company was about. Sites were marked down for inclusion of “1990s” features such as bouncing text, too much/unnecessary animation, black and heavily coloured/gradated backgrounds.
  • Ease of use: Scoring ranged from 3.5/10 to 9/10 with the best sites featuring simple menus and functionality. The worst sites were considered those with links that didn’t work or that went direct to .pdf files or even (in one case) closed the company site and opened a new proprietary credit card payment page. Buttons that promised “more info” but merely revealed twenty words, were marked down because of the unnecessary link (which incidentally is SEO unfriendly).

Table 2: Common Faults

  • External links not identified as such
  • No news
  • News categories / events categories with nothing or little in them
  • Long sentences,uneven and insufficient line spacing, all leading to poor readability
  • Illegible text because of poor colour / background colour choices

Sites with little (or old) news were similarly downgraded, some featuring unnecessary comments and category buttons prominently on each item, the comments button never having been clicked. Better sites were considered those that used “share” buttons.

Note that unresponsive sites were allocated a maximum score of 7.

Discussion

The nuanced differences between websites are not something most of us consider during the 8 seconds we reputedly take to make up our minds about them. Drop shadows and wrap-around titles are of little consequence to most of us, but those in the know say these things do matter. That’s why we embarked on this brief study, the cdbresults of which were somewhat surprising. For those of us whose formative years were spent in medtech marketing, in an age where the opportunities to put your wares in front of the audience were very limited, the opportunity to engage using the internet looks like, and is like manna from heaven. Why wouldn’t a company make its online offering as brilliant as possible? And why, after this survey are we left thinking that if medtech was getting a school report for its online effort, we’d be saying “could do better.”

Here’s our more detailed assessment: Firstly, website performance is something of a moving target, although the end goal is fairly consistent. Most companies would agree that generating traffic is desirable, with whatever ultimate objective that delivers. That’s what our key performance indicators looked at; the things a website needs to do to keep search engines happy (and therefore, if Google has got it right, customers).

Looking at the selection of companies reviewed, it’s not particularly helpful to generalise, not least because of the diversity of company type offered by this random selection. That said, all the companies have in common the feature that they’re trying to engage in markets in which they need to make more or less technical demonstrations of products or services to healthcare professionals or companies.

Sites ranged from those that looked like they’d been developed in the 1990s and not touched since, through to more impressive contemporary offerings. It’s quite clear that some sites were considered integral to a company’s operation, while others looked like more of a tick box exercise. Also apparent is what we might call the throwback effect, which is sites that load quickly and work well on desktop computers but are dreadfully slow to load on mobile devices and are often not responsive. Delivering a slow, painful page that doesn’t fit and has fonts that are too small to read doesn’t tie in with the eight second attention span. In a mobile world, especially when dealing with healthcare professionals who are hardly ever shackled to a desktop, delivering a well-designed page quickly is essential.

On desktop speed, to see 33% of websites flagging up red, so indicating an urgent need for attention, is surprising. For nearly double this number (62%) to be red flagged for mobile speed is a demonstration that these sites have not been attended to since the world went 50% mobile.

(Note that Google’s own tools were used to assess speed and user experience, meaning these results will correlate with the website’s search rankings – Google will always prioritise a faster, more usable website)

When it came to scoring user experience, only 3/21 were red flagged, perhaps confirming the suspicion that medtech folk are more interested in content than they are about speed and security (remember only 3/21 sites were encrypted).

On then to the parameters that required us to turn our own subjective opinions into a score. Again here it was apparent that a number of sites were undergoing something of a makeover. While we didn’t formally assess companies across different domains, we noticed that it was not uncommon to find those whose internet presence in one territory looked poor, while the experience in another was significantly better. So we can let these companies off the hook perhaps, but the pervading sense through a significant number of sites is that they were either afterthoughts or had been developed using the lethal formula of a human being with little experience of how to get the best from a website. Some looked like online catalogues, which is clearly what they were intended to be, missing the point that there’s so much more potential to engage with your audience than flicking through the pages of a catalogue with them.

It was also noteworthy that the sites that scored low on attractiveness also scored low on usability. That’s not likely to be because attractiveness derives usability incidentally; it’s a reflection of the lack of care or experience in one area carrying across into another.

Conclusions

Here we sit then, fascinated by the output from a survey that we performed on a random set of representative company websites. One could argue about our impartiality on the basis that we’d love to help people get their offering right, but the key statistics that sprang out from our use of Google’s PageSpeed Rank developer tool are undeniable. Speeding things up (by using and managing the right hosting solutions) and making them work on mobile devices, plus paying some attention to the way content is delivered, is not very difficult to achieve, yet could pay dividends.

Survey Results
See the InfoGraphic derived from this survey

The fact that 43% of the sites were created using WordPress rather suggests that either the company or the one it paid to develop its site, buys into the idea that this is important. In the authors’ experience WordPress is an essential component in delivering a website that combines functional effectiveness with flexibility of design and layout. Furthermore, when it comes to keeping your new pages up to date, its interface means that’s an easy task.

So whether we’re reporting here on an opportunity missed or a fundamental problem with the UK medtech industry is less relevant than the fact that it could be done much better if placed in experienced hands. Website development should not be an “off-the-shelf” affair if the desired result is for a company’s online presence to play its part in delivering for the business.

The bottom line here then is that If you want a website to deliver good search engine performance, and a favourable visitor experience, it needs nurturing like a baby. Unless your website scores all green on Google’s PageSpeed Rank tool, uses SSL and is hosted on a server that uses HTTP/2 and flies onto screens, be they big or small, connected to the internet via the data network or fibre optics, you need to take action.

Posted on: Posted on: Posted in: Categories