Bracknell Forest Digital Services


Leave a comment

Reviewing the (statistical) situation

It’s been over 7 months since we last posted on here, announcing the launch of our website. No excuses, but it’s just been one of those periods where there hasn’t been the time to sit and reflect and blog.

In brief, we’ve been moving from project to project, many of which were time sensitive and demanding. In addition, our team has been changing, with members leaving and devolved editors being merged into our centralised team. It’s been a time of change for the council as a whole as well and our team has been working to keep up.

That said, I’ve finally had a chance to sit down today and look at the stats from our site over the first 6 months. Whilst we’ve been monitoring the site stats, it was the first time I’ve put all the data together to get an idea of the bigger picture. I thought I would share some of that data with you.

Note: Technically we launched in mid-June, but for the ease of reporting, I’m working with data from 1 July 2017 to 31 December 2017.


Sessions

sessions.PNG

In 6 months, we had 1,036,510 sessions logged on our site. This is compared to 880,360 sessions in the same period last year on our old site.

Users

users.PNG

We had 548,475 users of our site during the 6 month period. This compares to 574,506 users in the same period last year.

This is a slight decrease in users, some of which is accounted for in changes to the way we monitor our stats as for our site as we are now blocking internal traffic from being reported.

Pageviews

pageviews.PNG

In terms of pageviews, we hit over 3 million hits in the half year. This compares to 2,388,654 for the same period last year.

Some of the increase in pageviews is likely accounted for by our move to guide pages. We are now splitting longer and more complicated content into multiple pages to help the customer journey. This means that what was once a single page, a user might need to work through 4 or 5 smaller pages on their journey.

Traffic sources

Traffic channels.PNG

We often tell our colleagues that most people use a search engine to find our content, rather than navigate from the home page. With over 70% of traffic coming from organic searches, our new site is certainly proving this.

Even with direct traffic, we see that most customers go to a page they know. Only 12% of our direct traffic is going to the homepage first. This really emphasises that the homepage isn’t necessarily the most important page on the site. This is made clearer still when you realise that our homepage, whilst the second most popular page on the site, is still getting less than 5% of pageviews overall!

Site search

search usage.PNG

Whilst we really like using Drupal for our new site, we have had some implementation issues with our search, particularly relating to content within guide pages. This has caused a number of complaints from customers and staff looking for content. We’re looking into resolving these issues as soon as we can, reviewing how we use guide pages to present content.

That said, based on our search report, only 3% of sessions used our site search in the 6 months from July 2017. So whilst it’s key we get our search functioning properly, it’s clear from the stats that most customers go directly to the content they need, generally using an external search.

Devices

devices.PNG

With a responsive site, it’s great to see that mobile traffic continues to outperform desktop usage. We’ve put a lot of work into getting our site mobile friendly and as these numbers continue to increase, it’s ever important to make sure future developments are mobile friendly.

(For those of you who want to know what devices our user’s love, that’s fairly clear: 45.28% of all mobile traffic was from an Apple product!)

Browsers

browser.PNG

Based on the prevalent use of Apple devices, it’s no surprise that Safari tops our browser list with 40.25% of sessions. In addition, Internet Explorer has significantly decreased in usage, although Edge has yet to make the top 5!

Top pages

With over 3 million pageviews across some 2000 pages, our site is well used. However, as you can see from the top 10 below, Coral Reef Waterworld significantly skews our pageview statistics.

top ten pages.PNG

Within the next couple of months, these stats will be radically different. Coral Reef, Bracknell Leisure Centre and Downshire Golf Complex are being outsourced and so 7 of our top 10 pages will be removed from the site.

This is obviously going to have a big impact on our pageviews (and probably sessions in general). In fact, I think we can safely say that our stats will decrease significantly and our next 6 month review will likely tell a different story!


If you’ve made it this far, well done! It might be time for a tea break!

We will hopefully be back before long with more updates for you as to what we’ve been up to. In the meantime, thanks for reading and have a great weekend.

Advertisements


Leave a comment

Publishing website statistics

Last month Public Service Digital published an article around the data that council digital teams share and how, this being 2016, we really ought to be making better use of the tools available for doing this. Further, the article raised the idea that perhaps there ought to be some form of standards around the publication of the data in order to make it comparable.

This raised some discussion in our team around what we are currently doing, and how we could improve the provision of data for our public website.

Like many councils, we use Google Analytics to monitor our website performance. We already publish some basic statistics on a monthly basis, but it is limited.

GOV.UK have some really interesting presentation of their website statistics, and whilst we are not there yet, it is something to aim for in terms of data publication.

Instead, we have opted for something quicker and simpler to get us moving in the right direction. We have used the Google Analytics Spreadsheet add-on to put together a new website statistics page.

This page currently shows some of our key website statistics in the shape of interactive graphs, like the one below.

 

We’ve added what we think are some of the key metrics for looking at website performance, but we would be interested to know what others publish, and whether providing more information would be useful. It would be great to have a discussion around what data councils should be publishing in this area and whether there ought to be some form of shared standards for publication so making comparisons becomes easier.

Finally, thanks has to go to Croydon council’s web team as their website statistics page and spreadsheet really helped us understand what we needed to make this work. Using this, plus the tutorial videos available from Google, we managed to put this page together quite quickly. The whole process is simple and these videos are a great starting point if you want to make something similar out of your statistics.

Over the coming months we will look to improve on this initial offering and share further data for our website performance. Is there any particular data you think we should share on this page? Let us know in the comments!

Orange and blue lines on a graphy


Leave a comment

Home sweet home? – a look at homepage usage

Last week Brighton published a great blog post about their homepage, and how really, Google is their homepage.

This prompted me to take a look at our homepage to see how it performs in relation to the rest of our website. What we found out was pretty interesting stuff (if you like this kind of thing anyway!)

Looking at a standard content report for our website shows that our homepage was the top ranked page in March.

Table showing the top page views in March

However, this presents a slightly misleading interpretation of homepage usage as when we look at homepage visits in comparison to all other visits, we can see that actually, over a year, 82% of sessions on our website did not visit the homepage at all.

Graph showing sessions with and without homepage visits

Where is home?

Approximately 15% of sessions start on our homepage. What about the other 75%? The table below shows the top ten pages that customers started on for March.

Top ten landing pages (excluding homepage)

Unsurprisingly, given it’s popularity, and the current work going on, Coral Reef is the most popular landing page, followed by The Look Out Discovery Centre.

All of these pages are some of our most popular pages on our site so it is not a shock that these pages would be entry points. However, this doesn’t tell us how customers are getting to these pages.

Do they Google it? Do they have a direct link? What about other sources?

According the the numbers, Google is the most popular starting point, and direct links follow.

User flow diagram

Even more interesting, and something we should probably explore at a later date, is that when we add the homepage as a landing page back into the figures, we see that large numbers of customers are actually still coming to it via Google rather than by other means.

Diagram showing usage of homepage as a landing page

Using the homepage

Let’s now take a look at those sessions where there is a visit to the homepage after looking at other pages. Why are the majority of our customers not doing this, but some are?

To help us understand, we can look at where customers were before they ended up on the homepage. The following table shows the top five pages that people navigated from.

Pages used before homepage

What is it about these pages that mean customers decide to head to the homepage after using them?

Is it because they do not find the information they want, or is it because they did find that information but wanted to do something else?

These pages are generally well used, but perhaps they are failing customers in some way that results in the backtrack to the homepage. Some focused user testing may be in order to see if we can work out why this is happening.

Another key indicator we can look at in relation to homepage usage is time spent on the website.

Graph showing average session duration

In March, the average time spent on the website when the customer did not go to the homepage was 1 minute 46 seconds.

If the session included the homepage this jumped to 3 minutes 41 seconds, suggesting that those customers who go to the homepage navigate around a lot more before they find what they want. They may also be interested in more than one thing.

Again, we need to be monitoring how our customers navigate much more closely in order to understand this.

What next?

What should we take away from all this? How can we make our new homepage more useful for those customers who have ended up on it? I think there are two key things we need to consider:

  1. We shouldn’t over engineer the homepage. It doesn’t need to be fancy but it does need to provide customers with a simple way to get to the information they need. Customers should also be able to find what they need quickly.
  2. We must focus on the content and ensure that customers can get to it through the channels they want – for example, let’s make sure that customers can find our content easily through search engines.

We’re already doing some of this for the new website.

We’re trying to keep the homepage as simple as we can, with clear navigation into services. In our first wireframe testing last week, users really liked this, and found it clear and easy to use. Once we have some stats on our new homepage, it will be very interesting to run a comparison.


Leave a comment

3 years and counting

Over the past few months we have done a lot of stock taking on our existing website. This has helped us inform the way we want to move forward with the re-development.

It’s also a good way of benchmarking for the future. Our current site is over three years old. In this time, we have gathered a large quantity of statistics about usage of the site. We can see what people come to our site for, as well as what type of information we present to them when they do – from content pages to PDF documents.

By looking at where we are now, we hope to get a good idea of how our re-developed site has improved the customer experience.

We’ve put together an infographic to highlight the most interesting (to us, at least) key statistics from the last three years. If infographs are not your thing, the statistics are also available below with a little explanation of the figures.

Key statistics from the public website since April 2012

Content growth

The figures around content show that the site has grown, despite attempts to keep page numbers down. This has sometimes led to unnecessarily long pages, something we are going to be looking at during the re-development.

Year Number of pages
2012 996
2013 1208
2014 1191
2015 1244

Likewise, the number of PDF documents on our site has also grown, getting up to nearly 3000 at some points. We have managed to bring this back down through careful assessment of the need for a document, but we still have a long way to go.

Year Number of documents
2012 1874
2013 2840
2014 1197
2015 2170

Overall key figures

Page views 17,752,383
Visitors 2,841,258
Visits 5,898,175

Yearly key figures

These figures show us that in terms of visits and visitors, we have had a steady increase in both, reflecting the general trend towards online services. Similarly, the number of page views has also grown over the years.

2012-13  2013-14  2014-15
Visits  1,581,131  2,172,729  2,354,903
Visitors 793,654  1,097,603  1,227,738
Page views  5,097,905  6,144,779  7,028,221

For a more detailed breakdown of our statistics, we update our website statistics page monthly.

We’ll also be back over the coming weeks to take a closer look at what these figures have shown us.