How Northern Parrots put our 20% guarantee to the test

Applied Web Analytics

The UK’s only conversion agency specialising
in multichannel retail. We guarantee to
uplift your sales by 20% or your money back.

You are here: Blog > Case study > How Northern Parrots put our 20% guarantee to the test

How Northern Parrots put our 20% guarantee to the test

Six months ago, Northern Parrots thought they had done everything possible to improve their website. As the leading UK retailer of food, toys and accessories for parrots with a seven digit annual turnover, they were already getting outstanding results thanks to e-commerce best practice and expert recommendations

They had even been told by a number of experts that the homepage was so good it couldn’t be bettered. Could Conversion Rate Optimisation really make that much difference?

The AWA digital CRO project guarantee logoBut then Northern Parrots heard about AWA digital’s guarantee to lift web sales by 20% in six months. Initial scepticism quickly transformed into temptation as they realised they had nothing to lose. Either sales would increase by 20% or they would get their money back. It really was a win-win situation.

34.7% sales increase in just 6 months

Six months later, and Northern Parrots are looking at sales increases of 34.7% from just one split test. Remarkable considering their conversion rate was already well above industry average. This outstanding result, and the process used to achieve it, are the reason this test scooped Gold in the Best Home Page category of the 2014 WhichTestWon awards.

Read on to find out how the AWA Conversion System™ was used to gain this uplift and smash their end of year sales target.

The key to achieving double digit results is talking to and understanding your customers. All our tests show that this is vital to make meaningful improvements. Analytics data, best practice and expert reviews can only take you so far.

How we got to the heart of their customers’ desires – Insight Generation

First we installed a range of optimisation tools, including Qualaroo, Crazy Egg and Optimizely, onto the Northern Parrots website. Using a combination of insight generation tools and techniques, our team quickly and easily got to the heart of the issues affecting Northern Parrots’ website visitors. The team found that:

  • There were very few clicks on the main area of the homepage, despite this being where they majority of the promotions were. The majority of clicks were in fact on the side navigation which listed product categories. This heat and scroll map data from over 1,700 visits was the first indication of the browsing behavior of Northern Parrot’s customers.
Screengrab of Crazy Egg Heatmap for Northern Parrots homepage

Screengrab of Crazy Egg Heatmap

  • Customers wanted to discover new, tempting items rather than search for specific items but the side navigation didn’t meet these browsing needs. Visitors wanted it to be clearer and more suited to their habits. These findings from user testing with five Northern Parrots’ customers provided a new perspective on the heat map data and shows the importance of understanding your customers’ needs, not just their clicking behaviours.
  • Visitors’ were happy to share their fears, concerns and pain points in user surveys. Respondents also told us what they loved about the website and why they chose this one over the competition.
  • There was no clearly identified value proposition explaining why a prospect should buy from Northern Parrots as opposed to a competitor. Using the responses from the user survey, we were able to clearly identify the main reasons Northern Parrots’ customers were buying and use these to convey the value proposition in our design variation.

How we chose the first split test – Triage™

After gaining insight from Northern Parrots’ customers and visitors, the next stage was to turn this wealth of data into a clear optimisation plan. To do this, we recorded each unique insight, the impact and the quantity and type of evidence supporting it. This unique Triage™ analysis enabled us to easily identify the best opportunities on the website and create a prioritized optimisation plan.

This is a great example of the importance of gathering voice-of-customer data in your CRO project. Traditional conversion funnel analysis and recommendations from experts had led the client to believe that the homepage was performing well and there would be little opportunity to deliver such bold uplifts on this page. However, our voice-of-customer evidence suggested otherwise. Our team could see that this was definitely the page to focus on for his first split test and it was the first test proposed in the optimisation plan.

How we developed the split test – the Creative Execution process

As a result of the insight into Northern Parrots’ customers, the hypotheses for this split test were simple:

  1. Presenting browsers with a clear path to products, relevant to them, will increase conversion rate
  2. A clear value proposition will increase conversion rate as a trust is built and visitors understand why they should buy from Northern Parrots
  3. Aligning the main navigation with popular and profitable conversion paths will result in higher revenue per visitor

To test these hypotheses, the team created a wireframe of the page we believed would better meet visitors’ needs and increase sales. De-cluttering the design, re-writing the key content on the page and catering for the browsers were all high on the agenda. Another innovation was incorporating an endorsement from a vet because user feedback showed was the main reason customers chose this site over the competition.

Our optimisation team then worked collaboratively with our copywriting and creative teams to refine the wireframe and content. The optimisation team went back to the user testers they had previously worked with and got their feedback on the wireframe. Their feedback was fed into the final design which was tested. This collaborative approach to creative execution together with the deep-dive research meant we were confident that our variation had the best chance of delivering double-digit increases.

The split test designs:

The original page:

The original Northern Parrots homepage

Our variation:

Our variation of the Northern Parrots homepage

What we found – the split test results

The experiment ran for a total of three weeks. During this time our variation consistently outperformed the control. The final results of the experiment were a site-wide increase in revenue per visitor of 21.7% at a 95.1% confidence level. That meant our 20% uplift guarantee had been conquered in the first split test alone.

But the really interesting bit was when mobile traffic was excluded. For these non-mobile users (who make up the majority of Northern Parrot’s visitors), the increase in revenue per visitor rose to 34.7% at 96.3% confidence level.

Screengrab of the split test results from Optimizely

Screengrab of Optimizely results

This was an interesting finding in itself. Why were mobile users less responsive? We developed two hypotheses for this.

  • Unlike non-mobile visitors the majority of Northern Parrots’ mobile users are ‘buyers’, not ‘browsers’.
  • The design changes were too complex for the smaller screen size

In order to test these, we have updated the optimisation plan with a mobile-only homepage test.

Northern Parrots were, unsurprisingly, thrilled by the result and are in the process of implementing the winning design to their desktop and tablet sites in order to take advantage of this proven uplift in sales. Mike Taylor, MD said “although we were initially sceptical that AWA digital would be able to increase our sales by 20% or more, the results speak for themselves. Now they’ve proved themselves, we look forward to achieving even better results with them in the future.”

Always looking for the next big win, our optimisation team are already working on beating this result with their next test on the category pages of the Northern Parrots website. We also plan to re-visit this homepage test once it has been developed to see what additional uplift potential there is.

Gold Award winning

As well as smashing the AWA guarantee, this test also beat off 136 other entrants from all over the world by scooping Gold for Best Home Page in the WhichTestWon 2014 Online Testing Awards.

Announcing the win, the judges commented on how impressed they were that:

  • such a huge uplift was achieved from a top-funnel page
  • the design successfully balances a strong message with navigation and product images
  • the problems were identified and addressed using a mix of data sources, including user testing, user surveys, heat maps, scroll maps and analytics

Put us to the test

Are you ready to increase your e-commerce sales by 20%? Looking for a risk-free approach to conversion rate optimisation? Then put us to the test! Get in touch today to discuss how AWA digital, the UK’s only CRO agency specialising in multichannel e-commerce, can make these aspirations a reality.

 

Not sure what a 20% increase in sales would really mean to your business? Use our conversion calculator to find out now. Want to see more case studies? Read our client testimonials page to see what other multichannel retailers say about working with us. Need more convincing? Find out more about how we guarantee to increase your sales by 20%. Watch this two minute video.

Posted in Case study

2 Comments

  • February 3, 2014 at 1:54 pm Julian Erbsloeh says:

    Hi guys,

    this is really interesting insight and a fantastic result – well done and thank you for sharing. I do have a question about the time frame you chose for the experiment – the result screenshot shows that the experiment was run for only 10 days during the arguably most ‘unusual’ time of the year: the run up to Christmas into the new year.

    How can you be sure that this uplift is sustainable in the long run after taking just such a short sample during such a volatile time of the year, do you keep a ‘stub’ running to verify the uplift over time and across seasonality peaks and troughs?

    Best regards,
    Julian

    Reply
    • February 5, 2014 at 5:17 pm Dan says:

      Hi Julian.

      Great question.

      The experiment ran for a total of three weeks over two different periods. The screenshot shows the results of the second phase, which includes the Christmas period. Results were consistent throughout both phases of the experiment.

      It’s also worth stating that this business does not experience the seasonal volatility often observed elsewhere, which you refer to.

      I hope this answers your questions

Leave a comment

We would love to hear your comments. To leave a comment complete all sections of the form below, your email address will not be published.

Comment Spam Blocking by WP-SpamShield

Receive "new post" notifications without leaving a comment