Computers and artificial intelligence have come along at an exponential rate over the past few decades, from being regarded as oversized adding machines to the point where they have played integral roles in some legitimately creative endeavours.
While you won’t find many marketers today disputing the need to embrace customer-oriented, data-driven marketing, the shift has raised on gigantic question: How do you know where to prioritise your data and customers efforts, and indeed, what areas you can feasibly improve digital interactions?
At the recent Adobe Digital Marketing Symposium in Sydney, three brands shared examples of specific data analytics and optimisation initiatives to improve digital exchanges with their customers, and what sort of success they had as a result.
Bankwest: Testing for improved customer relations
Raphael Nottebohm, senior manager digital performance and insights at Bankwest, said the banking group tasked a team of two digital analytics staff 18 months ago with identifying areas where customers were potentially struggling to engage with its products and services online, and to help improve those experiences. The group uses both Adobe Analytics and Adobe Target technology platforms.
The initiative waded through a mix of information including browsing and voice of the customer data to find problems as well as potential solutions. Nottebohm said the first optimisation question was: Can everyone buy from us?
“We asked the question at Bankwest and the answer was no, not everyone could buy from this,” he explained. An example was from its product page, where customers could fill in an application form online or via mobile devices, and had started the process but didn’t complete it.
The data team found that there was just 6 per cent conversion on Google Chrome and iPad compared to 64 per cent conversation rate for other browser and device combinations. The abnormality was indicative of a problem with customer experiences, Nottebohm said, and was duly fixed.
As a second example, he highlighted work done using Web analytics around an online credit card application form. While it took 18 minutes on average for a potential customer to fill in the form, 19 per cent were abandoning their efforts on the last page.
The question to ask was whether that was abnormal or business as usual, Nottebohm said. To do so, the Bankwest team needed combine the ‘what’ with the ‘why’.
A good way to approach it is to go big – don’t be shy. You need quick wins, and you need to persuade upper management that you’re doing the job
“We decided to run a survey to ask why they were leaving and found 70 per cent said that they thought the process was completed,” he said. “We updated the UI, changed the copy, and now we’ve already achieved 60 per cent of our 2014/2015 financial year targets growth within three days of doing that work.”
Nottebohm’s third example of how data analytics had helped improve customer experiences was around testing landing pages.
“It’s important to test but it’s not always easy to know what to test first,” he commented. “A good way to approach it is to go big – don’t be shy. You need quick wins, and you need to persuade upper management that you’re doing the job.”
In Bankwest’s case, the analytics team asked marketing if it could eliminate the hero image from a landing page test for one of the bank’s savings campaigns. The question it was striving to answer was: Does Garry the squirrel add value to the customers’ experience?
The results showed the page without Garry the squirrel outperformed the control page by 14 per cent, Nottebohm said.
“Running the test is easy; it’s asking the question and engaging with marketing to challenge existing practices that is more difficult,” he added.
Telstra: Helping customers move home
Working as a blended team with 2DataFish, Telstra has built an SEO practice using several Adobe tools including Adobe Analytics and Audience Manager. The program of work is driving by a continuous improvement methodology and is aimed at improving Telstra’s digital sales.
To illustrate the success of the work done to date, 2DataFish co-founder and director, Max Boucoiran, discussed several optimisation tests that triggered a double-digit uplift in conversion on the ‘moving home’ section of Telstra’s website. The section is about helping customer transition services when they are relocating.
The first initiative was around the UX team refreshing the section and an opportunity to test different layouts and content length.
“We engaged many stakeholders and came up with the agreement to test four different designs. Tested these experiences side by side over 30 days and let customers decide which they preferred,” Boucoiran explained. The winner was a shorter content format.
The second test was about “putting the magnifying glass” over the winning layout in terms of click through, he continued, looking at icons, button colours and link styles. This trial ran over a month and took about three months to set up and get necessary approvals.
The third test was looking at customer journey flow pages and doing some analysis on where problems of drop off was occurring. Boucoiran said the team noted a step where consumers were opting out in droves. They eliminated that step entirely and tested the new user flow with a simple A/B test, he said.
“Succeed breeds success and that really applies here. The first test took six months to get off the ground, but once we hit the first result, where we saw a near 10 per cent uplift. The second one then only took three months to get off the ground,” Boucoiran said.
“It’s a lot easier to engage with stakeholders once we prove the worth and value of our optimisation capabilities. Only a month after the second [test], we launched the third campaign.”
Boucoiran’s main takeaway was that there is value in continuous improvements when it comes to data-driven digital optimisation.
“As you get comfortable with one section, it’s much easier to engage with all the stakeholders, and to get much more comfortable with the process,” he added. “We reduced lead times but saw uplift in customer conversions.”
Seek: Increasing registrations
At Seek, recent analytics work on improving customer outcomes has taken place around increasing registrations to help improve user profiles, its head of digital analytics, Jed Elliott, told attendees. To do this, his team had a series of hypothesis around how to present information, the decisions consumers needed to make during registering and what it could make mandatory, facilitating privacy settings, the cost compared to alternatives, and initial registrations. The business uses Adobe Test and Target and Analytics suites.
Elliott explained the group had four options with different combinations – mandatory or optional – as well as choosing privacy settings from a dropdown list versus default options.
“Having mandatory questions showed a minor uplift, but not enough to warrant forcing people down that path… whereas the transparent, optional approach experienced a good uplift,” he said. “With regards to privacy settings, both gave us good results, one slightly higher than the other so we stuck with giving consumers the options.”
Seek segmented the results across all different industry segments before rolling out the changes and then going through a process of multiple variations across the site and 15 access points. Elliott said the first two waves of changes showed a 46 per cent lift in profile conversion rates, a key KPI Seek uses as a business. As a result, the company is now rolling across the apps and going out to market.
Elliott’s key takeaway for other marketers was to make sure you share the numbers with the right business executives.
“You must show the ROI on your tests to stakeholders to sustain momentum,” he said.
Follow CMO on Twitter: @CMOAustralia, take part in the CMO Australia conversation on LinkedIn: CMO Australia, join us on Facebook: https://www.facebook.com/CMOAustralia, or check us out on Google+: google.com/+CmoAu