A few years ago, there was lots of chatter about the elusive UX unicorn; a mythical person capable of delivering everything from research to design to development. It became an obsession for the industry, sparking debate about whether this was the metaphor for how unreasonable our expectations of designers had become, while some felt it was what all designers should be aspiring to.
During the IBM Smarter Analytics conference in Sydney this month, four of its key session speakers –Australia Post manager of business systems, Armand Mizan; IBM’s senior consultant for advanced analytics and optimisation, Tim Higgins; Telstra’s director of one-to-one marketing, Nick Adams; and American Express’ manager of marketing development for Global Corporate Payments, Simon Taranto – closed off the day by outlining their top tips for improving your approach to data analytics.
What shined through was the business intelligence behind the technology. While there are a range of technology solutions available giving marketers the ability to process vast amounts of customer and market data, any attempts to undertake them must be done with a clear alignment to business value.
After all, there’s no point in creating a model to analyse customer behaviour without first identifying the parameters and pain points you’re trying to overcome. This includes acknowledging the business concerns and opportunities the rest of the company and executive team is focused on.
“You have to get the buy-in and sponsorship of c-level management. It’s essential,” Mizan said. “This ensures all analytics projects align to the strategic objective. Whatever initiative you embark upon, understand the business requirements and pain points being experienced. If you misunderstand those requirements, you won’t get great take-up.”
For Higgins, companies that employ analytics as a core function to answering questions within their organisation will do well. “What I’d also recommend is that your analytics environment is flexible and responsive to suit the timeframe of the business executives,” he said.
The panel at IBM Smarter Analytics in Sydney (from left): Australia Post's Armand Mizan, IBM's Tim Higgins, Telstra's Nick Adams and AMEX's Simon Taranto
Analytics people are not always great at articulating the value of their insights to the rest of the business, which is another hurdle all leaders need to be aware of, Adams continued. “You need to build a new brand of people on your team who can work with the other business executives to show your analytics projects as powerful business experiences,” he said.
There must also be partnership between the analytical and creative marketers, Taranto said. “The analytics model needs to be governed and aligned with your go-to market plans if it’s going to work,” he added.
“You also need to modify expectations should the analytics suggest something different to what you anticipated. Your analytics initiatives should also be positioned in terms of incremental knowledge, and not replicate what’s being done elsewhere. The low hanging fruit is probably already being dealt with.”
The panel were also asked their views on how a company can get started on the smarter analytics path. For Mizan, ownership plays a significant role. “You must take ownership of hierarchies, the quality of the data, testing, communication, accountability and so on because without ownership it’s hard to get a model that meets all the expectations,” he said.
Earning the respect of the creative marketing team is worth paying attention to, Adams said. In a bid to overcome the scepticism of analytics, Telstra has brought in mathematics tutors to tutor its 90-strong team in analytics, block models and solving problems statistically.
Taranto meanwhile, said the best way to get started is to pick one problem, get it to a point of maturity and go to market. “Once you have proven the analytics work well, you’ll see the requests coming through,” he said.
And don’t forget having access to appropriate data in the first place is key, Higgins said. “Eighty per cent of the job is getting the data into shape,” he added.