How Does It Work?
Simple! Each conference delegate participates in five discussions – two on day one and three on day two of the conference. Each discussion lasts for 90 minutes.
There are 55 discussions to pick from, split into five time slots.
Following registration you will be directed to the discussion selection page. You complete your selections any time following registration but we recommend doing it sooner rather than later as places are limited for each discussions. Some discussions will be booked out early.
For each slot pick your first, second and third choices. In over 83% of cases delegates get their first choice. You will get your itinerary upon your arrival at the conference.
The hardest part is picking your top choices from a very tempting discussions list.
You will find below the discussions list split by the time slot in which each will run.
Discussions Day 1 - June 8, 2016
The growth of experiment-based methods for measuring, optimising and innovating with digital products has driven an explosion in managers wanting to go big on velocity and scale. Few companies are successful in getting the throughput or quality from these programmes to justify the costs involved. Some even give up on experimentation and feedback altogether!
What are the things that allow us to scale A/B testing across silos, departments, divisions and products whilst making it responsive to the needs of multiple stakeholders? What are the enemies of velocity and quality of learning in testing programmes?
Craig Sullivan has tried, failed and iterated testing at many companies. Whilst discovering that success is rarely down to tools – it is a human, organisational, process and methodology driven problem (at least when it is not working). Identifying the easy and big wins from many optimisation opportunities, hypothesis development, idea generation and testing prioritisation is a big challenge for companies. Trying to orchestrate this at scale requires new approaches to building, prioritising, iterating and analysing experiments.
This discussion will prove invaluable if scaling up your optimisation programme is a key objective for your organisation.
If your career is not in transition, it should be. This does not mean you need to change jobs but without growth and change - we stagnate. And then it is only a matter of time before change will be brought upon us. Why not proactively seek it out?
This discussion, led by David McBride of Intel, will focus on topics like detecting early-warning signs that it is time to re-engage, deciding if you need to make a career change, finding the right way to seek additional responsibility, seeking opportunities at your current company, cultivating a network and starting out in a new role on the right foot.
Note: Attending this discussion does not mean you are looking to change companies or that you do not like your job. All who seek to optimise their work experience are welcome and encouraged to attend.
Organisations have woken up to the power of data and how they can exploit data for a variety of business benefits. As a result they are demanding more from their analytics teams. They want analytics to be distributed, better embedded into business functions and better integrated. As more businesses place data at the heart of their competitive advantage they are seeking out more sophisticated ways to exploit data for that advantage.
Meanwhile the transition to advanced analytics poses a significant challenge for many teams. Finding the right skills is a long and often fruitless process, reshaping and ramping up a new team comes with unsustainable costs. Integrating the complexities of data science into the business can sometimes end in deadlock. So what should one do?
In this discussion Dwayne Browne of Accenture will explore ways to ease this transition. Questions covered will include:
- How do we go about reshaping our teams?
- Is your business really ready for data science? Do the right operating models and data governance policies exist to support uninhibited data exploration?
- Should we retrain in-house, use strategic partners to flex up in the short term or go all out to employ fresh talent?
- How do we move from ad hoc projects to data science at scale?
This promises to be an exciting discussion that should help you chart your next steps, whether an analyst or a manager, in your analytics journey.
For decades companies have used consumer segmentation to drive their products to market, identify consumer needs, and tailor marketing messages. Now big data empowers companies to go beyond traditional segmentation models into profiling individual consumers, understanding their preferences, interests, purchasing intents, and physical location. In-depth consumer profiling allows companies to target their marketing messages, personalise digital channels, recommend products, target 3rd party advertising, and produce better products.
In this discussion, we will exchange views and experiences regarding various use cases and opportunities for consumer profiling and their prerequisites such as data science, infrastructure, data sources, and consumer privacy. We will discuss how to build the bridge between digital consumer profiling and traditional segmentation in a company that operates in multiple channels.
The discussion will be led by Ulla Kruhse-Lehtonen whose analytics team at Sanoma, a Finnish-Dutch media company, has developed and commercialised consumer profiling use cases in its online, print, and television operations.
Up until recently digital analytics relied heavily on standard off the shelf tools. However, changes to customer behaviour and advancements in technology necessitate and empower a move towards bespoke and customised solutions. How far are you willing to leap from the comforts of the standard to the benefits and pains of the bespoke?
Tom Betts of the FT has been grappling with this question for the past few years. Two years ago in X Change 2014 he was a lone voice in a big data tech huddle interested in building a self-service digital analytics solution from scratch. Last year at the DA Hub, about a third of delegates in the room were either considering it or had started.
The FT has recently completed building its own bespoke solution. In this discussion Tom will share insight for this project but will also solicit the group’s opinions on questions such as:
- What is the business case for building an in-house stack?
- How do you obtain management buy-in?
- How far can we go with off-the-shelf products?
- What are the potential pitfalls you must consider before embarking on this journey?
- How do you continue to grow your solution to address changing conditions?
Whether you at the consideration stage or already into the process of building your own solution this discussion will help shape your thoughts around your next steps.
Data mining and data science are the current buzz terms. Applying data science methods to digital analytics seems to be the next logical step on the maturity scale for digital analytics. However, the description of specific use cases and challenges where data science can help, as well as the outline of methodologies and toolsets to use, is rare.
So let's get together and share experiences on initial or more advanced steps on where, when and how data science methods can be used in digital analytics. We will discuss interesting challenges and match them with the most suitable methods, look at the prerequisites and transformation requirements on the data side. Our goal will be to gather ideas and directions on practical application of advanced analytics.
In a recent academic research conducted by Emre Soyer and Robin Hogarth three groups of economists were asked the same question relating to the same dataset.
- The first group was given the data and a standard statistical analysis of the data; 28% of these economists got the answer wrong right
- A second group was given the data, the statistical analysis, and a graph; 39% of these economists got the answer right
- A third group was only given the graph – 97% got the answer right
The above results suggest that visualised data on its own could be much more powerful than the same data accompanied by analysis or the analysis (descriptive or numeric) on its own. So why are we still using analysis? In this discussion, led by Lukas Grebe, we will explore ways to transform our approach to presenting data and insight. We will ask whether data visualisation is best utilised for explanatory purposes or could it also be used for data exploration and insight discovery.
If you are interested in using visualisation as , replacing static reports with bespoke interactive data tools and finding ways of opening up 'train-of-thought' analysis beyond the analytics team then this conversation is for you.
As the line between the digital and real worlds becomes increasingly fuzzy, digital analytics is being used to help with decisions beyond just online optimisation. It is now deployed to help in areas such as stock control, in store footfall, ROPO sales, stock buying decisions. In some cases digital analytics data may be more granular than back office systems data.
Digital analytics is starting to move closer to its mainstream cousin, business intelligence.
In this discussion we will draw on the group’s collective experience to answer key questions such as:
- What are the similarities and differences between BI and DA and why should we care?
- Is digital analytics a subset of the BI function and if so what should we be doing about it?
- Why do digital analysts often struggle to make our voice heard at the ‘Top Table' and what can we learn from BI analysts?
- What do we need to do to thrive in the broader business environment?
Join what should be a thought provoking discussion on the coming of age for Digital Analytics.
…yelled the Kaiser Chiefs in 2004. Now we’ve got your attention it’s worth highlighting that our digital analytics tools are moving in the direction of predicting said riot. Statistical and predictive modelling features such as automated insights and cluster analytics are becoming increasingly available. You only have to look at product roadmaps to see that more are on the way.
The situation poses big questions. Do we already have a need for such analytics capabilities? Are they being adopted by analysts or simply left to gather dust? Are we able to learn these new features “on the job” or do we need to reskill? Are they potentially game changing?
For this discussion, Dan Grainger of travel giant TUI Group invites delegates to bring forward their views and opinions. Together we will compile a picture of where digital analytics is moving and how we could better use emerging tools to improve business value and customer satisfaction.
The debate relating to creative vs. data led marketing is a long standing one. The emergence of programmatic has only served to heighten the debate. But is it really an either-or thing? Should we focus on funny cat videos and sparkling unicorns cause “data says so” or does high quality content still hold the key to marketing success?
In this discussion, Christoph Hell from Sky Deutschland will ask whether data is a cure or a curse for creativity and is there a line, thin or wide, separating them. We will examine the risks of switching to a pure data driven content creation strategy and ask what changes are required to identify and escape the potential of a self-fulfilling prophecy.
With an ever increasing landscape of sophisticated data visualisation, analytics and BI tools at our disposal building effective dashboards should be easy, shouldn't it?
Dashboards are supposed to be tools that drive decisions, action and improve business performance, yet so few businesses seem to have achieved this. Some dashboards even inhibit business performance while other dashboards end up in 'data limbo' with managers considering them as irrelevant, inaccurate or ignoring them completely.
In this discussion, Adrian Nash will invite attendees to share their challenges and successes designing and building dashboards, and propose different approaches taken from digital analytics, BI and data strategy - including how to:
- Select relevant metrics aligned to the dashboard purpose (e.g. strategic or operational)
- Generate insight that can and will be acted upon
- Ensure your dashboard enables improved business performance
- Anticipate unintended consequences (e.g. first and second order impact metrics)
- Combine data from multiple data sources - with and without data warehouses
- Identify lead versus lag indicators
The advent of new computing technologies is transforming machine learning. Whilst some machine learning algorithms have been around for many years, the ability to automatically apply complex mathematical calculations to big data – over and over, faster and faster – is a recent development.
If you have had a chance to experiment or do real digital analytics work in machine learning, please attend (we expect a small group). Machine learning is at the cutting edge of digital and this is a chance to swap knowledge, learn about tool experiences, and discuss strategies around machine learning – from the right tools, to structuring the data, to the most interesting problems. Not for beginners.
Our landscape is largely owned by product managers both for online pure players and the digital teams within a mixed business. These managers know their product inside out, often own the roadmap and know exactly what the goals are for a successful digital project. So it is the perfect match for digital analysts to work with right. Or is it?
In this discussion, run by Benjamin Stephens of ASOS, we will consider whether it is beneficial for analysts to align with product teams and whether this really means analysis will be done in line with the people responsible for delivering change. Would this setup benefit the product team, your analysts, both or neither? We will look at the pros and cons of having analysts focused on specific components of a site or app and whether this is creating a dependency on another teams, their structure and resource or actually giving the analysts teeth?
You will come away from this discussion with a newly informed way of thinking about product analytics.
How many times have your projects overrun or under-delivered because of reliance on third-party tools or consulting? Could the relationship be managed better?
Suppliers have limited resources – and you are not their only client. Some clients shout louder than others to try and get what they need, while others maintain a friendlier relationship. How can you make sure the suppliers you deal with meet your needs efficiently, on time and inside budget?
From negotiating contract terms to managing custom development work, Nick Redding will lead a discussion aimed at sharing tips and tricks for getting the most out of third-party suppliers and making sure the tools you buy or develop deliver what you expect them to.
Analytics without a process is doomed. How do you go from an ad hoc, immature, learn as you go approach to a fully-fledged, disciplined, matured set of processes with improved quality and effectiveness? Basically, how do you repeat success?
In this discussion, digital analytics thought leader Stéphane Hamel will argue that we can learn from other disciplines. One such example is Six Sigma, an approach developed in 1986 and put forth by Jack Welsh himself at General Electric. Defined as “a disciplined, data-driven approach and methodology for eliminating defects in any process – from manufacturing to transactional and from product to service” it can readily apply to our field of interest.
You will hear Stéphane repeat the Six Sigma DMAIC mantra - Define/Measure/Analyse/Improve/Control - do you agree? Where does Six Sigma, Lean, Agile concepts, Kaizen and those other approaches with fancy names fit in? What’s YOUR approach? What works or does not work?
When combining customer behavioural data, customer satisfaction, session replays, usability testing and ethnographic fieldwork data, one could claim to be capturing the full customer experience. However, have we left anything out? Are there any best practices for a complete/standard customer experience metric? How many different types of data are needed to paint a picture of customer experience?
Brand affinity research, social media analysis and man-in-the-street surveys can be valuable but which data sets have you found to be the most revealing? When does data become more expensive than it is worth? When does it become un-correlate-able?
In this discussion, Jim Sterne wonders if we can identify, sort and rank the best/most necessary data elements and attributes to develop a recommendation to the rest of the industry. We will share experiences, good and bad, of using diverse data sets to augment traditional digital analytics data.
You should come out of this discussion armed with new ideas to track your customer journey.
Personalisation has been a common buzz word for years now. But what does it really mean? Businesses have grasped how important it is but still not sure exactly how to do it.
In this discussion Tim Robinson of Marks & Spencer will look at different examples of how to drive online personalisation and some of the pitfalls in doing so, around structure, process and tools:
- Structure: how should we set up our resources? Should personalisation be the responsibility of a central team?
- Process: how can we get buy-in where we need it? What level of input is required? Do we need to test everything? How can we ensure we don’t ‘optimise to nothing’ and still showcase breadth?
- Tools: what are the third-party toolsets we can use? What can/should we be doing in-house? What are the limitations – where do we stop?
This would be a great opportunity to share success stories as well as challenges and disappointments so we can all learn to do it better.
First party data is promising to be a game changer in improving digital marketing ROI. We are at the point where technologies relating to first party data are robust enough to deliver tangible value. Digital Management Platforms (DMPs) are a good example of this trend. Now marketers can leverage first party data to improve targeting accuracy and serve customers more relevant content at the right time.
There are challenges hindering these promises from realisation including privacy concerns and the speed of changing organisations cultural DNA. This discussion will focus on tackling the trickiest questions organisations face when dealing with first party data including:
- How does first party data utilisation change work flows and processes in the omni-channel organisation?
- What requirements first party data usage creates for digital data collection?
- Do we need a DMP in our organisation and how would we benefit from it?
- Should we purchase a commercial DMP or build an in house solution?
- How do we tackle increasing privacy demands due to first party data?
- What best practice digital marketing strategies can we implement to achieve results quickly?
- How is data science different from traditional BI and digital analytics?
- Which business questions should data science answer that we cannot answer with traditional analytics?
- What are the prequalifiers for your organisation to move into data science?
- What are the skills, capabilities and processes required to make data science successful in your organisation?
- Would data science eventually take over traditional analytics or will the two disciplines coexist somehow?
A pre-requisite for analytical maturity is automated reporting. This allows people in the organisation to spend more time on add value initiatives rather than cranking the engine. But even when we use out-of-the-box analytical collection tools, we find that we need to integrate that data with other data sources - marketing cost, CRM data, inventory data, promotional and merchandising activity, and competitor intelligence.
Every new data source increases complexity and the fragility of the data pipeline. Integration is only the first step. Invariably, once integrated, the raw data needs to be modelled in order to have any value to the business. This adds more complexity and more potential for error.
In this session we will discuss potential scenarios that can affect the accuracy of the final reports that end users consume on a day to day basis. We will also discuss effective troubleshooting and safeguards that can be put in place to catch and minimise errors.
Finally, we will analyse which of these scenarios are applicable to the digital analyst, not just to the BI architect.
Too many personalisation and optimisation programmes end up as damp squibs because nobody thought about reporting properly upfront. In addition, technology vendors are still far too focussed on execution, glossing over the need for flexible, well-understood reporting features. Ask to see a reporting suite in a sales pitch and watch the panic in their eyes. All sounds too familiar?
Data and reporting are your reward for all the hard work you put into your optimisation and personalisation programmes. It is a fiddly, error-prone endeavour, reliant in equal measure on process and technology. Sandra White of Barclays Bank recognises that reporting and stakeholder engagement are crucial for the success of her personalisation programme. In this session, she will prompt us to contemplate how best to engage our internal stakeholders. We will ask which metrics, tagging and reporting tools we should leverage? We will consider the importance of conclusion reports, as well as how to report effectively for ongoing personalisation programmes.
Whilst this discussion will focus on the reporting associated with personalisation and CRO programmes, there will be important lessons for digital analytics as well.
Digital analytics implementations are difficult. Technical debt, competing priorities, lack of resource. When it comes to implementing tracking of video and audio streaming, take all the challenges you find in a standard implementation and multiply the complexity by a factor of five ten. Organisational inertia, multiple (and ever increasing) platforms and players, a confusing array of metadata, edge cases coming out of your ears, player weight, privacy, conflicting tracking domains… the list goes on… and on… and on.
John Larder heads digital analytics for BBC (they have a video or two on their website, you know!). John will share some of his experiences and will ask how can we finally conquer stream tracking. From planning through metric selection to execution – we will cover the key aspects of tracking streaming content.
You will come away from this discussion with new ideas on how to improve your streaming tracking strategy.
Discussions Day 2 - June 9, 2016
This has all of the hallmarks of our experiences in the rise of mobile platforms. Except this time the change is likely to be quicker and platform providers hold the keys to the data.
Join Tom Betts, Chief Data Officer at the Financial Times, for a discussion about data ecosystem management in a world where distributed platforms rule. We will look at key challenges and successes with accessing and integrating data from distributed platforms. We will also examine how best to leverage this data.
Your company is collecting a lot of data. Unfortunately, it mainly uses this data for tactical purposes. You know you could do more but how do you sell the C-level management on the explosive potential of data?
Ulla Kruhse-Lehtonen has established and led company-wide data and analytics programmes in both Nokia and Sanoma. In this discussion she will explore how to place analytics on the C-level agenda. We will discuss how to identify the highest value use cases, create a capability gap analysis, and calculate financially-modelled business cases to convince and excite company executives. We will also discuss the role of company-wide incentives and a good organisational setup for analytics.
We strongly recommend analytics and optimisation team managers (and those aspiring to become ones) to join this discussion which will be as much about management as analytics.
As the popular saying goes, content is king, and companies continue to invest vast sums of money in flaunting their products and services. However, it is not unusual for analysis of content to be nothing more than a meaningless data dump made to look pretty.
From images to videos, banners to blogs, they all play a key part in a customer’s journey and we should be measuring and valuing them appropriately. But how? You will find out in this discussion led by Dan Grainger of travel giant TUI Group.
We will discuss appropriate methods for measuring content impact, with Dan sharing his experiences of both the good and the bad. We will also discuss prioritisation of content analysis against other demands on your analytics team, addressing where it should sit and whether it should indeed be considered a specific role.
Most enterprises have struggled with digital. As much money, time and resources as they have invested, traditional businesses often feel – and really are – slower, less accomplished, and less adept at digital than their pure-play competitors. Is that fixable? Are there ways big companies can adapt their culture, their organisation, and their capabilities in ways that make them truly competitive?
In this discussion, led by Gary Angel, we will look at how analytics can/should be integrated into the enterprise to achieve deeper organisational transformation. Topics covered will include:
- Why analytics fails
- What capabilities need to be integrated
- What is the role of Agile
- How to transform enterprise culture around analytics
We all have a responsibility to ensure we act in an ethical manner. Sometimes the boundaries become opaque as we come close to a breakthrough or under pressure from the business to deliver.
In this discussion we will use the group’s collective experience to explore the question of ethics in digital analytics. This ranges from how we collect data, to what we find out about customers, how that is communicated internally and how we should act upon the insight.
It is a pretty wide ranging area that touches on many aspects of digital analytics from split testing to predictive analytics. We will start by trying to answer questions such as:
- What are the main areas in digital analytics where we are likely to be confronted by ethical issues?
- What are the main areas in digital analytics where we are bound (voluntarily or otherwise) by ethical standards?
- Where is the line drawn between practicality and ethical best practice?
- How do we manage our obligation between the needs of the business and the rights of the customer / visitor / data owner?
If you don’t attend this discuss you are either a saint or you have no moral compass at all.
Ecommerce, application (e.g. banking) and lead generation websites offer tangible conversion measurement points. The same cannot always be said for other website types such as media, entertainment, informational and brand websites.
These days managers are expected to demonstrate ROI on any digital activity including non-transactional websites.
As the manager responsible for analytics on the non-transactional part of the LEGO.com website, Antje Wolter faces that question every day. In this discussion Antje will look to explore with you how to evaluate the success of a non-transactional pages, how to create new KPIs such as engagement scores and how to calculate the value proposition of a pure marketing page to the overall business value.
Attribution modelling has been around for a while. All the same interest is now shifting from focusing on the customer’s touch points and conversion paths to understanding the customer holistically with the aim of optimising their life time value.
Advances in attribution technology raise new challenges. Which data sources should we include; what models to use or should we build our own models to fulfil specific business needs?
Teemu Relander has been leading many marketing analytics projects over the past few years. In this discussion he will guide us through both opportunities and challenges marketers are currently facing and will face in the near future when using attribution modelling for optimising the customer life time value.
As a discipline User Experience (UX) is focused on understanding your customers’ needs and delivering the easiest possible series of actions to complete an action. For years UX claimed to base its decisions on data. However, in reality the number of UX designers using behavioural analytics on a daily basis is quite low. Clearly this is a sub-optimal situation where designers are taking decisions without fully understanding their customer’s behaviour.
In this discussion, led by Eric Bernhard of Dixons Carphone, we will consider methods to bring UX and analytics to collaborate with each other and share insights. We will look at various aspects of this challenge including the professional language we use to communicate to each other, cross training and physical proximity. We will also ask whether the ideal solution is building integrated UX/Analytics teams. What are the benefits, risks and challenges we face? What sort of analytical challenges we could expect to resolve? What are the errors we made on the way? What could be improved?
In doing so we aim to help those looking to extract more value from their UX and Analytics resources and introduce new ideas for those already in the process.
Multi-channel businesses face the increasing challenge of ROPO – Research Online; Purchase Offline. Companies look to offer customers more choice but in doing so we create a measurement challenge for ourselves. Consequently, recorded online sales inevitably under represent the actual value of our digital assets which in turn threatens to undermine the ROI on digital investment.
Insurance is one of those businesses greatly affected by ROPO. In this discussion, Benjamin Erhard, BI & Digital Analytics evangelist at Allianz Group, will look to explore methods for measuring the impact of ROPO in hybrid business models. He will invite participants to share their experiences on technical concepts and functional assumptions to merge the online and offline worlds (e.g. fuzzy matching, cohorts, latency times etc.), newly arising challenges (e.g. data integration, data privacy, data accuracy etc.) and organisational changes (e.g. impact on remuneration models, commissions etc.).
Finally the discussion group will consider means to leverage data to persuade project sponsors to maintain project funding by understanding the full impact of their investments into the online channel.
The big dirty secret of the optimisation industry is that most organisations have broken tests. These tests are either biased when they get launched or are collecting the wrong data and testing the wrong things. Companies don’t know how long to run their tests for, when to stop and how their business and purchase cycles impact the data they see.
Last year, Craig Sullivan led an exciting discussion around this topic. The discussion was so popular that we asked Craig to run it again this year taking into account all the new developments in the market.
Using his experience of breaking A/B tests over 70 different ways, Craig will explore what painful lessons the group has learned through making testing mistakes. Rather than focusing on great results or amazing optimisation claims, this discussion will provide you with priceless examples of how to avoid wasting immense amounts of time or resource.
Craig will use examples and common problems to open up discussion in this area. Participants will be asked to share the mistakes they have made in organising, running or analysing tests – to help us all improve our work in this area.
Competition in the digital world is intensifying as barriers to entry keep dropping. With the advent of digital-savvy management the desire to consume data is increasing. Data is quickly becoming the next battleground. Data analysts along with data scientists are the most sought after soldiers. So are we delivering the expected return on investment in analytics or RoA?
Anna Denejnaja is constantly challenging her team with this exact question – looking to tackle the challenges of prioritisation and selection of analytics initiatives. The measurement of RoA, as contribution versus cost, is not an easy one. Digital analytics teams commonly serve multiple stakeholders and depend on them for any recommendations’ execution.
In this discussion we will work on examples and experience from attendees to explore the methods and challenges for a RoA calculation. Topics will include:
- How and whether it is necessary to quantify the distinct contribution of the analytics team
- A look at the investment and organisational processes required to enable any RoA
- A review of the desired outcomes and how they would vary between different types of organisations
- Why knowing your RoA can further leverage the strategical position of digital analytics
In addition, we will look to examine the merits and risks of accountability and create an understanding of how focusing on RoA can be used in our daily work to improve our analytics impact on the business.
Market forecasts for Internet of Things (IoT) and wearables show a hockey-stick growth - up and to the right. Whether the forecasts are conservative or optimistic, these devices offer the measure community an exciting opportunity coupled with great challenges.
In many cases analytics would be extended from the sales process to the usage of the product. How often is my product used? Where is it used? What features are most applicable? This can aid in product improvement as well as customer understanding. What's more, optimisation now extends beyond the digital sphere into the real world where these devices reside.
Will the frameworks that have helped us organise our digital measurement efforts apply in the IoT data world? Are these just another cross-channel data source or are they inherently different? How will we secure the data streams and ensure we remain compliant with our privacy policies? Join us for a discussion on metrics, consumer expectations, security, privacy and data architecture for the next generation of analytics.
There has been a significant growth in the number of tools and the sophistication of methodologies for shaping the customer experience over the past few years. Digital analytics, voice of the customer, qualitative research and predictive analytics are used by many organisations to gain insights into their customers' behaviour, reactions, motivations and propensity to make certain decisions.
Great emphasis is placed on the technologies and competencies required to master each of these areas in its own right. However, it is only by driving synergies across them that we can gain the necessary understanding of customer experience to influence more positive outcomes. In this session, led by Federica Ancona of Sky, we will explore:
- The various insight team structures for driving a holistic understanding of the customer experience
- Methods for encouraging interdisciplinary collaboration
- Business change implications – how do you make the transition from a siloed insight team structure to one that drives synergies?
This is a discussion for those looking to learn how to extract greater business value from customer insight teams.
Do you really know how much incremental value your development is driving? Can you really trust your A/B and MVT results? What about the changes you cannot test? What if you are continually releasing changes a number of times a day?
In this discussion we will look at how much we really think we can measure (and, more importantly, prove to senior budget-holders!) when it comes to the value of website changes. Can we isolate these improvements from other influences like marketing, promotions and stock availability?
Optimisation tools often tell us our changes will make millions but how much do we trust them? Do we actually see these benefits realised? Tim Robinson will share the approach Marks & Spencer have taken and discuss what opportunities exist to enhance our understanding of the value of development.
Whether the projected future looks bright or dim it is an analyst's job to help measure where the organisation’s performance is heading. More importantly, they should be able to identify the combination of factors that puts the organisation on the most favourable path. However, factors keep changing - strategy, tactics, product mix, marketing mix, ongoing tests, promotional activity, product launches, competitor activity, etc. All of these make it very difficult to make informed projections.
In this session, led by Carmen Mardiros, we will discuss how forecasting is currently being tackled in digital analytics, the different methodologies and models used and of the level of accuracy these forecasting methods yield.
We will debate whether the increasing volume and richness of data and rising adoption of tools such as R amongst analysts are changing how we approach forecasting within the organisation.We will also ask whether the usefulness of the predictions for the organisation is increased as a result.
Given some KPIs, could you define the best dashboard ever? How would you know it is the best anyway?
Let’s tackle the challenge together - from Stephen Few best practices to the harsh reality of the HiPPO veto, you will manage to learn something new, certainly argue a little bit, and even have fun doing it!
Industry veteran Stéphane Hamel will be the leader of this discussion and you are expected to play an active role in defining the sexiest dashboard ever. Bring your own dashboard templates (ex data) and share what has worked and what has not worked for your organisation.
Technology for digital optimisation is rapidly changing. As the size and scope of problems that can be solved steadily increases, the complexity of the underlying technology is growing. This poses a significant challenge for businesses that are already struggling with turning sporadic testing into structured optimisation programmes.
In this discussion, led by Ashish Umre of Tesco, we will consider some of these challenges and ponder what the future might hold from a strategy and a tools perspective. We will also look at the following questions:
- How should businesses approach real-time personalisation / recommendations?
- What is the case for server side optimisation?
- How do we optimise for a world of Internet of Things (IoT) and wearables?
Looking around my business I count five high profile teams who are considered ‘analysts’ and owners of a data source, not to mention the renegade “I know what GA is” contingent throwing a spanner in the works. As a result, it becomes difficult to stay in the loop with analysis conducted, what the business is asking and whether the right questions are going to the right people.
In this discussion we will examine whether the role of the analyst demands skills so varied that it can set teams apart. Would it be more effective to have separate teams for digital (behavioural), sales and customer data? Is it better to have specialists in different teams who play nicely together or are we then missing the big picture - structure vs process? Have you gone too far when you are teaching Finance what cookies are and what a tag means?
Join us for a stimulating discussion about how to manage the analytics process in light of increasing demand for data insight.
Of all online marketing channels social media is most frequently where the prevailing decision making model is the “HiPPO” (Highest Paid Person Opinion) rather than a data-driven decision model. Given we have access to near infinite data that can be processed to get descriptive, predictive or prescriptive insights, shouldn't data always win against HiPPO?
Social data contains a mix of irrelevant noise and invaluable signals which, if measured precisely, can generate the insight required to evaluate your business from your customers’ perspective. How do you increase the signal to noise ratio and use the resulting information to deliver return on investment, transformational change and performance improvements?
In this discussion, led by Christoph Hell from Sky Deutschland, we will share experiences in social media analytics and how to deal with the characteristics of social media. We will also look to define and understand the various possibilities of integrating analytics across social media teams, discuss the most valuable KPI sets and how to use the resulting information to increase the value of social media without causing creative clashes.
Digital analytics was born as a cloud technology and has always been agile and responsive to emerging business requirements providing analysts the capability to radically improve business performance. Yet I often hear analysts expressing frustration at perpetually working on short term tactical solutions, in digital data silos or wanting to see more strategic impact from their work. In this huddle we are going to discuss how to change that and, with BI platforms becoming increasingly cloud-based and agile, how to avoid being side-lined or even becoming irrelevant.
Adrian Nash will propose that Business Information (BI) capabilities are transforming and digital analytics practitioners will need to develop new data or analytics (insight) skills to benefit from the resulting opportunities.
Like any analytics capability, BI should be viewed as part of a broader data management strategy. In this session we will discuss some key components of data management strategy and the opportunities this creates for digital analytics practitioners to become more strategic:
- Data warehouses: why projects no longer need to fail (or take 18 months) and why you do not need to keep your digital analytics data in a silo
- Self-service and Meta data: how do you help business users generate their own reports (so you can focus on generating insight)?
- Organisational design: digital analytics capabilities are increasingly being combined into BI platforms and more centralised BI teams - what is the impact of emerging roles and accountabilities for digital analytics practitioners?
- Data governance: new data creates exciting new analytics possibilities, but how do you manage the new risks of data protection and security?
- Data quality and master data - what does it mean in an integrated data and analytics environment?
Data management strategy is an ongoing activity; you should leave this discussion with increased knowledge about how to benefit from it and a combined digital analytics and BI skillset to enhance your career and deliver sustainable business performance improvement.
Most companies consider themselves or at least outwardly claim to be “data driven”. Lucy Butler, responsible for all digital intelligence at retail giant Argos, questions whether this claim holds water.
In this discussion she will challenge us to collectively define what does a data-led culture really look like, and how can we embed one? We will also look at the following questions:
- How do we ensure stakeholders seek data-led insight before decisions are made; and are willing to reconsider their opinion when the data tells them something that surprises them?
- What to do about the "HiPPO" in the room - what is the most effective role of senior leadership in a data-led organisation?
- What is the role of an analyst in a data-led organisation? Does having a separate analytics team give other teams the opportunity to "pass the buck"?
- What org structure and set-up works best to effectively integrate analytics into all decision-making?
No two customers are the same. They have different interests, behaviours and value to your business. One customer may be loyal; another at risk of churn. The same customer may behave differently at different times of day or week or on different devices. With the technologies now available, treating customers identically is not an acceptable option and is most likely costing you in lost profitability.
In this discussion, Nick Redding of William Hill will look at some of the methods available to bring customer modelling into the world of digital analytics and optimisation to deliver targeted promotions and messages. This session will explore how using value, churn and propensity models can enrich customer experience and offer a consistent message across channels. It will also explore the challenges in integrating traditional clickstream, CRO and data science.
You will come away from this discussion armed with new ideas for starting or enhancing your personalisation programme.
A recurrent theme across the analytics and optimisation communities is finding the right mix of people, process and technology. There is no one structure that fits all yet most organisations struggled to strike a good balance.
As the Optimisation Manager for Tesco Ashish Umre is constantly looking to get that balance right. In this discussion he will explore with you the following questions:
- How do you identify the problems worth solving?
- How do you identify the right people, processes and technology mix?
- How do you engage stakeholders in the optimisation process?
- How do you collaboratively create/share best practices?
- How do you create a scalable programme?
You will come away from this discussion with a set of insights and ideas for moving your programme from basic testing to structured optimisation.
The world of the digital analyst is converging rapidly especially with its non-digital counterpart. Once a stronghold for technical, “vendor specialists” digital analytics is now changing as businesses demand more. This is partly driven by a need to connect with and understand users in an omni-channel way. Digital analytics teams now have to venture out of their comfort zone to understand the full picture. They must broaden their skill sets to extract insights and value from a wider range of data technologies and bespoke tools.
Many of these tools are much more advanced with an increasing amount requiring developer skills. In this discussion Dwayne Browne will explore what it means to be a digital analyst in 2016 and what this trend means for practitioners in this space.
Questions covered will include:
- Buck the trend - is the time for tool-specific skill sets truly in decline or will it become even stronger over time?
- Do we expect our analysts to swap the “Digital” with “Data” in their titles?
- Digital analytics has always been high demand low supply – will there be brighter days ahead as people cross skill?
Join what is promising to be an exciting discussion about our immediate future and how to keep ourselves relevant in a fast changing business environment.
Growing your role into an analytics services advisor requires a change of behaviour on your part in order to effect a change in the behaviour of your organisation. How we represent ourselves determines how the rest of the enterprise sees us.
Leading the discussion form the perspective of senior management, Jim Sterne will look to unearth ways an analyst can present their occupation so that the rest of the company sees them as specialists and authorities instead of sentient calculators.
What does an analytical statement of work look like? How does an underfunded analytics team properly prioritise the tasks they can accomplish? What can we do and what should we avoid doing in order to gain respect?
Our output might include a list of positioning guidelines for analysts or advice on helping one’s company become more data driven.
Customer interactions with products and services are increasingly more entwined across online and offline channels. Consequently, quantifying the effectiveness of digital marketing initiatives must evolve to take these market changes into account.
Digital attribution started off as painstaking Excel wizardry whilst marketing mix modelling used to be the esoteric knowledge of niche econometrics teams. Today both capabilities have become commodities; marketers expect technology vendors to provide these capabilities as out-of-the-box functionalities or demand them from media agencies as a core service.
In a commoditised landscape, what sets your marketing measurement strategy apart? In this discussion Federica Ancona of Sky will ask how organisations go about measuring their digital ROI in an omni-channel world. Topics covered will include:
- What does "Return" mean in your ROI equation?
- How do you measure it? What challenges do you face?
- How can you gain an accurate understanding of online/offline customer journeys at a time when privacy concerns feature prominently in our industry's debates?
If marketing analytics forms a large part of your job then this discussion is for you.
Digital marketing technologies continue to evolve and grow. We are loading more and more analytics, tracking, social and advertising platforms to adjust marketing activities and to maximise revenue.
However, are we truly conscious of the price we pay? Sure, this is impacting the performance of our marketing efforts but to what extent – are we really in control of it all? How can we reduce or completely avoid the negative impact that third party trackers impose on our online presence?
In this discussion, led by Holger Offermann of Microsoft Mobile, we will look into the issues that third party implementations pose to our digital marketing efforts. We will share experiences, investigate the challenges across the various technologies and discuss possible solutions for implementation and operations related to third party trackers.
Finally we will conclude on a set of best practices that shall help us gain control of the issues that third party solutions present.
The term “Website Performance” is mostly associated with the technical aspects of the website – load times, response times and infrastructural bottlenecks.
However, what if your page loads fast but your content, the user experience or marketing channel are performing badly?
In this discussion, Antje Wolter of LEGO will argue that the "performance" of a website is much more than the sum of its parts. She would like to discuss how these different parts interact, how they correlate and how can they be measured, then analysed and optimised with classical digital analytics tools or specialised performance management applications.
Data accuracy is a big issue. We are constantly challenged over data quality and accessibility. It is potentially the number one factor holding back our success. So how do we overcome this challenge?
In this discussion, Lucy Butler of Argos will ask how does great data governance looks like and how can we ensure our organisations adhere to it, to ensure relevant data is available, accurate and accessible when we need to ask questions of it?
We will also consider the following questions:
- Why do so many organisations struggle to maintain useful and accurate data?
- What examples can the group share of organisations that do it better - and what are the key elements which support this?
- How do we ensure measurement is even considered up front in every new initiative - and where should the accountability sit?
- How do we ensure data governance is prioritised, when competing with other uses of resource which might appear to have more immediate customer benefit?
The continuous rise in broadband speed and download capacities (at home and on mobile devices) is fuelling an explosion in video and audio consumption via online channels; From marketing videos through socially-shared video content to full streaming services such as BBC iPlayer, Sky Now TV, Netflix and Amazon Prime. Whilst the shift in consumer behaviour is clear, many organisations still struggle measuring streaming activity successfully.
As Head of Digital Analytics at the BBC, John Larder has to deal with various scenarios involving streamed content. In this discussion he will examine some of the key challenges with measuring customer engagement for streaming.
We will ask whether metrics such as completion ratio could be used as universal benchmarks or should we apply different measurement criteria based on the context, type and location of the stream? How do we blend technical performance metrics such as Bit Rate and Buffer Zone with conventional engagement metrics? How do we represent the various factors influencing streaming content (stream length, streaming quality, customer broadband speed, consumption on stationary vs. mobile broadband etc.)? How should our management reporting look like?
If streaming is a key part of your content strategy then this discussion is where you should be.