Home TypeFeatures Data for Dummies part 2: structure, people and creating value

Data for Dummies part 2: structure, people and creating value

by Simon Parker

In his latest feature, EN guest editor Simon Parker continues his discussion on data with Mark Parsons and Stephan Forseilles.

Two weeks ago, I shared the first part of my data discussion with Mark Parsons and Stephan Forseilles, two data experts within the exhibitions industry.  Mark runs a data business focused on the exhibitions industry, and Stephan is the CTO of Easyfairs. When I set up the discussion, I was concerned about how to make a fairly dull subject (to me anyway) interesting.  Is it not all about advanced maths, distribution curves and practiced by intensely bright people who do not get out much? How wrong I was as it is really quite interesting. The promise of data, and specifically artificial intelligence (AI), is something where it is clear that expectations often do not meet the reality, but there are significant changes occurring and some of the promise is turning into real world money. I have surprised myself in how interested in this area I have become, and how often this subject comes up in my discussions with other leaders in the industry.  Today, I would like to share the second part of our discussion, which builds on the themes of our first article and talks more about how to make this real.

In the last article we talked about how data can give great insight and can help organisers profile their customers to enabling them to tailor products. But this sounds really theoretical – what does an organiser actually have to do, and how should they organize themselves to deliver a great data strategy?

Mark began by speaking of his scepticism regarding data strategy. He said: “I’m somewhat against the concept of defining data as a strategy; a strategy is a plan of action to achieve a long-term objective. Strategy should be something concrete and related to current and potential customers. Data is typically a record of what has happened, it’s a snapshot at a specific point in time of something of interest.”

He added: “In my opinion no one should have a strategy to become ‘better’ at data, this is a road to spending lots of money without much benefit. It is useful to picture data as if it’s a liquid – it’s a flow of information from business processes, data from one process flows to the next. Organisers need to craft processes to capture more relevant data and build the systems to help understand the data flowing through their processes, over time they may build automated processes using this flow, but this is not a strategy.”

Building on Mark’s comments, Stephan emphasised the importance of creating systems and processes to understand data. He said: “I feel very strongly that it is critical to centralise data management and data operations. Data silos present a massive hurdle and when data is fragmented it is often difficult to achieve good insight, and impossible to spot trends across the wider portfolio.”

He added: “There are many financial and technical arguments as to why it makes sense to centralise, but I’ve found that it just makes a difference in terms of the people you can attract and retain. Good data people are hard to find, work best in a team, and need new challenges. It is easy to lose people because they get frustrated doing the same thing, and centralising data operations really helps create best mix of variety and novelty for the team.

“Now having said this, it is essential that we give the event teams the tools to understand their data, gain insights and make data driven decisions. The challenge is always striking the balance between what is done by the data team and what is done by the business. I’m not sure we need the event teams to have the level of expertise of our central data team, but it is critical that the insights from data and ways of thinking aren’t centralised – those closest to the customer need to be able to understand their data.”

Standardisation and building a data business

So, what I take from this is to let the data team deal with the complexity of managing the data, as it is best centralised and essential that tools exist that sit over the top of it to allow teams to interrogate and gain business information from it. Right, got that but what does “good” look like?

Stephan explained: “EasyFairs have made huge strides in building significant revenue from data and digital products over the last few years and have built a business where, even in non-Covid-19 times, this contributes over 10% of revenues with the goal to go above 20% in the upcoming years. We have been providing digital products for 10 years now and we have a range of products that is consistent across the portfolio and sold across the entire business. We have recently developed new products related to online events, which are gaining real traction.”

He added: “We have a specific team that is responsible for developing the technologies that sustain these products and while there is some personalisation by product, they are all underpinned by the same technology. One thing which helps is that for online events we use a limited set of platforms. By restricting the number of partners that are used across the whole group it has made it far easier in terms of integration, data centralisation and efficiency – I understand it is not the same for every organiser but this approach works for us.”

Mark stressed the importance of keeping things simple: “Stephan talked about how Easyfairs used a limited set of third-party platforms, creates technology which can be reskinned, and has centralised his data operations and analytics team. I think these are really important decisions for organisers. In the current environment it’s easy to drown in data, especially if there is no glue to bind data sources together properly – the more connections, the more complexity. In my experience the best organisers are those who are to use data (and automation) to do the day job in an efficient way, but also make sure that data is captured and stored in a way which allows it to be understood after the fact. Now I’m fully aware that few have the budget for data scientists, but putting in place the building blocks to store data and interaction signals in-house is an investment few can afford not to make.”

I then asked Mark about how an organiser might approach getting better at data – is it better to have a data revolution (a big bang) or is this a slow evolutionary change we are going to see with event organisers? He said: “It’s a bit of both, I think you need a big bang in terms of the ways of thinking, but in practical terms changing any businesses is a complex business, and something which occurs best over time. There is a maturity to how companies use data which takes time to build.

“Early on the quick wins of reducing costs or increasing revenues are the right objectives, but as organisers increase their data maturity, the desire and capability to ask more complex questions emerges. Rather than how to effectively merchandise digital inventory, there are questions such as how do I use data to increase loyalty, how do I use data to allocate capital more efficiently, what sectors should I play in, etc. These are the sort of questions I really enjoy focusing on at Events Intelligence. I’m not sure you can get there without going through the first stages, but these more strategic questions are more impactful on longer term performance.“

We are in the visibility business & hacking serendipity

Stephan provided a real-world example of how Easyfairs are thinking about how to use data. He said: “We’ve been doing a lot of work on the product landscapes for an event. This initially started as something which was designed to find companies we were missing, but as we developed our understanding, we were able to train models to assess the attractiveness between visitor profiles and specific exhibitor stands based on their products.”

He added: “We’re starting to explore how we can use this ‘attractiveness’ matrix to more intelligently make recommendations. Now we could build better lead lists using this insight, but we’re not in the data selling business – we’re in the lead generation and visibility business. Selling a raw list of attendees who we think would be interested in your product doesn’t feel respectful of people’s privacy. Nor is it respectful of the trust that is placed in us as community builders. What we’ve start to do using this insight is to help our exhibitors to increase their visibility and attract the right visitors to them. Before the show to help them market their products more effectively. During the show to help attendees find those companies which sell products which we think they might be interested in. And after the show to help our exhibitors make sure that they follow up effectively.

“Serendipity matters, and is a big part of the live offering, but with data we’re exploring whether we can hack serendipity – what can we do to increase the likelihood that it occurs? How do we improve the visibility of those companies to those customers who are most likely to be interested in their products?”

Mark affirmed that Stephan’s point on the value of recommendation is key. He said: “With every exhibitor survey that organisers do, the feedback is always consistent – companies want more higher quality leads. Using recommendations marketing can become more effective via personalisation – matching content and companies with their ideal customers. On the show floor, recommendation can help shape the traffic flow, to the point that it may improve the efficiency of lead generation – increasing the hit rate for exhibitors of ‘high quality’ leads.

“As live returns however there are new sources of training data which can improve models which understand behaviour and relationships, both from digital and hybrid offers and the impact of technology on the show floor. Data to help understand the social graph of a community can be obtained from passively using companies like Crowd Connected or more actively via ‘tap-in’ solutions such as that used by Easyfairs (in partnership with GES) or provided by companies like Konduko.”

While this is getting quite complex, at the heart of this is a respectful and sophisticated approach to data that enables exhibitors to market themselves ahead of our shows and then follow up effectively afterwards. It also provides ways by which we can accurately “recommend” product that is very likely to be of interest to the buyer in a non-invasive way (that effectively hacks serendipity) and creates higher quality leads – sounds pretty good.

There are approaches to data that fit every size of organisation

So, just one final question to Stephan and Mark, our readers range from small independent organisers through to the majors. A lot of what you have talked about is great if you have got the budget, you can just throw money at the problem as an investment in the future. What about the smaller organisers, the ones with say 10-15 people?

Stephan opened with: “I think it is super important to find the right platforms and partners for smaller companies, don’t try to do everything yourself because these are very specialised skills – and with only one or two shows to spread the costs across – this approach can become very costly. Most of the platforms are not expensive and offer solutions to build data and insight. Be mindful, but not to the point of paranoia over the contractual terms over data ownership and the ability to get at the raw data. In terms of building skills, encourage those within your business who show an interest in this area to explore using real world data. If possible, find a freelancer with some data science capabilities (or someone studying this) and share some of your pain points. Remember for someone building their skill in data science, using real world data is really interesting.”

Mark added: “I think it’s really important to realise that data is often not the answer. It’s easy to lose sight of the fact that we make money by bringing communities together. For a small organiser think in a very focused way about what data might help you understand your communities better. Start designing processes to collect features and characteristics which help you understand your customers better. In time this will become a very powerful dataset to help allow you to segment communities and build interaction models. With my mergers and acquisitions (M&A) lens I can also see a situation in the future where data is used as a differentiator in sale processes – data diligence doesn’t really factor into the deal process currently, but as major organisers become more focused on data, the quality and depth of data is likely to be a reason to pay more.“

Given the range and scope of what we have discussed it feels almost impossible to summarise in a meaningful way. My main take outs are that not having a robust approach to how you manage data is not going to wash in today’s increasingly complex, volatile and ever-changing business landscape. If we accept that we are in the business of making connections and providing information for communities to do business, then good data is a pre-requisite and underpins everything we do. Not only that it enables our teams to get closer to their customers, provides better return on investment (ROI) for everyone including business who want to sell for bigger multiples (if you accept Mark’s point on valuation).

There are great technologies and platforms that can help us build this capability regardless of the size of our organisation, but we need to keep it simple at first and build from there. As Stephan and Easyfairs have demonstrated, there is a sizeable business to be built this way providing we have the vision, skills and determination.

I would like to say an enormous thank you to Stephan and Mark for their time and for bringing great insight into this fascinating subject. And not sure if it is it just me but does “hacking serendipity” sound like a mid-70’s Genesis album?

Related Articles