26 October 2022

Expert Profile: Madeline Lisaius - Data Science Specialist

Madeline Lisaius

A data scientist and spatial thinker with degrees from Stanford and Cambridge, Madeline splits her time between academia and consulting. She has worked with communities across the globe and specialises in using big data to address and solve complex environmental problems.
 

We spent an illuminating hour with Madeline, discussing her consulting portfolio, her approaches to team-building (she's a former National Geographic explorer) and techniques for getting to the real heart of a problem; and the question of whether data science has all the answers..!

 

Let’s start with a bit of an overview - what types of projects do you take on? 

 

Well, top level, I work at the intersection of data, the environment and people. I’m a data scientist, yes, but more broadly I think of myself as a solutions builder! 
 

My portfolio covers everything from refocusing team dynamics to capability building to technical training on remote sensing. I’m also brought in as an Executive Advisor - for clients looking for smart ways to lead in the climate space. I get them up to speed on everything they need to know about, say, climate justice, potential brand reputation risks, and recommend what they should explore in more depth. I’ve done a fair bit of speaker coaching too - again, typically on climate topics. 
 

Client-wise, I work with the private and public sector organisations, as well as taking on projects in the civil society / philanthropic space. Recently I worked with a small team to create a product for the British Antarctic survey - essentially a tool to map sea ice extent. The project’s in conjunction with my PhD - it's an incredibly exciting piece of work (and very public - so I can talk about it!) 
 

Data science / machine learning / data driven solutions - they’re real buzzwords, do they get misused? 


Well…! The short answer is yes. Data scientists are asked to do all sorts of data visualisation / computer architecture / systems engineering projects. But the core of data science is machine learning, and building products using huge quantities of data - i.e. starting with the data and using it to build a solution, as opposed to building a solution which the data can then fit into. 


I get approached for many projects where there simply isn’t enough data for a data science solution - what the client is actually after is a visualisation tool, so they can look at the data that is there. Other times, the data’s not accessible: what’s actually needed is a people person, to negotiate access, and then a systems engineer / team to build, say, a data management platform, which will let the client move forward on their original project idea. 

 

In these contexts I become the bridging person - I’ll help the client work out what they actually need, and then I’ll use my network to help them find it. At which point I’ll generally offer to step back - but most of the time I’m asked to stay and transition into more of a project management / guiding role. For me, this is the “full circle” of solution building.
 

Occasionally, I’ll have a conversation with a client who wants to look at, for example, remote sensing data of night lights over cities, as a proxy for a development index - it’s something a government might commission in order to locate key areas of deprivation, and then use the information to inform policy decisions / types of intervention. This kind of project is absolutely in my sweet spot! But it’s rare!!
 

Is it hard to balance your consulting projects with your PhD? 
 

It works really well actually! Being able to freelance while I work on my PhD means I can engage with some really interesting projects and use my brain in a different way. The PhD is completely self-driven, and it’s a very ‘exploratory mode’ way of thinking. When I’m consulting, there’s a problem and I’m looking for a solution. It’s a great contrast - a really rich balance. 
 

Where does your passion for the environment stem from? 
 

I grew up in the Pacific North West of the United States. It’s one of the few temperate rainforests on the planet. Enormous trees. The landscape is spectacular. We’d take an annual trip to the coast - and one year, a whole patch of rainforest had just disappeared. That's what started me thinking about the relationship between people and the environment - and all the complex dynamics involved - social, political, economic, ecological. And because I "see" spatially (generally if I've been somewhere once, I can go back mapless!) - I also started wondering: what if I could see all those different dynamics, those different layers, on one map? What if that map could then be used to find new models of interaction...? 


Then, after high school I spent a year in Ecuador, exploring the Andes and the Amazon. I lived in a place that was close to a recently capped landfill, and I saw the impact on the local community. On the food. On the drinking water. It was so palpably clear: for people and communities to thrive, to be well, the planet must be healthy. I wanted to create solutions to promote this health - that would serve both the human and environmental interests. 
 

My degree was interdisciplinary - social sciences, ecology, computer science, statistics - and my first Masters focused on applying statistics and spatial data science to large landscape systems, like agriculture and forestry and the urban margins. I went back to Ecuador, and spent a year working in partnership with National Geographic and an indiginous women’s organisation - we were mapping deforestation and degradation and building tools that would help the community to advocate for better policy and management.


You worked with National Geographic in a teaching capacity too - is that right? 
 

I did! In Costa Rica, then in Peru. Essentially the programmes were about equipping young adults to enter a new place and learn with curiosity. So they could encounter differences with an open mind. Engage, rather than judge. National Geographic took care of the logistics and I created the programme with a colleague. 


It was fantastic - a really great balance of support and freedom. Of course, the first few days were always super intense - we called it ‘getting to altitude!’ - working out values, a group contract, a collective sense of norms. I learned a lot about how to bond and build teams - the importance of creating a shared language and an open space for discussion. 
 

Have you taken this into your consulting work?
 

Definitely! I draw on these skills a lot when I’m consulting, and team building / troubleshooting work is part of my portfolio. Often it’s not what I’m hired for initially - I bring it in because it’s necessary to fix something - like tension in the team / across teams. Sometimes taking half a day (or longer!), to really focus on the team - unpick tensions, rebalance, create a space where everyone can hear each other - is a fantastically efficient way of clearing away obstacles. Then you can move - together - towards a solution!
 

What helps you to succeed as a consultant - what are your top tips ?
 

  1. Build a network of peers


In one of my first jobs, I had an incredible boss - she’d cultivated so many relationships throughout her career, she always knew someone who could come in when we ran into issues, and essentially, I wanted to be able to offer the same thing to my clients! I’m networking constantly: connecting with new people and organisations at events etc. When I meet with people working in a similar field, I’m always asking myself: who’s a team player, who would I like to work with - and I keep in the loop - where they are, what they’re working on - I stay in touch. 
 

It means that if I reach a point in the project where I’m no longer an expert, I’ll know someone who is. Someone who not only has the skills required, but who, as a person, will be able to fit seamlessly into the project - which is key. It means I can be confident of taking my clients right through - to a full solution. 
 

  1. Work with sensitivity - customise solutions for the particular people they’re aimed at

Give time and space to genuinely understanding the people you’re working with, so that the solution you’re building can actually land. Think laterally. Read between the lines. What’s not being said? 
 

I worked with a team a while back, and there was a key technical tool that they weren't able to use. It was clear that they weren’t comfortable admitting this - perhaps because of embarrassment. So, I created a set of workshops where the tool training was built in, and I presented it as ‘this is my standard offering.’ It sounds small, but it meant I could deliver the training in a way that was socially comfortable for the team. End result:  the engagement was fantastic and the learning efficient!
 

  1. Make sure you’re addressing the right problem 


In my experience, clients sometimes haven’t fully explored the root cause of a problem before bringing you in. They’ve felt pain, which they’ve identified as the problem, but in fact it’s only a symptom. So before you start building any solution, it’s really important to step back and make your own diagnosis. 
 

I start with an exploration phase. Firstly I’ll make sure I’ve clarified 1) where exactly it is that the client feels pain 2) who is involved / impacted - both internally and externally - who are the key players, what’s the relationship landscape? Then I’ll present a plan for how I'm going to dig deeper into the pain point - to get right to the source. 
 

When I have a handle on the problem source, I’m able to present an efficient solution to the problem rather than a bandaid. Clients are usually relieved - often, when I describe the source, it makes sense to them, they just hadn’t seen it from that perspective. I’m careful to listen to people, and I’m careful to make sure that they feel heard. 
 

As a project progresses, my aim is always to make sure people feel the solution that they are getting - that the problem is being fixed for them, for the people in the organisation - as well as for the organisation.
 

Thanks so much Maddy - any final thoughts to add? 


Just one - and I say this as a passionately committed data scientist! Artificial Intelligence and Machine Learning are so hyped as solutions for everything right now, but I think there will always be a critical human component to the great solutions. I don’t think that will ever change. 
 

No tool will fix everything - if someone says that it will, they’re either lying, or under a lot of pressure…in my opinion!!










 

Give time and space to genuinely understanding the people you’re working with, so that the solution you’re building can actually land. Think laterally. Read between the lines. What’s not being said?