Episode 307 / Drew Hampton / Varo Bank / Head of Performance Marketing
Optimising Data Communication
Subscribe to Our Newsletter
Simplify your data as much as possible and make it easy for people to digest it.
In the world of performance marketing, data is more than just numbers; it's a narrative waiting to be uncovered. Drew Hampton, Head of Performance Marketing at Varo Bank, shares why it’s critical to know how to communicate data and create a story, on the latest episode of the Shiny New Object Podcast.
Drew's passion for data began early in his career, driven by the realisation that you don't need to be a data scientist to unlock powerful insights. "I just really loved being able to parse data together, tie a story, identify what needs to be done," he explains.
On the podcast, Drew emphasises the critical importance of digging deeper, warning against taking data at face value. "It's truly important to understand one or two levels deep beyond the data," he advises. This philosophy has been a game-changer, helping him save millions by identifying potential investment errors and challenging flawed assumptions.
When it comes to presenting data, Drew is equally meticulous. He advocates for clear, engaging presentations that capture and maintain audience attention. His strategy includes creating slides with clear headlines, key takeaways, and a focus on what truly matters. Drew is also exploring the potential of AI in data analysis and presentation preparation. As he puts it, “AI is basically your thought partner, your brainstorming partner that you can use to refine and bounce ideas off of.”
Learn more from Drew in the full episode.
Transcript
The following gives you a good idea of what was said, but it’s not 100% accurate.
Drew Hampton 0:00
So I think it's truly important to understand one or two levels deep beyond the data, and not take it at face value.
Speaker 0:12
Are you a fitness loving foodie? Then defy snacks is for you. Defy offers real high quality chocolate you love, combined with functional ingredients, giving you as much as 26 grammes of protein per bag. So you can be both a little bit bad and a little bit good at the same time as a female owned brand, defy dedicates a portion of its profits to other women owned businesses currently sold in the US. You can find us on defy with an i snacks.com and follow us on social at defisnacks.
Tom Ollerton 0:45
Hello and welcome to the shiny new object podcast. My name is Tom Ollerton. I'm the founder of the creative effectiveness platform, Automated Creative. And this is a podcast about the future of data driven marketing. And every week or so I have the pleasure and the privilege of speaking to someone exciting from our industry, and today is no different. I'm on a call with Drew Hampton, who is head of performance marketing at Varo Bank. Drew, if anyone doesn't know you, who's listening to this podcast, can you give us a bit of background?
Drew Hampton 1:14
Absolutely. I've been in the field for about 15 years and touched nearly all aspects of performance marketing, including life cycle marketing. I started at a self mode startup, working for just equity back in 2007 2008 and was trying to find a new career. And while working at that startup, I fell in love with performance marketing, especially the data angle and the tech angle. So and I've had a career that's taken me through, exposed me to a lot, and taken me to where I am today.
Tom Ollerton 1:48
What was your falling in love moment with data driven marketing or performance marketing? Sorry, what was that? What was that moment where you got love hearts in your eyes?
Drew Hampton 1:53
You could be heavily involved in data without having to be a data scientist or like, have a master's in math. I've always been more prone to being on the analytical side. I just really loved being able to parse data together, tie story, like, yeah, like, parse different sets of data together to identify a story, to identify what needs to be done, to do the investigation. It was always kind of fun to, like, try to find the narrative, what's working, what's not. I just really fell in love with that aspect.
Tom Ollerton 2:21
So what's been the best investment of your own time, energy or money in your career?
Drew Hampton 2:28
Well, not surprised, but really it has been understanding data and how to tie it together. It's truly incredible how far you can get with just VLOOKUPs, some if statements, pivot tables, knowing how to use mapping and just using simple brain logic. Every job I've gone to, I've used the same set of skills and kind of built upon them, and it's helped me quickly advance with whatever role I've been in. Helped plug gaps whenever there arises, and we haven't, like, We're short of data analysts, and it's just really, I can't really stress enough, like, how important it is. Another thing too, is, in reality, I hate manual data entry and building reporting templates. On the idea of building reporting template you can automate can be fun and exciting, especially if, you know, eventually use our AI agents to help automate some of these processes. And you know, even my previous chief proc officer, he once joked he's never seen someone enjoy Excel as much as me, where I don't know if that's true, but I just love to build and automate things and then try to find the story and the narrative. So that's just definitely been a game changer in my career for up to this day, to be honest.
Tom Ollerton 3:38
And did you have any other advice you find yourself sharing when it comes to data driven marketing. So in the role now or in previous roles, what's been that bit of advice that you hear yourself sharing most often?
Drew Hampton 3:49
It's a great question. Actually, my answer ties into my previous answer and one that was provided to you by Mr. Sobrado, who gave a talk on predictive modelling. He talked about the value of looking at charts and tables and the ability to investigate and see how data is tied together. I actually want to expand upon that, especially for the younger people in the industry, is the value of being able to manipulate the data yourself, understand one to two levels beyond the top layer layer, and by extracting manipulate I'm just talking something as simple as downloading reports and manipulating it in Excel, knowing how to create functions like CTR and CPM and CPC in Pivot Tables, which is surprising number of people don't know how to do. And the reason this is valuable is because one too often we assume data and tracking is clean, when I think most of us know that it's not. That's not the case, and there are error, frequent errors, incorrect assumptions and data at times at best can be directional. You have to almost triangulate it. And by being able to manipulate the data yourself, tie two data sources together via key play around for yourself, you'll not only be able to better understand how it's connected and what affects what, you can spot errors, figure out insights, identify trends. You know, like in my career, I have saved probably millions and millions of dollars by either identifying data errors or preventing, like saying no to certain investments based on flawed logical thinking, by understanding how like, our multi touch attribution model works, or understanding the logic, or kind of looking at the data, double checking, going, Wait, is this too good to be true? Why is it saying this and diving in? And when people or engineering says, No, it has to be this way, you present them the data, and you force them to look at something. And so it's just being able to really, one, tie that data together, one or two levels deep, and doesn't have to be something fancy, it can be something simple, but you'll understand how it moves around, what impacts what, what's real and what's not, what's too good to be true, what needs more investigating, and also what is, maybe, in your opinion, flawed assumptions. So I think it's truly important to understand one or two levels deep beyond the data, and not take it at face value. If you do that, you'll better understand what. Maybe things are being overweighted. Maybe all your ads are 90% view through conversions based on like a 320 by 50 ad, you can instead of like No 50 or 80% click through conversion. So it really can help you, kind of, keep organised, keep abreast of the situation, understand how the data interacts with one other, and be able to dive in and, more importantly, eventually, figure out the story and identify the story.
Tom Ollerton 6:36
It's really interesting advice, but I want to hear a story. Can you ... you don't have to name names, but it sounds like you've got the attitude of asking the question of the data and not taking it as correct because it's come out on a spreadsheet or a deck or whatever it is. It sounds like you go: really why? Well, like, prove it to me, go two more layers. So can you tell me a story about when that worked for you?
Drew Hampton 6:57
Yeah, absolutely. One of my favourite examples was a company where we, where I helped work with our data analyst, and she and I became close friends putting together a multi touch attribution model. The first time I ever worked on a multi touch attribution model, and it was called a starter player closer model. For those who are not familiar, basically what it means is the starter, the first touch that gets 25% of conversion credit. The closer is the last touch that also gets 25% of the curve conversion credit. The player gets 50% of the conversion credit. It's all the touch points in between. So let's say there are 10 touch points, the first touch point and last touch point get half the credit. The remaining eight touch points in the middle get the main 50% of the credit. And when, initially, when we started using this, I was using Criteo, which was a programmatic display remarketing service to run a remarketing ads. And the results for Criteo were incredible, but they didn't sit right by me based on some of the data I've seen, and also knowing how like Criteo worked, and also, more importantly, how of our what such attribution model worked. So I talked to our data analyst, and I found out that the head of data valued an impression the same as a click, which I fundamentally still do not believe in this day, because a click shows action, usually intent, whereas impressions, you know, can you remember how many ads or what ads you saw yesterday? Does it could be below the fold. There's no guarantee. The other thing too is I also saw Criteo, like, had a frequency of 50 at times for some of her campaigns. So basically what was going on is just through view through attribution, Criteo was getting the majority of that player attribution, so like 89% that player attribution, thus basically taking a larger part of the pie that truly deserved, because we're giving so much weight to the impression versus a click. And there was a time where we want to invest, like, a half million dollars, and I had to set up and say, No, I do not think that's an investment is worth, worth it? I said one because of the way the mall set up where you're over, like we're over attributing Criteo; two, if we look at the data, it looks like we have a reach rate of like 90% right now. Three, we ran incrementality test and did not reach stat SIG. So using all those three bear like data points, I was able to change that. I was able to change the goal and focus on investing the money into a much more profitable channel that could scale more whereas remarking channel, while very valuable, and want to be very clear, was a very good channel. It was tapped out, and it was also taking too much credit, and that additional $500,000 would have gone to waste, in that, in the remarketing channel.
Tom Ollerton 10:04
This episode of the shiny new object podcast is brought to you in partnership with Madfest, whether it's live in London or streamed online to the global marketing community, you can always expect a distinctive and daring blend of fast paced content, startup innovation, pictures and unconventional entertainment from Madfest events, you'll find me causing trouble on stage, recording live versions of this podcast and sharing a beer with the nicest and most influential people in marketing. Check it out at www.madfestlondon.com.
Tom Ollerton 10:41
So we're gonna move on now to your shiny new object, which is how to communicate your data. Why is that your shiny new object and help the audience understand what it is and what you mean?
Drew Hampton 10:52
Absolutely, I know at face value, sounds like a dull object, but it can also help surface your shiny insights. The way I look at is, in the last 50 years, we have gone from typewriters to computers to internet to AI. All these have been productivity enhancement tools, but also frequently reduces people's ability to focus for extended period of times. Conversely, it also kind of feeds into the hyper inquisitiveness of people that want to ask a dozen questions, and both either the malaise or overt inquisitiveness can really derail your presentation when trying to present the data, the insights and the learnings, and I think now more than ever, when presenting need to be able to keep people's attention, control the narrative and clearly articulate what you want people to walk away with, not just present them something and hope to digest it. And you know, there are all types of presentations. This could be weekly, or it could be one offs. There's a lot of simple things that to this day I see people fail to do in their presentations. And I mean first, and so I can give a quick overview of the key items, but one it goes down to your approach, structure, formatting and the data. So before you start your presentation, you really need to, like, take a step back and go, What am I trying to communicate? What do I want them to walk away with? Who is my audience, and what do I really need to share with them? If your audience is more broad and people from different departments, different departments, there's a lot of data that they may not need, need to see, or may just confuse them you may need to simplify. So really understand who your audience is and what they really need to know. If you're reporting on a project or a big you know, or quarterly review or something, have one slide that starts with what the goal of the project was, that can keep you anchoring the entire presentation. I've been on... I remember once I was given a significant budget increase and had to report the results to the entire C staff at a company, and I anchored the presentation, beginning with the agenda and then a recap of the goal. And it was the best thing I did, because right when I went to the next slide, to the exact... so I never worked with start asking a dozen questions that were not related to the project or what I was working on, and even the chief proc officer jumped in, goes, he had to go in, like, if you saw last night, that's not the point of this presentation. We're going to be focusing on this. And what it does it gives you ability to anchor what you want to talk about and what you want communicate. So basically, you structure, you're gonna have an agenda. Give a quick overview so people know what to expect. Then with your slide, have a headline and one sentence summary of the goal of the slide. IE, is revenue up. Have a quick sentence reminder of what you're doing. Don't assume people are always going to remember. Have two to three key takeaways, um on or two three key takeaways and, or wins that you put in bullet point. Have additional two, three blockers and, or learnings. Then list next steps. It's very simple. It's very to the point, and it can really just, again, simplify the slide and make it easier for people to digest the data. So even if they're distracted by slack, SMS message, they can also zoom in and see quickly what they're supposed to digest. The structure, as I said, sort of the formatting for key sentences you want to focus on, that want people to focus on, bold a few of the word, words and text and bright colours. So if people again get distracted, turn away, they turn back to the slide, they can zoom in on the key talking points. If there's a table or graph, have one or two boxes highlighting where you want them to focus. Make sure it's the box a little bit thicker, and the colour is very bright. So again, you got their attention. You're focusing them on the key area you want them to focus on. Organise table rose by impact, not alphabetical. Too often I see all this, all these like tables where it's just all done alphabetically. I'm trying to find, okay, what was your biggest spender? Like the top one. Could be, you know, 5% of our budget, what was spent 30% of our budget, or versus 40? Have it in order of impact. It makes it easier for people to follow. And then have a fine print section that shows the acronyms, the extra details to pay on the audience they may not be familiar with the terms, you know, don't hesitate to add footnotes. They can say data is from this range or this is comparing to this. You put small footnotes everywhere, and it doesn't have to be invasive. And again, so when people ask you, one, you're cooking off the top of your head. Two, you can point on slide so they can follow it and review it. And again, they can easily digest it. And then, when it comes to data, it's really about simplifying the data as much as possible. What do people really need to know? For example, so many presentations on that people include impressions, clicks, etc. That's really not necessarily important when you're presenting some of these other metrics. Simplify and streamline the data. Now, remember, the appendix is your best friend. All the data that you can think, that you think people may ask about put in the appendix, you can quickly reference if those questions come up, so you don't have to worry about, if you're worried about, again, not presenting, not having data up that you want people to see, just shove in the appendix. So you can quickly reference. It allows you to keep your slide more simple and more engaging. And you know, overall, this requires a new habit and but once you have a template, and especially with assistance of AI, there's really no excuse for not having slides that can really communicate what you're trying to say and allow you to format so your shiny object, the learning of what you're doing, the key takeaway of what you want these people to like focus on is what comes away with, and they don't overlook it or underestimate its impact. So I mean, that's just something I see people consistently fail to this day in a lot of different ways. And I think with the reliance on AI to automate some of these content creation everything, even while I can summarise it, one, taking an extra initiative yourself to, you know, struck, you know, refine what if you want to use AI to rely on the insights or whatnot, make sure you still refine the content, to make it more, you know, human based, and so people can relate. And also, there's a value to while there is definitely a little bit of an extra work one, it kind of removes that malaise that everybody gets into when they're just washed once, repeating with the same decks over and over, or, um, these are I'm talking about, like, for the weekly presentations, also, even for, like, your big presentations, it shows that you're a control, that you've taken ownership. It shows that you're organised, and people go again. We'll have better takeaways, and we'll also feel more confident in you. Allows you to at the end, let's say you like, I had a presentation where I did not spend the full allowed amount, but our budget I was given, however, I more than increased what our revenue goals were. And at the end, identify some of our blockers and like and or organise them in order of impact. And then at the very end, I said, with this additional budget, I would like to hire an um, no third party to help fix our data pipelines. This will enable me to do X, Y, Z. It will cost about this. And right away, in that same presentation with X, I was able the exact so I was able to get that approval because I was clearly structured in what we did, what we learned, where we succeeded, the next steps, the blockers and just people feel a lot more confident when you're on top of things. And being able to communicate how you're on top of things, communicate what you've done, communicate what you're going to do next, can just make a day night difference in your career and when working with people.
Tom Ollerton 18:50
So how do you use AI specifically, when you're pulling together a presentation?
Drew Hampton 18:55
You know, it's a great question. I'm still in the nascent stages of using AI for context, because I work for a new bank, we have heavy security, and all AI up until a couple months ago, was blocked at our company. We could not access any of the AIs. And only recently was Google Gemini, like basically approved, and we can't use agents or anything like that, which drives me nuts, because that could really help but let's just say you're not sure what insights there are. If you play around and all you have to develop a prompt. You can upload a data set and ask, you know, Gemini or whatever AI you're using to define what are the three biggest trends, what happened week over week? What creatives are doing this? What channels change you have to, you know, iterate on the prompt and work on it. But that can just give you the three key takeaways. They can plug in place in, you know, it's and then, once you have that, and then also, you use AI to go, I am presenting this to this group of people, all this exact it's going to be around this. What are some sample presentation formats, or how I present it? Can give you an idea. Of how, like, I think the biggest block for people is getting the pen to the paper and how, what template they want to use in order to present their data. And it can really be like a thought. No, it could be brainstorming tool for you to use and how you want present what the narrative is. And then once you get in the rhythm, you can use it to help find insights. And more, kind of increase, reduce the amount of time you have to do the analytical rigour yourself. And I want, I won't say always, because there's all I still think there's a huge human value to doing something else as yourself, again, depending how frequent these tasks are. But it could just really help. It just helps consolidate and just like make the process more efficient and allows you to do other things too. So you're still doing your job, you're still providing insights, you're still communicating the key takeaways, but you're also saving time by not having to sift through everything yourself, necessarily. Or, you know, if you're having trouble framing something, it's very long winded, how do I simplify the statement into three bullet points. How do I like here's a slide I created. What is the key takeaway from this slide? If you can't figure out how to articulate yourself, so it's basically your thought partner, your brainstorming partner that you can use to refine and bounce ideas off of.
Tom Ollerton 21:15
So Drew, unfortunately, we've come to the end of the podcast, and you gave such a well considered and passionate view of how to represent your data and story tell with that data. And I think you gave like, five or six very clear points and actions, and that was inspirational to me. So thank you for that. I concur specifically with your first point is that I used to work for agencies for years and had this thing I call slide salad, where everyone was just piling in slides like this, like, Oh, I'm not sure we're going to use these, but here's five slides, here's... So you end up with this ginormous thing, and then, like, six people are all stood around like arguing the toss over every slide. It's a disaster, whereas it'll automate it great, if we go to your point, what is the one thing that we want people to remember from this presentation? If they forget about it in three hours, three months, whatever, and then every single slide, every single title image, ladders up to that thing. But if you don't have that thing, you're just doing slide salad, and you articulate it more beautiful than that. But I would say that I'm very much on board with that. Yeah, be clear what you want to say and represent it in a way that the audience will will pay attention, but I also really like the admission that people are going to get distracted and they need to be able to get back into a deck once they've looked at a Slack message. And I've never thought of it like that before. So thank you for that insight Drew. If someone wants to get in touch with you about anything you've talked about today, where's a good place and what makes a message that you will respond to?
Drew Hampton 22:37
It's no surprise LinkedIn, and I think for a message I'll respond to make it engaging. And if you're looking to sell me, don't make it so overtly sales oriented. I really want gauge I'm like to learn. So love to hear from people. Love to hear their insights, their experiences, and you know what they've learned over their career. So please feel free to reach out. I'm more than happy to talk.
Tom Ollerton 23:00
Drew, thank you so much for your time.
Drew Hampton 23:02
Thank you for hosting me.
Subscribe to the ‘Shiny New Object’ Podcast on Apple Podcasts, Spotify, YouTube and Soundcloud.