OWOX BI Research On The State Of Digital Analytics: interview with Damion Brown

Our research continues! Thanks to Mariia Bocheva, we’re meeting talented analysts, marketing magicians, and data scientists who are true examples to follow. Here’s our next interview with Damion Brown, the founder of Data Runs Deep in Australia.

As always, here are the main topics for navigation:

Mariia Bocheva: Could you say a few words about yourself and your previous experience?

Damion Brown: Sure thing! I was born in England and came over to Australia when I was 28, which is a really good age to do something like that, by the way, because you’ve got an old enough head to properly appreciate what you’re doing, and you’ve got a young enough liver to put up with all the partying you’re going to do!

Anyway, after some bumming around, two failed but brilliant startups, and some odd jobs, I found myself in a sort of marketing job where the remit was “figure out how to use the internet for marketing.” So there was some PPC, some blogging, some A/B testing and, of course, some analytics using things like Webalizer and AWStats.

When I first saw and used Google Analytics, I found myself so intrigued that I realised I’d been sitting at my desk with a full bladder and I really needed to stop doing what I was doing and find a bathroom… the software was incredible, and absorbing, and ridiculously fun. So I figured that I wanted to do [it] for a living.

That turned into a journey of freelancing that took me disastrously close to the dole office more than once, and then somehow solidified into Data Runs Deep, which is an agency with about 15 people who are all incredible.

MB: What hard skills are most important for analysts today?

DB: It’s a tough one because while we often say that the hard skills aren’t important – they’re just tools, right, so how can the tools be the thing that matters – they really do matter. Anyone who’s tried to hire someone with good technical knowledge of Google Analytics will tell you that! So the hard skills that are important, and that someone should invest their time in if they want to land a job in the field, are skills related to the platform you want to use.

Learn Google Analytics properly, implement GTM, get it wrong, get it wrong again, then fix it, expand it, customise it, break it, fix it… rinse and repeat!

MB: What soft skills should a good analyst have?

DB: Free-thinking. That’s all it comes down to. Analysis isn’t a sequence of instructions – it’s an “instructionless sequence.” And if that sort of thing fills you with dread, then you’re not going to have a good time.

MB: What’s the biggest mistake an analyst can make? Can you share some of your analytical mistakes?

DB: Mistakes are what analytics is all about!

I don’t think there’s a word for it, but everyone has done this: you investigate a problem, you craft a really long email with loads of screenshots, and then just as you’re rounding it off with a recommendation or two about what’s wrong and how to fix it, you realise that you’ve misread the report or misinterpreted the question or something. So, embarrassed, you delete all the text and start again. “Hey Paul, this is actually expected behaviour…”

We’ve all done it.

A really big personal mistake was early on when Google Analytics was becoming Universal Analytics; I’d never implemented it on ecommerce. A largeish fashion brand here in Australia had developers who were enthusiastic about the beta [version of] Universal Analytics, and we geeked out on it, and together we got the business to agree to implement their brand-new site with Universal Analytics installed.

Trouble is, at that time the documentation for ecommerce tracking either wasn’t updated, or wasn’t there, or we just ignored it, so the transactions were set up using the old ga.js, not Universal Analytics. That meant that we simply weren’t tracking transactions.

The client rang me at 10pm, I assured them that it was all okay and “it’s just GA being GA, everything else is tracking fine, the transactions will show up tomorrow, nothing to worry about.” After two hours of abject panic, me and one of the developers managed to update the transaction push and we had it working and live at 5am the next day. I thought the client would be angry, but they were delighted. Owning the mistake – and putting it right – was just what they wanted, and it made me realise that often in life, the difference between smelling of sh*t and smelling of roses is a very fine line indeed.

MB: Do you think miscommunication between analysts and marketing teams is common? Do you have any recommendations on how to overcome it?

DB: Miscommunications make the world go round. There was a flood in Australia a few years ago where a river in rural Victoria burst its banks, and a reporter on the scene phoned their report into the news desk at their office and said there were “about thirty sows and pigs floating in the river.” That got reported in the newspaper as “about 30,000 pigs floating in the river,” which is terribly and hilariously different. That always makes me smile when I think about it, though naturally of course I hope the pigs made it to dry land okay.

Sorry – what was the question again? Right, miscommunication!

It’s a fact of life in any professional environment, particularly today where you have distributed teams that rely on the nuance-free communication of email, Slack, and f***ing Jira. Having a briefing template helps – a single-pager in Google Drive where requests from marketing can be standardised with comments about what they need, why they need it, what they plan to do with it, and when they need it by.

But there’s no real solution other than to take the time to properly sit down with people, unpack what they need done, and have a meaningful conversation. Who’s got time for that? [Laughs]

MB: What professional resources or events can you recommend for analysts?

DB: Hands down, Superweek. It’s amazing, and I don’t think it’s much of an exaggeration to say that my little company wouldn’t be what it is today without Superweek.

Being thrown into a closed ecosystem of web analysts from around the world, to learn about how people do things, is just brilliant. I made the trek there myself in 2014 and have been back ever since, and altogether four people from Data Runs Deep attend the conference each year.

That’s unmissable. Measurecamp is great too. If Measurecamp is a gig then Superweek is a festival. So my ideal year would be one Superweek, a dozen Measurecamps, and Marketing Analytics Summit thrown in for good measure :-)

MB: What knowledge are analysts and marketing specialists missing in order to make companies data-driven?

DB: We could always do a better job of understanding scale, I think. We know great stuff about this month’s numbers, this quarter’s performance, or the annual increase in revenue for a business year-on-year.

What we’re not so great at is understanding how that moves the needle on a company’s overall profitability, whether it can make more jobs for more people, whether the share price is up or down, that sort of thing.

I’m not too worried about the gaps in knowledge, though, and we shouldn’t beat ourselves up about it. It’s easy to forget that analytics is a very young industry; it sits within digital, which is also comparatively young.

We’re all just figuring it out as we go along. We don’t need to have all the answers yet.

MB: Does an analyst have to know SQL, Python, and R and build compiled dashboards?

DB: I know none of those things, so either I’m not an analyst or I guess the answer is no!

MB: What are the most important things analysts need to do at different stages of business maturity (startup, SMB, SME, enterprise)?

DB: That’s a really interesting question. I don’t think the problems are necessarily different at the various stages. The scale is different, of course, but a lot of the time we find ourselves sort of shocked that it can take an enterprise eight months to implement a dataLayer, when it takes a startup or small business two days. It’s the same f***ing thing, guys!

MB: What analytical challenges do you have at your company right now? What tools do you need to overcome them?

DB: There are challenges like ITP [Intelligent Tracking Prevention] and ad blockers, which are starting to be addressed by things like TraceDock, a really impressive solution that fills in the gaps that ITP and ad blockers make happen in your data. Solutions like this are wonderful, but a couple of side steps with ITP and the whole thing breaks again.

So data quality and accuracy are still big issues, and will remain so until we all either move to server-side tracking or we make everyone wear an implant chip in their skull so we can track them properly. That might sound silly, but remember it’s only slightly more work than implementing server-side tracking!

MB: What difficulties do you see when it comes to implementing analytics and how would you assess the overall development of the market?

DB: Well, apart from ITP – which is boring because everyone talks about it – I still think there’s a lot of work to be done in getting the taxonomy and structure of Google Analytics Event Tracking and Custom Dimensions. People are still getting it so frickin’ wrong.

How many times do you see things like Event Category = Click, Event Action = Click and Event Label = Null? And this will be something implemented by a GA Partner as well. It’s utter bullshit, and it’s a problem we should have all solved by now.

Snowplow (now) and App + Web (not yet but soon) support deeper Event syntaxes, and people get excited about moving beyond GA’s three arbitrary Event Tracking layers, but jeez... get it right with three before you move to thirty!

MB: How can an analyst have a greater impact on marketing? How can they be useful for the marketing team?

DB: Don’t report. Prescribe, predict, and pre-empt.

MB: How do you evaluate the current maturity of marketing analytics?

DB: Well, as we were saying a few minutes back, this is still a really early industry.

We think we’re advanced, but we’re still getting Events wrong, and ITP sends us into a tailspin. We’re basing all our work on, essentially, JavaScript putting a text file on a browser.

That’s incredibly flimsy and incredibly top-heavy, with so much reliance being put on numbers that are being generated in a really old-fashioned way.

So I think we’re still a very nascent, early-stage industry with a lot to learn. But with such an open and share-oriented community at its heart, we’re equipping ourselves with the right dynamic to learn whatever we need to. Hopefully.

MB: What do you think is the future of marketing analytics? What trends do you see coming and what’s in demand?

DB: The biggest thing for the next few years will probably be machine learning and artificial intelligence, which is a boring thing to say because that’s all anyone talks about these days. But the technology is within reach for all of us now. Once you have your data in BigQuery, you can load up machine learning processes right in your data table.

That’s huge.

There are people right now that are running prediction algorithms on their sites, using BigQuery data to predict what people are likely to read or purchase based on what other people have done. That’s huge too.

It kind of feels like the entire history of web analytics has been leading up to the point where you have a massive table of data and you throw AI at it to come up with recommendations. That’s the really exciting stuff – doing things like blending store stock data with online behaviour and predicting which shops are going to run out of which products on which day. Mental!

MB: What problems do you see on the market today?

DB: I guess the one big problem, the one gigantic elephant in the room, is that we see news stories about election tampering and populism and ad targeting, and we’re all disgusted but at the end of the day, web analytics is part of the same thing that gives rise to all that horrible stuff. 

I’m not saying that we’re all involved in electing Trump, but we all make a little bit of cash from tracking people on the internet, and we track them so we can understand their behaviour, and we want to understand their behaviour so we can change their behaviour.

As individuals, of course, we don’t want our behaviour to be changed. Who would? So you think, okay, I’ll install Ghostery or use Brave or DuckDuckGo and I won’t be tracked. But then you have a sea of people who still are being tracked, because they’re ignorant about that tracking, which is a morally wobbly place to be.

People have been saying “data is the new oil” for years. But maybe it’s not oil. Maybe “data is the new asbestos”. Think about that. We mine it, transport it, install it, build with it… and it works, it’s fine... but what happens when we realise its toxicity, its’ dangerous nature, its damaging effects on humanity?

Just like asbestos, with data there’s going to be lawsuits, big fines, and if you want it removed, you have to call someone. What if we’re all part of the problem?

I’m not saying we should all quit, nor am I going all tinfoil hat and smoking weed in a basement to hide from the world. We should, though, think very carefully about the work we’re doing and who we’re doing it for. Data Runs Deep has an ethical policy about the sorts of people we will and won’t use this incredibly powerful technology with, and sometimes I wonder if the whole industry needs a sort of code of conduct about this sort of thing.

Thanks, Damion, for such a sincere interview! 

We hope you enjoyed this read and got as great of an inspirational boost as we did. Analytics is a young industry, but it’s being developed by a community of talented and kind people.

Subscribe to our blog to read the next interviews as part of our research into the state of digital analytics and adapt the experience of top analysts to your business.