Danny Gardner

Social Intelligence Lead | Haleon

Inducted 2024

Danny Gardner

What’s the one thing you wish you’d known when you started social listening?

I wish I knew how to communicate its value better. I do now of course but as a junior analyst it wasn’t my job to communicate its value, I was a number cruncher. I wish I was exposed to more use cases when I first started out, it’s something I make sure the people I teach know right off the bat.

 

Have you had any experiences that have made you want to quit? What made you keep going?

I’ve had some long nights and tough conversations, but never enough to make me wanna quit. As good as my team is we still struggle to get the budget we need to take this to the cutting edge, it’s a thorn in my side but not enough to wanna quit. This is a hot space to be in and will be for decades.

 

What role does tech play in your social intelligence process? Where do people contribute?

It’s about 50% of the battle. I think that’s fair, some people will probably think that’s too much, or maybe too little. It’s not lost on me that I do the work of 5-6 people, most of which is due to the scalable nature of the technology. My individual skill helps sure, but without the tech, I’d drown in data, and bad data at that.

People take us to the finish line, which is hotly debated today. ChatGPT and Gemini are both really good for analysis and summarization, they can read more data and just as good as I can; the models struggle with interpretation and what to do with the information, which is where I as a human come in and can’t ever be replaced but the topline stuff is great. No human can read 10,000 posts a day, AI has a huge role to play.

Prescriptive analytics in our field still need work, I haven’t seen a tool (licensed or open source) that can tell you EXACTLY what to do with the data. Gives us something to work towards for sure.

 

Who have you seen as a mentor in your career?

Love this one, I reserve some of my hardest questions and challenges for my mentors, especially those in the social listening space.

Carrie Stern who has my role at Google I’ve looked up to for many years, she’s an excellent tactician, strategist and friend. She’s one of my favorite soundboards and people in this space.

Brian Wright (who’s at Coca-Cola now but I met when he was at Wells Fargo) is an excellent advisor of mine. He’s been in the space a long time and his wisdom shows. He always knows something new that I don’t.

Aynat Ravin is another great mentor of mine, and fellow University of Maryland alum 😊. She’s run the social listening center at Marriott for years and helps me regularly expand my horizons of what’s possible.

Todd Grossman the former Americas CEO at Talkwalker has been a great friend and advisor. He’s shed so much light on the industry’s history for me and has championed me and my work for years.

I’m very lucky to have these folks (and others) in my corner.

 

Most embarrassing mistake you made in a social listening project?

Haha I was talking about this the other day! A couple years ago I was building our dashboard page for Lamisil (a brand we sold but when we owned treated athlete’s foot, a fungal infection) and never checked the media wall widget we have (where it was on the page and what it was pulling in).

I happened to show that page to someone in a walkthrough and grossed out them and myself when it was the first widget that popped up. We get porn in our data too (you’d be amazed) and I’m pretty sure I’ve forgotten to turn that data off once or twice in a readout.

Needless to say, we didn’t have a media wall for Lamisil anymore. Might’ve deleted the query for a while too LOL. Check your data people, don’t do what I did.

 

How do you see the future of social listening evolving?

I touched upon this in the other interview, but video intelligence is huge, has been and has yet to peak. It probably won’t for the next 5-10 years, but ever since I started, I’ve always had video as a media format living rent free in my head.

I just ran my end of year portfolio report and video usage across our portfolio (i.e., UGC) of queries is up 200%+, with static image down about 50%. I’ve always known this but now I have the data to back it, and we as an industry lack the technical ability to do this. Especially at scale.

I also think commerce will be a big part of social listening, not your traditional listening but more social analytics. Two of the biggest trends in IT are social media and commerce, if people aren’t buying things online then they’re looking for things to buy. This is very clearly the future, and these two huge datasets are converging.

Text-based listening will always be relevant, unstructured text is everywhere, and it’s arguably the hardest data type to analyze. It’s important that we get really good at this (which I think as an industry we are!).

 

What’s the most useful data source? Are there any you find useless? Why?

Tough question, Twitter and Reddit complement each other very well, and on top of that they’re typically the two biggest sources. Is there some skew, yeah maybe, but you can’t argue the fact that they both are two halves of a whole.

Blogs are useless, the blogs data I find in tools today is just bad. Most bloggers have diversified into new apps, Pinterest, Snapchat, Instagram, TikTok, etc. Tumblr data I find useless too. I met someone last year that loved it who I’m blanking on, and they had a good use case for it too, for their business.

Twitch is cool, our one tool gets them, but I don’t get a lot of value out of it just yet.

 

How have you been able to win over ney-sayers throughout your career?

I’ll debate the really stubborn people I meet but generally speaking most people understand it’s value, they get that analyzing social media for any of the 10-15 major enterprise use cases we have is valuable. And that we can do it at a relatively low fixed cost.

I’ve always been a vocal champion for this data in our company, and only recently on LinkedIn where I’ve built a following saying the same thing. That goes a long way, it’s added even more credibility to the years of showing people behind the curtains.

And now you just have to point to the biggest viral moments in social over the last couple years, they are all the result of basic but passive ears-to-the-ground social listening. Stanley, AirBnB, North Face, they weren’t the most technical social listening cases (far from it), but it brought great visibility to its importance.

This is less a concern for me than it’s ever been. Companies want to be social-led brands so bad and social listening is the ultimate hack.

Get Social with SILab