Strategy Sessions Podcast | Britney Muller

Britney is Marketing Manager at Hugging Face, the AI community building the future. She’s an SEO expert, machine learning whiz and amazing conference speaker.

Episode 11 Season 2

Listen // Watch // Download

Find this episode in all your usual podcast places: Spotify | Apple | Google

In this episode we discuss:

Hugging Face… what is it and where did the name come from

The difference between Artificial Intelligence (AI) and Machine Learning (ML)

Why people don’t need to be scared of ML

Why people should be scared of ML!

What’s the main goal of the marketing at Hugging Face

Why you should always prioritise by impact in your marketing activity

What it’s like working at Moz

The fun of being a Search Scientist and watching the SEO community take your work to new levels

Learning lessons from running your own marketing company

Why imposter syndrome is your friend when it comes to public speaking

TOPTIPS for aspiring conference speakers

Surviving lockdown with a pet snake!

Entrainment theory and why you’ll end up going down a YouTube hole because of it!

Britney Muller

Britney Muller is a Marketing Manager at Hugging Face, SEO Consultant, and proud Aunt. Britney was previously the Senior SEO Scientist at Moz and founder of Pryde Marketing, a boutique medical marketing agency.

Find Britney on Twitter

Book Recommendations

The Data Skeptic Podcast

Machine learning with python

Natural Language Processing with Transformers – Building Language Applications with Hugging Face by Lewis Tunstall, Leandro von Werra and Thomas Wolf

Digital Marketing Strategy Course

My Digital Marketing Strategy Course in partnership with the University of Vaasa in Finland is available now via Teachable for just €249.

It’s perfect for small business owners, entrepreneurs and those who want to get a better understanding of what marketing strategy is and how to embed that strategy across an organisation.

Andi Jarvis

If you have any questions or want to talk about anything that was discussed in the show, the best place to get me is on Twitter or LinkedIn.

If you don’t get the podcast emailed to you (and a monthly newsletter) you can sign up for it on the Eximo Marketing website.

Make sure you subscribe to get the podcast every fortnight and if you enjoyed the show, please give it a 5* rating.

Andi Jarvis, Eximo Marketing.

Interview Transcript

This transcript has been done automagically using Happy Scribe and hasn’t been checked by a real person, so there may be some hilarious mistakes where the AI can’t work out our accents – I’m sure they’re trained on just the American accent.

Eyup and welcome to the Strategy Sessions.

My name is Andi Jarvis, and this is episode 11.

Today. My guest is Britney Muller.

Britney, how are you doing?

Good. How are you doing?

I’m great, thank you.

We’re smiling, we’re laughing.

I’ve cocked up the beginning of this

podcast several times already, so I’m just going to let Britney introduce yourself.

Britney, you work for Hugging Face, which is possibly the company

with the greatest name we’ve had on the podcast so far.

So tell everyone who doesn’t know, who are you?

What is Hugging Face and where does that great name come from?

Yeah, that’s a great question.

Andi just quick side note, this isn’t what my voice usually sounds like.

I’m battling a bit of a cold, so bear with me.

You’ve been doing karaoke for the last nights

I’ve been practicing now that the pandemic is finally letting up, getting ready.

Yeah. So Hugging Face is incredible.

Previous to working there, I had been a huge fan of their different

libraries and the things that they provide.

So essentially,

if you know anything about open AI, which was created by Elon Musk,

and it was this mission to make AI and machine learning models open

and readily available to the public that was very quickly closed and monetized.

Unfortunately, Hugging Face is sort

of a bit of the opposite of that in terms of it’s really on a mission to democratize

machine learning and continue to keep some of these large, state of the art models

open source and available to the community.

And that’s really what it’s revolving around anyway.

It’s this unbelievable community of machine learning engineers

and hobbyists that really dedicate their time and energy and resources to building

data sets and models and incredible things.

And then the way that we are able

to continue that effort is also through some of our paid solutions.

So we currently offer things like access to some of our experts in the field

through an expert acceleration program that’s really incredible.

Andi huge companies everywhere are using

it to accelerate their machine learning roadmap.

And then we also have things like Inference API to speed up the actual

compression Andi optimization of these large models.

And then we serve private model hub.

And that’s essentially

we’re essentially on a mission to become the GitHub of machine learning.

And through doing that,

we’re offering things like Private versions of that as well.

So that’s kind of Hugging Face

in a nutshell, the way the name came about, this is so funny.

So I recently was in New York meeting some

team members Andi had a chance to talk to all three founders at once at a bar.

And we were all having fun and hanging out.

And I finally got to ask the three

of them, could you share your story of why you called it Hugging Face?

And my impression was always that it was

a play off of these large machine learning models.

There was a phase of Parsi Mcpars Face and Bodie McBoat Face.

All these engineers thought it was just

funny to name these models ridiculous things that we would have to say out loud.

Andi it’s not that case at all for Hugging Face.

It was this situation where Clement was saying that he had this idea

Andi was brainstorming it for a while, and he was so confident that

this company would go public that he was already thinking of the stock symbol.

And he thought, what a cool

kind of moment in history to have the first emoji as a ticker symbol instead

of an actual, like three or four lettered symbol.

And so that’s how Hugging Face came to be.

I suppose I probably shouldn’t be surprised that somebody who’s working

right at the forefront of machine learning and AI is already thinking 18 steps ahead.

That shouldn’t be a surprise that someone’s got that falls out.

But I love that. Think big, right?

Aim for the moon. That’s where we got to go.

Fantastic. Well, look for the idiots who are

listening today, and I include myself in that.

Right. So artificial intelligence, AI,

machine learning, ML, what’s the difference between those two?

Because when I read things,

I’ve seen them used, and I feel like they used interchangeably,

but I suspect that they’re very different things.

Can you kind of get us all up to level

zero so we all know what we’re talking about from here on, it.

Yeah.

So they are often used interchangeably, but the difference is that artificial

intelligence is the large umbrella under which machine learning sits.

And then within machine learning, you have things like deep learning,

reoccurring neural networks, you have different types of machine learning.

But AI as a whole,

it’s a bit of a controversy, actually, because there’s still this old school

of thought that we have yet to reach artificial intelligence.

There’s the whole principle of the Turing Test that hasn’t yet been.

Well, arguably not been accomplished yet

stating that we haven’t yet reached that level of general artificial intelligence.

So, yeah, it’s an interesting topic.

You see, it used all the time in websites and kind of a buzzword.

But in my opinion, it should be used a bit

more carefully and we should be leaning more so on the term machine learning.

There’s a previous guest on the podcast, J.

Castle, who wrote a great piece about the language of marketing being so imprecise.

Now, granted, we’re kind of outside

the scheme of marketing a little bit here, but he was making the point that when I

say a marketing strategy Andi you see a marketing strategy,

we could mean two very different things compared to he has a legal background,

every term in law has exactly the same meaning.

And every legal person, every lawyer,

every advocate knows exactly what that term means.

Marketing is not like that. It’s quite varied.

And I think you’re kind of speaking

to the same thing that there’s probably a great debate which might feel a bit neck

Beardy about has artificial intelligence been reached yet.

But it’s actually a really important

debate, especially when you get idiot marketers who like to throw artificial

intelligence into a solution that isn’t to make it sound cleverer.

Would that be kind of is that one

of the things that you guys look at over coffee sometimes at Hooking Fish?

You’re like, they’ve set up artificial

intelligence on this website and no, it’s just a prediction algorithm.

Is that the sort of argument that goes around your place?

I haven’t heard that just yet, but I did just have a very fascinating

long conversation with Margaret Mitchell, who started the ethical AI board at Google

and brought up I’m going to say her name wrong, Timmy iBrew.

I think I’m butchering that.

I need to work on that.

So she has all this incredible experience with recognizing bias in AI.

Andi one of the key takeaways I took from that conversation is we need to get

better about the lexicons, about the language we’re using.

We need to all be on the same page because inclusion Andi diversity are two totally

different things, and we use them incorrectly all the time.

And I think the same principle can really be applied to machine

learning and marketing Andi the way that we refer to these different systems.

We need to do a better job just to improve our level of communication.

Well, you’ve touched on it there, and I want to jump into this question.

This is a very positive podcast.

I’m not trying to catch you, I promise you.

But there are lots of people worried about

the rise of machine learning and scared about it.

Andi there’s all sorts of

fairly semi well publicized if you’re in the right twist of threads and Reddit

threads and things about when machine learning goes wrong.

So with picking black faces open, doing it incorrectly.

Or I think you’d mentioned in a recent

blog post how when you search for jobs for men, it pulls up some sort of like

politician and soldier Andi fairly high ranking things.

And when you put women in, it throws out cleaner, nurse, prostitute or something.

Andi this is an example of kind of

am I right in thinking, though, that the machine isn’t biased?

It’s the data set it’s learning from that is biased,

Andi therefore we need to try and work out how we unpick that.

Is that the general discussion that’s going on in the industry at the minute?

Exactly.

So it’s important to be aware of garbage in, garbage out.

It really comes down to that training data.

But something I’ve learned just as of recently is all data is biased.

And unfortunately,

even in recognizing that, we’re then applying our own values

and belief systems on what we think should take precedent over other things.

I think the bottom line of that is ensuring that individuals aren’t hurt or

marginalized or discriminated against through these systems.

And we need to do a better job of machine learning engineers and anyone that’s

deploying these models to have a better way to recognize the discrimination.

I sat in on a really fascinating talk by

Lynn Sue, who works at Microsoft, on different

bias,

making sure there’s ethical AI involved at Microsoft,

and her examples of recognizing discrimination were just fascinating.

So I think we have come a long way

in terms of having some benchmarks and ways to go about it.

There’s tons of room for improvement, obviously, but we do have to be careful.

One thing that I also learned recently

at Hugging Face is Margaret Mitchell’s earlier work at Microsoft.

She was doing lots of vision AI for helping blind and visually impaired

individuals see things through different applications.

And she fed a model.

They were working on an image of a really large, horrific explosion.

And the model’s output was beautiful.

Stunning, lovely. Wow.

Because it had never seen an image like that before.

Right.

It had seen sunsets, it had seen fireworks,

and it had no previous information about anything about it.

We don’t tend to take pictures of horrific

things as often as we do really great things.

And so

her thoughts on that or that was just one click away,

if that was connected to a system that could continue to create that chaos,

thinking it was beautiful and wonderful, like how terrifying that actually is.

Andi let me tell you, I used to laugh at the question,

will AI become dangerous?

And just even this last year, the more conversations I have with some

of the leading experts in this field, the more fear I have going forward

in that this used to be a question where I would get laughs and funny responses.

And now all of these individuals I’m

currently talking to that are at the head of this cutting edge technology.

They really take pause, and then

they’ll explain their fears and their

concerns because there is rising concerns as it gets connected to more and more

entities and more things that the risk of harm will increase.

Andi that, to me, gives me a real

positive outlook on this, because I think when new technologies like

this come along and even things like social media when it first came along,

the Champions of these technologies are often people at the forefront of it

who are so passionate about the benefits that they are blinded to the concerns

that other people might have or the things that could go wrong.

So when the people are at the Vanguard of this are really aware of the issues,

that gives me hope that we’re going to be able to spin back into and go.

Right. Okay.

Let’s put the right checks and balances in place, because I think to go back

to your example of the photograph, my understanding is that the data sets

for this sort of thing to work probably have to be huge.

So by the law of big numbers means

that when you have a lot of the pictures or lots of words about things,

you’re going to get better results, more accurate results.

We don’t have as many. The results aren’t going to be as great.

So things like explosions or edge cases or even small countries.

So I think of a country like New Zealand,

which is 4 million people, punches above its weight globally.

But actually the data set of, say,

faces that you can train on of people from a New Zealand heritage is actually

tiny because it’s only a country of 4 million.

There’s probably more people within an hour’s drive of where you sit right now

than there is in all of New Zealand.

So it’s a really interesting thing of not understanding.

How do we focus on the small things as

well as the big things to the fact that the people who are behind this are

worrying about it gives me hope for the future.

I love that perspective so much Andi could not agree more.

We’ve got the things we’re worried about.

But what are the big positives in the takeaway?

Because people aren’t just going

into machine learning Andi AI because it’s fun and there’s

something to do and there’s some investment dollars going around.

People are in this because there’s great

things that can be done to improve life in society.

So what are some of the cases where

machine learning can really help us and improve what we do?

Yeah, there’s so many really exciting spaces.

A couple that I get really excited about

is in the medical space, the ability to more accurately identify

disease and problems early Andi help save people’s lives is already happening today.

There are really powerful applications

Andi Pharmaceuticals Andi identifying new drugs that haven’t been used for different

applications that could be really successful.

Some other spaces.

I know there’s a group

at the Allen Institute working on wildlife protection in terms of they’re

using machine learning on drones to identify poachers

on elephant sanctuaries and in elephant areas in Africa.

How cool is that?

Like, what amazing?

Yes, I love that application so much.

And then also they’re using underwater sonar and machine learning to monitor the

endangered population of whales off the Pacific Northwest Coast.

And they’re able to follow and through

the sounds are able to monitor the health and kind of swoop in with different

resources to help protect the set of whales.

There’s different crazy kind of rumors about

old cell phones being used in weatherproof

boxes that are put way up high in the rainforest.

And they’re using just the microphone

on the cell phones to identify illegal deforestization.

So anytime, like a chainsaw is picked up, it alerts local authorities and they can

stop this process of illegal chopping down of trees.

I mean, there are tons Andi tons of really cool applications like that.

This must be like a really old cell phone, like a Nokia 3310 or something like that.

The ones that you could charge once and the battery will last for a month.

You couldn’t do it with an old Apple device, could you?

I mean, those things last about 3 hours when they’re not plugged in to charge.

One of the things we’ll go and check,

we will go and check in their own forest because I’ve never been Andi want to go.

But moving on, moving on.

Otherwise we’re going to get stuck here with me talking about all tech forever.

This is a marketing podcast and you are the marketing manager for Hogging Face.

So you’ve got all this amazing technology

behind you and all these great things happening.

But what are the key objectives for you as marketing manager there?

Would you wake up in the morning and go,

Right, I need to smash this out of the park.

Yeah.

So there’s a bit of a dissident happening

to be totally candid between the things that we’re a marketing team of two.

First of all, so we’re quite limited in the things we’re able to do.

That being said, it’s such a small team

at Hugging Base that we can move really quickly.

So between the two of us, we’re able to execute projects on our own

relatively quickly Andi have a lot of fun in that space.

But there’s this other side of the company

where we really also want to use marketing to leverage our paid solutions.

A lot of people aren’t even aware that Hugging Face offers these paid solutions.

So as a marketer, instead of just focusing

on like fun SEO stuff, I sort of have to put on different

marketing hats and think, oh shit, how do I do that?

Right? In a way that also

cultivates awareness and qualified leads for these paid solutions.

And that’s a much, much harder thing to accomplish.

I would love to sit at my desk all day and come up with really fun content

marketing ideas that use our tools in a way that we create virality through.

Here’s what we learned analyzing the top

thousand plus machine learning and engineer job posts,

or here’s the Twitter conversation by region taking place on X.

Like you can do really cool things with the models we have.

I know my marketing lead, Federico,

was the cofounder of Monkey Learn and over there they would apply natural language

processing to their content marketing in a fascinating ways.

Like he applied natural language processing to Twitter information about

hotel reviews and he isolated it to regions.

And people in France most often complained about croissants.

Andi the French do love a croissant.

They do love it.

That was the biggest negative sentiment.

And then I think it was Bangladesh,

it was cockroaches, and it was very regional specific

and fascinating to apply that information in a way that I think

people are so interested in. Right.

We’re interested in timely events, we’re interested in geospecific information.

And I think marketers as a whole have not

taken advantage of that and had fun with that as they could.

I recently pitched an idea hugging face that they immediately told me no.

And I did in my spare time where I have

this theory that we can learn a lot about Craigslist free listings.

I saw you did this. Yeah.

Okay. So stupid.

But it was something I had so much fun doing.

And I knew people would get a good laugh

of learning about their city and what’s the top unique free listed item.

And it really does reflect some of the local cultural perceptions that we have.

So things like that I think are so much fun.

So the last guest I had on the show were two guys from a company called STATSports.

They do performance monitoring for professional athletes.

So when you see sports,

usually soccer players, they whip the shirt off when they score.

And they wearing this vest with a performance tracker in the back.

So these guys based in Ireland,

they make this tracker and they’re working with Major League Soccer,

a lot of baseball teams, a lot of football teams, Premier League soccer team.

They work with kind of if you name the ten biggest sports organizations in the world,

they’re probably working with eight of them.

So they’ve got a whole their business

makes money in the B to B space, right, by signing these companies up.

But they also started selling in the last couple of years to consumers.

So you’re an amateur sports person like myself.

You want to track your data and you can

compare yourself to the pros as well, how fast you run.

Andi we talk to them about, look,

how do you do this BTC stuff with the B to BB together?

Andi look, they’re in a sexy,

fun industry, but they didn’t really see a difference.

They were like, look,

this stuff helps drive this stuff and this stuff helps drive this strikes me when

you’re talking about the two parts of your role.

You’ve got all this fun interest in sort of almost consumery type marketing.

But the bit that pays the bills is the business to business type marketing.

And it feels like there’s maybe a little bit of tension between the two.

So what are you doing to square the circle?

Are you kind of going to do a little bit of this or are you keeping them separate?

What’s your thought process?

I think it’s very difficult.

I think that’s sort of directional.

It needs to take direction from the top.

So what is the North Star?

What is it that we’re trying to accomplish this quarter and this year?

And how can our efforts impact that?

So for us right now, we are focused on monetization.

We are focused on figuring out some of our paid solutions.

So growing awareness, growing SEO has taken a backseat to that.

I always have stuff on my desk just to keep this stuff in mind.

I always prioritize by impact.

I don’t know if you

constantly drilled in my brain and I think it’s important.

I think employees across the board, you need that North Star.

You need something by which you can prioritize what you’re working

on on a daily basis and making sure that you’re driving that impact.

So that’s something I think about regularly.

I love the sticky note on your desk.

I’m a big believe when I work

in companies, you create the strategy with them.

I’m like, what do we do with this now?

It’s like stick it on the wall next to everybody’s desk, everyone’s machine.

They can see it. Right?

What’s the point in hiding the strategy?

You hear this sometimes?

What if somebody leaves and takes it to a competitor?

Like, I don’t care, right? Yeah.

They can have the bit of paper if you

like, but they don’t have all these great things.

And people here driving it forward.

So share it, publish it online, let people know where you’re going.

We’re not quite there yet.

Some companies are a bit still like, we don’t know about that.

Andi.

But no, look, stick it there every day you come in, prioritize my impact.

I love that line.

I’m going to pinch it and pretend it was mine.

It’s so important.

Yeah. So what sort of tactics are you using them? When you’re thinking

prioritized by impact, can you maybe tell us through, like,

one thing or what sort of tactics do you have in mind for the next quarter?

Yeah. So it’s interesting.

So we’re still such an early stage startup that my opinion is that we really need

to get a bunch of experiments under our belt.

We really need to figure out what sticks

in terms of distribution, strategy, content.

How are we creating these qualified leads?

Who are the key stakeholders that even

make the decisions to purchase one of our paid solutions?

What does that look like?

What are those titles?

We’ve done tons of research with LinkedIn.

We’re creating content to target those individuals.

We’re testing out different distribution strategies.

And then ultimately

we need to have a point where we take a step back and we evaluate.

Okay.

Throughout this process, what has worked well and what hasn’t.

How can we kill our darlings and focus on the things that have driven impact?

Yeah. Kill your darlings.

You have to. Yeah.

It’s great advice for marketing,

and we’re going to talk about conference speaking soon as well.

And it is great advice for conference speaking as well.

Isn’t it like just kill you darlings?

So we’ll come back to that in a moment. Okay.

So you mentioned the S word during that as well.

You mentioned science because one of your

previous roles was the senior SEO scientist at Moss.

a company I’ve got a lot of respect for.

I know and love a few people who work or have worked there.

you’re not just an SEO from senior SEO scientists what’s the job there?

What did you do? What tests were you running?

What took us to do that?

Oh, my gosh, it was so much fun.

You got to give Maz props.

They give so much autonomy to their employees and trust for you to kind of run

wild with what you feel will be interesting research or valuable webinars.

Andi so through the Smith team which was myself, Dr.

Pete and the late Russ Jones I still just can’t believe

we had so much fun coming up with,

you know, this is what the community has been talking about lately.

We see these trends with our data.

This could be really valuable to put into some sort of research that we then deliver

Andi are basically a catalyst of that information to the incredible community.

And then the cool part about all that is

the community really takes it to the next level.

So like I said we sort of served as the catalyst of deeper research

and information from the data we had access to and then the unbelievable SEO

community continues to bring all of those things and level it up,

asking the right questions, applying it in more interesting ways, creating tools.

I mean that’s kind of the beauty of the industry as a whole.

There’s a word for it which I can’t remember at the minute but the way

that the community builds on things is just fantastic.

We all stand on the shoulders of. Don’t we?

We’ve done something and then jump on that and somebody builds on it.

There’s a real collaborative, collegiate spirit.

Most of the time in the SEO,

there is sometimes a bond fight between the community Andi throwing things at each

other, but we’ll we’ll worry about that another time.

You had some great times,

which is so we’ve had Rand on the show like a thousand years ago now,

and Rand talks about his exit from there and some of the things.

But as a company, it kind of grew, maybe evolved a little bit and changed,

but it’s still one of the leading SEO tools out there, right?

Yeah. But you loved your time there.

Running search science tested.

I suppose what I’m looking at here is from an SEO side of things.

Was it just the greatest place to go and learn?

Being there surrounded by Dr. Pete, first of all.

But everybody working there feels like they’re just an amazing set of people.

Honestly, yes.

And also the fact that,

again, that aspect of trust and it felt like such a safe place to make mistakes,

which I think is so essential in innovation in and of itself,

is having that safety net of go break things, go learn.

Andi so the fact that I was able to test different things on Maz.com and had

the freedom to experiment and break featured snippets that we had and figure

out what would break them and what wouldn’t and kind of reverse engineer

the whole SEO process was so incredibly fun, and it didn’t feel like work when

you’re doing things like that that you’re genuinely interested in.

And I’ve sort of had the mentality from the start of my SEO career of this is

an industry that you can really have fun with in terms of gamifying it in my head.

There’s always a game going

on with different search terms and information and how to hack this or

how to improve the Mark up here and feed it to Google so they’ll add your schema

markup or maybe honor a featured snippet better and all these things.

So, yeah, I feel incredibly grateful to Maz and all the employees there that I

got the privilege of working with to learn Andi just have fun in that space.

Brilliant stuff.

Andi there’s definitely a lesson there in making mistakes.

I mean, the whole point of science really is not knowing if she knows something is

going to succeed or, you know, something’s going to work.

That’s not science.

Setting a hypothesis and testing it, which by its very nature should lead

to failure, is what science is about, at least in my eyes.

Anyway,

if they let me make the description of science, that’s what it is.

But you’ve also worked freelance as well,

or you’ve set up your own consultancies and agencies before

different world,

moving from working in a team the size of Mars to working on your own.

How did you find that transition I really loved it.

I feel incredibly

lucky and grateful to be in a position where I got the privilege of choosing some

of the clients I got to work with moving out of Maz and they think I’m lying and I

can’t say who they are, but there’s one particular client.

They are my favorite team in the world to work with.

They are part of a large company that everyone has heard of and their

ability to learn, receive feedback, experiment on different items with me

has been so incredible in the growth that we’ve seen in our time working together.

As far as traffic and monetary numbers, it is just outrageous.

I’ve never been a part of something

that large before and it feels really good to apply all of my previous knowledge

and experience to something a bit bigger and through a consulting framework.

It’s been a nice step back for me to not

be full time SEO in house and just kind of have that experience as well.

There is something quite lovely about

being able to pick and choose the people you work with, isn’t it?

The All Blacks rugby team have this part of their policy called no Dickheads,

so it basically means Jim Collins maybe uses a moral problem.

You got to have the right people on the bus, that sort of approach.

But I have a bullshit detector that when I

go for sales meetings with people who wants to work for me.

I’m like, this isn’t feeling rightaway.

You can feel it.

You learn it early.

I don’t want to maybe want to call it freelancing.

You set up a consultancy,

but that was the second time you’ve done that before you went to Mars.

You worked or you set up an agency as well?

Correct.

Do you feel like you’ve always got the entrepreneurial itch you need

to scratch? Do you always feel like it was something

that keeps coming round as you get older and older and older?

Totally. It’s always there.

I learned so much starting that agency.

I was so young when I started it and at the peak I had seven full time

employees and we were doing very well in the medical space.

But to this day I have a lot of guilt around how I operated it.

I wish I would have been a better manager.

I wish I would have been a better leader.

There are so many things I would go back

and change, but ultimately, through my experience moving forward,

I realized where my strengths are now more than ever.

And I don’t necessarily love managing people.

It’s not a strong suit of mine.

I’m better suited in a role that can get

in the weeds and get a little scrappy and provide things here and there.

So I think learning what you love to do and what you don’t necessarily love to do

is just as important as anything in the growth of your career.

Absolutely.

The phrase I use is a building phrase.

But I like being on the tools.

I love being in working with clients,

digging, finding the problem, working out how we’re going to fix it.

I love that there’s two of us at XML.

I’m in a sort of a growth phase at the moment.

But it’s something I’m really aware of, that some people love it.

And just like, the sooner I can stop doing

this and work and then I can run the company, the better.

The sooner I can get big enough to bring somebody in to run the company.

So I can just go back to doing the thing I love.

That’s what I want to do.

Really helpful.

One thing, I suppose, you mentioned how young you were when you started that.

And there’s two people I’ve met ever who look the same age when I first met them.

And when I met them five years afterwards, it was you and Ross Simmons.

Like Ross Simmons has a portrait of himself in his attic.

Right.

That man has looked the same age for 30 years.

He doesn’t age. He’s about 195, and he never changes.

He’s no wonder he’s so good at marketing.

He’s been doing it since before marketing

was invented. But yeah, Ross is like, he does not change, does he?

No, he does not.

Drink from the same cup of eternal life or whatever as he has

anyway. Right.

So the reason I know that you haven’t

changed in a number of years is we met a number of years ago in Dublin.

You were keynote in Learningbound.

And this is turning some awful psychophantic kind of fanboy thing.

But you are one of my top three or four conference speakers of all time.

Right. I love watching you speak.

It’s just interesting.

It’s a show.

It’s educational and it’s one

of the things I mean, I always think when I speak

at conferences, you’ve got to entertain people as well as educate them,

because you can educate people all you like if you bought them to tears.

Nobody is listening anymore.

And you put on a show. Right.

Stacey Mcnard is another one. She puts on a show.

I absolutely love that. Will Reynolds.

I mean, Will could read me the phone book and I would be like, keep talking, Will.

I love it.

Andi you’re in that. You’re up there with me.

conference speaking, you enjoy it, don’t you?

You get a kick from it. I do.

I enjoy the people.

I enjoy connecting.

And again, sort of serving as a catalyst

to provide different and new information to individuals to help them then level up

the industry and level up the work that they’re doing.

Yeah.

Now

you also share advice with people as well

about dealing with nerves and imposter syndrome and things like that,

because there are a lot of people who maybe want to do conference speaking or

even want to come on as a guest on a podcast.

The number of people I asked to come on said, Will you be a guest?

And they’re like, I don’t know if I’d have anything to say.

Like, no, I’ve seen you writing, I’ve seen what your work does.

Come on Andi talk about it.

Andi like, no, nobody would be interested.

What sort of advice do you give to people about that?

I think that’s a very normal thought process.

I think we’re all human and we all have this fallacy of we have no idea what we’re

doing, and we aren’t as experienced as we think we are

Andi think it’s just important to realize that we’re all on the same page with that.

No one really knows what they’re doing.

There’s just a level of getting more

comfortable in the speaking space to talk about the things that you’re comfortable

and confident talking about and have experience in.

And quite frankly, everyone has things to talk about.

Everyone has unique marketing stories and case studies and experiences

that would be valuable to be shared amongst others.

So I think to give yourself a little bit more credit and to just go practice.

I believe Toastmasters are all around the world.

They’re such a great and easy free way to dip your toes in.

And I also say this as someone who was

initially so terrified of public speaking that I was so embarrassing.

Any, I would literally be walking to Toastmasters when I was living

in Denver, Colorado, Andi would get to the building Andi would just keep walking.

I would be sweating.

I’d be like, no, I’m not.

Today

it gets easier. It gets better.

Like, you just have to that’s the hardest part is just getting started.

And once you realize, oh, this is a fight or flight response, that’s not necessary.

These people aren’t Lions that are kind of come eat me.

It’s going to be okay.

Yeah.

I’m in the speakers group for Brighton SEO.

I’m keynote in first day this year.

Oh, it’s amazing.

Listen, I’m sorry, did I mention that

I’m really excited and also absolutely terrified, right?

Yeah. Andi

come across as quite a confidence.

In fact, I am a confident speaker when it comes to being on stage.

I make sure I know my material Andi get across well.

Andi there’s other people in the speakers

group have maybe not done things, but have maybe seen me speak before as well.

But maybe this is their first gig.

Andi they’re like, So I’m talking in this

group about being nervous and being petrified.

And some people are like, you’re not nervous, are you?

Like, absolutely terrified.

Before I go on stage, my hands are going.

My palms are sweaty.

I can feel I always get the vomit on his sweater already mum’s spaghetti.

I’m like, he’s nervous.

I’m like, my heart is going.

But once you’ve done it once and you realize that you didn’t die when you were

on stage doing that thing, you start to realize, I didn’t die.

That time. So I’m not going to die this time.

And actually, that’s just my body’s way

of making sure I know my shit before I go on stage.

So I start to feel that like we’re a month

out and I can already start to think I need to go back to my slide.

Andi those nerves are just my body’s way of telling me, go back to your slides.

Make sure you know them.

Make sure you know your material. Make sure you know what you do.

You got it. You got it.

You got it. And once I know that I can have

that rational conversation with myself stage side when the little one in my right

ear starts to go Andi you need to run away, run away quick,

the guy on my left shoulder is like, going, Nana, you know this, you got this.

You know you got it. You did it yesterday.

You did it the day before. You practiced in the bathrooms earlier,

you were talking to yourself on the way here.

You got this.

But realizing you’re not going to die on stage is probably set number one.

Isn’t it enough to that? So I’m rumbling about this number.

Oh, it’s so good.

Yeah, that’s such a common experience.

And I think the coolest part about what

you just said is that whole notion of you feel those nerves so far in advance

and then leading up as they get stronger and stronger,

that physiological experience is identical to arousal or excitement.

It’s been proven.

So what I was told by a speaking coach when I started at Maz was,

you need to just tell yourself every time you think I’m nervous.

I’m nervous, I’m excited.

I am excited.

Just kind of like, slap your thoughts and say, no, that’s excitement.

I am just really excited to share this information.

It’s going to be great.

Andi so you have to kind of talk yourself

down and also to kind of get out of that headspace where once you get

carried away with your thoughts, you’re not as in tune or in your body.

You’re less grounded, which takes away from your speaking.

Andi so something I work really hard on is kind of grounding myself before a talk.

I’ll wiggle all my toes in my shoes Andi mentally think ten toes down,

and I try to feel every toe, and it just kind of grounds.

You

can’t believe it. Yeah.

Weird.

I thought everyone had twelve like me, right?

I know. It’s special.

Let’s go back to something you said

earlier as well, just before we wrap up about kill your darlings.

Andi now you’re presenting now about

machine learning Andi fairly technical heavies use this stuff.

Andi what you do really well, from what I’ve seen, is

you make it accessible, but you also keep the story to what you’re doing.

You can kind of bring people on a journey.

Any idiot can present technical data and bore you to death.

Not everyone can present technical information and bring it to life.

So how do you tackle the kill your darlings?

Because I know I struggle with this.

I have great jokes in mind.

I’m like, this is the funniest joke ever, and I just know it has to come out.

I’m like, it just doesn’t work here.

But how do you face that fact?

So everything in terms of speaking

revolves around the audience, so they come first.

Before anything, I put my ego aside, my humble brags in my talk.

I will kill those so fast if I don’t think

it’s serving the larger purpose of these people came, and they’re providing two

of their most valuable resources, their time and their attention.

And I take that incredibly seriously.

So from the start of framing up a talk,

I’m constantly thinking about how will this be immediately applicable to the work

that they’re doing and accessible in a way that helps them.

Andi so I try it really hard to kind

of craft that story around, inspiring them to take that action.

And then oftentimes I get a lot

of my improvements and feedback from the audience after the talk.

That’s the part I miss the most Andi is just like connecting with people.

When we had a beer, that was so much fun.

I missed that stuff because that’s when

I learned more about what people are working on, what they need,

how my work can better serve that larger purpose.

Brilliant.

I’m sick to death at virtual conferences and I’ve been doing them,

I’ve been presenting at them, but I’m so sick of them.

So being back at Brighton in person is just like,

just before we wrap up

one question quickly,

before I ask you about books and recommendations, you’ve got a pet snake?

Yes. Tell us about the snake.

Briefly.

She’s the best.

Her name is Pumpkin.

She’s a little cranky right now,

which means she’s just quiet and hiding in her hide.

But, yeah, she’s a ball Python.

She’s about four and a half feet.

She’s very overweight.

She’s almost £6 and she should be like two.

So she’s on a bit of a diet.

Is that down to somebody over feeding her, perhaps.

She’s just so happy Andi she’s so happy.

We’re all happy when we’re eating, Brittany.

We are all happy.

I know this is the best.

I honestly don’t think I could have gone through the pandemic in 2020 without her.

We had so much fun.

Yeah, she’s one of my best buds. Brilliant stuff.

Brilliant.

Well, look, before I let you go, two things.

Number one, tell me about books, podcasts, newsletters, any resources that you would

recommend to people listening to the podcast.

I’ve really been working hard to kind of increase my knowledge in statistics,

data science, machine learning, all of that stuff.

And some of the things that I really enjoy are the data skeptic podcast, where

complex data science Andi machine learning

concepts are described in such simple digestible ways that just from listening.

I mean, it’s incredible that even some

statistical information like that you can get from listening to a podcast

has helped me so much in my journey to learn more about the field.

Highly recommend that there’s so many great books out there

on Python, Pandas machine learning with Pythons, one of my favorites.

My fellow coworkers at Hugging Face just released

a new book about Transformers in NLP and it’s absolutely fantastic.

It’s the one book where reading it, I feel less stupid being like,

oh thank God this information wasn’t available before where I was so confused

about different things with Transformers architecture.

And they finally described it in a way where I’m like, I get it,

I get why padding is necessary in a model, I finally get it.

So things like that,

yes, things like that are being provided in better and better ways.

There’s so many great resources online.

I’m personally working on different accessible material for marketers.

I don’t know that I’ll be able to launch

it on Hugging Face, but I think I’ll launch it regardless just

to help people getting more in tune with Pandas.

I think there are really fascinating ways,

honestly, to get insights from Google search console data.

And just through this more advanced

process, without even realizing it, you will learn about statistics.

You will learn about data science.

So I’m working on a course that would do that so super excited about things like

that in which it’s applicable in nature immediately valuable.

And people are also like, oh Holy shit,

I just learned how to import data into a Colab notebook or a Jupiter

notebook and how to clean out that data very quickly and how to do this.

So I think that will become more and more available to the marketing and SEO

community as a whole, helping to really kind of level up our field.

Because I don’t think we have enough statistical knowledge.

The amount of data that we work

with and the way that we work with it, it’s a little frightening that we have

gotten this far without a stronger hold on statistics distribution.

So I think.

There’S a huge opportunity for all of us

to it’s a huge bug bear of mine that marketers, especially content marketers.

I’m calling them out, they’ll do research and then make these

ridiculous claims extrapolating data that just does not you cannot extrapolate

your survey of twelve people to Twitter poll.

And we’ve decided that you’re like, no man, you didn’t

do this.

But there are charlatans.

I worked on a project, I’m going to have to tiptoe around all of the names.

But a very big agency based in London had used Twitter data and then extrapolated

that and said, this is what this country is like.

I can’t really get any more specific than that.

And I just remember sitting reading the presentation going, man, this is shit.

And then I found out how much they paid for this.

Andi anyway,

the fuzzy logic needs to die.

It’s very bad.

I’m just as guilty of it.

I’ve done stuff like that in the past

and I think we should all be on a mission to get better.

We need to be better. We need to be better.

Excellent.

And lastly, then one last question for you.

What one question do you usually get asked that I haven’t asked you today.

Okay. You know what my favorite question is Andi

will ask this at parties, happy hour on podcast.

It’s one of my favorite questions and I think it’s such a simple question.

But the quickest way to better get to know someone.

And the question is, what have you been interested in lately?

And it could be anything.

It could be a project, a theory, a book, a podcast, whatever.

And to just give people the freedom to really think about in their spare time.

What does that look like?

I have so much fun asking that question

and getting interesting feedback from people.

And the topic I’ve been so stuck on for like months now is a little embarrassing.

It’s this principle called Entrainment.

And it’s this insane

phenomenon that occurs when objects in motion are near each other.

So this was first discovered in a room

of great grandfather clocks back in the 18 hundreds.

And a man in the room was going to lunch and on his way out,

he swung all the pendulums randomly as he walked out the door.

And he came back like 30 minutes later.

And they’re all in sync, right? No.

Yes.

There’s YouTube videos that I will watch just fascinated.

But now I have all developed all these theories around entrainment in that we now

know women close to each other sync up their cycles.

People, when you’re with a loved one,

oftentimes your heartbeat will sync up as you talk to each other.

Andi so I think this is just another huge

case for be so careful with who you surround yourself with, right?

You are the average of the five people closest to you.

And to choose those people wisely because their thoughts, their feelings,

their emotions, their energy get entrained with you.

And so it’s something I think about way too often.

But I have so much fun, like looking into it.

Brittany, that is hands down.

Hands down, dammit.

The best answer to that question I have had on this podcast.

This is incredible. Shit.

I’m going off to start reading now.

You’ve ruined my weekend, right?

I read about this all weekend.

I’m also going to watch Irobot,

the Will Smith movie, because in there where the robots all

converge in the corner of where they are being stored.

And there’s just these bits like, why did they do that?

Yeah, I know the name for Entrainment.

They even train soldiers not to walk in unison when going over a bridge because

it will collapse bridges if they walk together.

There’s like stories of bridges coming down and killing soldiers.

So they’re taught, like, don’t walk in the same pattern.

Isn’t that crazy?

Literally, I’m going to get to blow my mind,

emoji out and bang that’s what’s going on with the email with this episode.

What have you been interested in?

What have I been interested in lately?

Anything.

I’m turning the world’s most boring man at the minute for various reasons.

Andi so I’ve been interested in storytelling again Because

cards on the table. Brighton is my first in person event

that I’ll be speaking up for, like, everybody else, two and a half years.

Usually if I was doing something that big, I would probably do it.

I would have the presentation tested

in a couple of different, smaller events first.

That’s not happening.

So I’ve gone back to the basics of storytelling, Looking at how

the different types of story Greek philosophy.

Greeks would say that there’s only,

I think, four different types of story tragedy, comedy, love story.

Andi what’s the other one? I can’t remember.

But I’ve been looking at how they build

stories and what the overlaps are and how they all stick together

and trying to find those elements and actually then sort of saying, okay,

well, how do you build that into a marketing presentation?

What are the bits Andi why do they use those mechanics?

And how do you use that across the 30

minutes keynote to keep people with you to bring them along with you?

Because I’m trying to get away from talking

at people And I want to bring them into the presentation.

For audience participation is difficult and raise your hand if no.

So I have elements where I want to bring

people in and kind of breach that wall, but I want to do it through storytelling,

so I’m looking at that and how it stitches together.

That’s what I’ve been interested in.

Oh, that’s so cool.

Now I feel like I’m a guest on mullercast or whatever your podcast is.

Honestly, the hands down everyone should be.

If you listen to the podcast and you

coming on soon, you need to aspire to be this good, right?

This is how good you need to be on board.

Brinny, thank you.

Thank you very much for your time.

This has been truly amazing. Thank you very much.

Andi Michelle, oh, sorry.

Yeah.

How do people get in touch with if they want to talk to you?

That’s a good question.

Probably most active on Twitter

can connect it to me on LinkedIn, but probably Twitter.

There’s a link in the show

Strategy Sessions Podcast with Britney Muller

Listen // Watch // Download

Britney Muller

Book Recommendations

Other Links

Digital Marketing Strategy Course

Andi Jarvis

Interview Transcript