Hey friend – Rob here.

I keep going back and forth between how much i pack into this newsletter.

And then i realise, i fucking hate long newsletters, so who am i to produce one?

Especially when i get feedback like this:

No, thank you Brian, this truly does mean a lot.

So yeah, remember kids: always write the things you'd want to read.

💌 This week on Salmon Theory:

• The problem with urging positivity
• Reduce variables before enhancing them
• The identity crisis of (pure) trend-led brands
• AI hallucinations as a killer feature
• How Heinz's red hot work is working

Let's get into it.

​

​

There is a great lesson i've learned in therapy, and more recently from child psychologists.

It's not very helpful to say:

You know what's helpful instead?

Why?

Because "it's ok" means whatever you're going through is what you need to go through.

Feeling pain is ok because it's your body giving you signals about something that matters.

It's a useful reminder next time we think we need to urge others to be positive because we need them to.

The problem?

This doesn't boost their resilience, it simply denies their reality (to paraphrase Adam Grant).

Which is one heckuva way to get someone to repress what they feel and having this come back to bite them later.

And if that happens with someone you lead, then chances are, that experience will come back to bite you too.

Normalise people's emotional reality, before trying to change it just because it benefits your own.

(Also a tremendous note to self, of course.)

​

The greatest breakthroughs i've ever seen in my life have been the result of reducing variables, not adding them.

• The strategy project where we had one framework too many, which was creating unnecessary confusion
• The creative brief where we were trying to be so clever we forgot one good thing is more valuable than three
• The anxiety episode where i was trying to do too much to have to do less later (rarely works that way)
• The depressive funk where i was putting so much focus on my failings i'd forgotten how much i'd grown
• The chronic overwhelm where i thought i could keep up with hundreds of newsletters every single week

Coaching is teaching me this as well, around how to best design my business.

(Did i mention you should hire Dana, and yes she also works with men 👋?)

That every time i want to change something in my life, the answer is rarely what else i can slot in.

It's much more likely that i need to take things away, often more things that i thought were necessary.

Yes, we can have it all, but not at the same time.

In that sense, you don't need to be Rick Rubin to appreciate the importance of being a "reducer".

Reducing variables sounds like a less ambitious move, but all it does is let you enhance the ones that survive.

Darwin in 2024 self-employment-while-being-a-friend-to-yourself narrative, baby.

​

Imagine someone only did what others thought were cool and popular at the time.

You'd say they're having an identity crisis.

They don't know who they are, and therefore only define their value based on what others value at that time.

And yet, we normalise this as being the 'proper way' for every brand to behave in a social media environment.

Being purely trend-led feels interesting until you realise it also may not be sustainable for 90+% of businesses.

'Sustainable' being the core word here, because yes it's hella sexy but if it's short lived, then you crash and burn.

By all means, be all over the latest trends if you think that's your most effective path to commercial growth.

But be careful of the sacrifices you need to make in your team's sanity and brand equity to get there.

​

A while ago, Contagious published one heckuva meta analysis, but the real kicker is this bit:

Pair this with the previous note if you should feel so inclined.

Two short reflections on it:

1. There is immense power in recognising a popular thing just wouldn't fit your ways, and so you don't do it
2. As we worry about the great skills commoditisation, one great benefit of AI may be its hallucination rate

This last point matters not because of what a hallucination rate does to the trustworthiness of how an AI 'thinks'.

But because of what it does to the level of forced randomness and divergence in how we think.

Remember, neurodivergence has short term and long term brand and business effects too.

​

Cannes Lions, yay!

But did any of the work actually, er, work?

Well, this is why my favourite section of their entire affair is the Creative Effectiveness Lions.

So, i spent some time dissecting the winners and bring to you les super petit summaries over the next few weeks.

​Let's start with Heinz, shall we?​

​

​