graph of vanity projects

Is B2B advertising a vanity project

I don’t think it’s anything to do with AI. But everything to do with ambitious people wanting to leave their mark on an organisation.

Here’s what my ‘vanity project’ would be if I ran a B2B advertising team. Scrap most of the bland clickbait adverts that most B2B brands use – and focus on excellent creative and brand building.

graph of vanity projects

Don’t track the clicks, the metrics which are more likely to be bots than humans. Use your deep instinct and experience to design eye catching advertising that talks to brand values, what consumers want or need to know and get rid of the rest. You probably won’t miss it. And you may lose your job.

But in 5 years time, I bet your advert will still be remembered by the key B2B decision makers….. because only 5% of your target audience are in buying mode at any one time. Use your advertising to build a strong brand; rely on SEO and website pages to backfill their research needs before the RFP.

 

Inspiration for this post from the Uncensored CMO interview with Tom Goodwin

Email subject lines can make or break open rates

I like to write to deliberately appeal to a small portion of the audience. The preview and subject line are the first, top-level filter available to email marketers.

I realised this after a client did an event which specified the age participants should be. It was a roaring success – because people knew immediately if it was right for them (or not).

The downstream benefits

  • audience segments can be built based on responses, opens, replies and clicks
  • deliverability improves driven by open rates

Remember, good email marketing software allows you to edit the preview text. This is a secondary audience filter and it does not have to be the same text as your opening sentences.

#1 B2B marketing skill for 2025

Building your audience is THE key marketing skill which will help your marketing strategy in 2025.

The “gurus” say one thing..

Let me explain.

Not all online marketing ‘teachers’ are wrong. But their methods paid dividends to them because they were early adopters.
– Most people teaching blogging started blogging between 2004 and 2008.
– Most people teaching content marketing started between 2008 and 2012.
– Most people teaching podcasting started between 2010 and 2014.

My first blog was 2006; first content marketing website was 2009 and my podcast was 2013.

These methods worked then because they were early adopters and because we marketers (and Google/Meta) had not yet started to algorithmically enshittify the platforms we used for natural search advertising and social discovery.

What works now?
Audience building – and direct email marketing and direct response copywriting.

And yes, I can do that. And you should know how to do it too.

H/t to Brian Clark for the bullet lists.

Agentic AI is a joy-killer for NBD

A “cold call battle”

I love a contest as much as any athlete… but this makes my blood run cold.

It’s an advert to promote agentic AI compeing alongside CSRs to book appointments.

You may not know this, but I’m a demon cold caller. It’s a service I provide as part of B2B Marketing – yet the underlying premise of this service is going to drive enshi*tification into yet another part of business marketing.

I know I can spot many online fakes, bots and robo-voices.

Soon we won’t be able to do this.

Agentic AI cold calling bot service

It will be ‘fun’ to try to get the automatic agents off the line quickly – but where’s the joy in beating a machine? It’s man-to-man mortal combat that gets my adrenaline surging.

That’s only a sorta-joke. I do like a competition. I do like to pit my skills and my wits against another person. But only in a fair fight.

I’m sure that these services will shortly be incorporating human-like aspects of speech and language to persuade the listener that they’re not talking to a bot.

Yet again – where’s the fun in that.

Does agentic AI risk taking all the fun out of doing business? Your thoughts….

Here be dragons

The allure of the speed-to-answer is great for early users of AI. Those of us who have been longer in this particular swimming pool Realise that critical assessment still remains the most important input of the human actor.

1960 Porsche – Dragons of a bygone age. Image Credit: Rebecca Caroe

As Neil Perkin quotes John V Willshire

Where technical debt for an organisation is

“the implied cost of additional work in the future resulting from choosing an expedient solution over a more robust one”
Cognitive debt is where you forgo the thinking in order just to get the answers, but have no real idea of why the answers are what they are.’

This in part reveals why enterprise finds it so hard to incorporate AI use into daily workflows. You must have the human input in order to ensure that you’re not making a fool of yourself in the rush to deliver speedily.

Anyone fancy a brown bag lunch to share experiences?

New ideas often receive pushback

New ideas often receive pushback.

That’s because they’re new, they represent change and that can be threatening to jobs, budgets and ‘the way we work round here’.

It doesn’t mean they are necessarily a bad idea. I find that frequently it takes a couple of goes to overcome the pushback and also it takes some time – maybe a year or more.

There’s nothing more powerful than “an idea whose time has come”. But that doesn’t just happen. The marketing communications outreach that builds the comprehension, that challenges the status quo, that offers alternative pathways is where us B2B marketers can prepare the ground so that you overcome the pushback, get acceptance for a pilot or a trial and then move towards adoption.

Here’s a case study.

I’m the MC for a B2B marketing careers event at the Marketing Association which I first proposed as an un-conference a year ago. Now the event is re-framed and is happening next week.

Let’s talk?

Not everything is enhanced with AI.

Use your experience of life, of human communication and decision making to work out times in the customer journey when personal, individual, manual actions will have greater effect.

I love using letters, phone calls, SMS – all great tools for use cases when you need cut through and to effect human-to-human contact as a context for making a sale.

PM me and I’ll tell you one use case I’m doing right now with a CV for a job application.

No it won’t do the whole job for you, yes you will use your human insight and creativity to frame up the AI supplied text. It’s so satisfying.

Getting a “no” is a good result for the sales pipeline

The inexperienced new business development marketer will disagree with that statement. Why wouldn’t you want more prospects in your pipeline?

The customer doesn’t owe you their business – but they do owe you an answer.

Creating marketing tactics that provoke the prospect to answer “no” is a good thing. Sales know not to pursue them and can focus their efforts elsewhere.

And remember, no is just for now.

95% of B2B prospects aren’t ready to buy at any one time. So going back to them later, may mean they’re ready to hear your brand message and review their plans.

Rebecca Caroe and Sam Irvine, two people on screen discussing AI

Marketing with AI – a warning

Sam Irvine is CEO of Copyright Licensing New Zealand – as a B2B marketer he understands the use cases for AI and marketing. There are two main ones, you can use it internally to manage processes as organisational AI and secondly for external outputs like campaigns and marketing material creation.

Good and Bad AI Actors

Sam acknowledges the time saving benefits of AI and he warns that many LLM training sets were created with scraped data taken without permission. He notes some legal cases in USA and Europe currently challenging this – Books3 which has 200,000 books pirated from the internet has been taken down after a  law suit.

The days of tech startups seeking forgiveness not permission are over.

New Zealand has a cross-party political group looking at AI right now but there aren’t yet any legal cases being heard.

Sam’s advice to marketers

Be careful about which tools you choose to use – are they ethical and responsible?

Nvida and Adobe have licensed all the images they use from Shutterstock and other photo libraries. The original creator is paid when you use their AI generated images.

A closed AI is one developed internally and trained on the organisation’s own data.

An open AI uses public data.

Three risks of using open AI

  1. Your input data will be added to the training set. Is this confidential information that you are authorised to put into the public domain? Sam warns against adding client data into an open AI system.
  2. The output you get could include copyright material taken from the LLM training set which could put you into the courts.
  3. You might get an answer which is wrong. Be careful around fact checking those outputs.

As the data sets are being withdrawn, the information used by open AI models is now getting older and less relevant. It’s incumbent on all of us marketers to use AI in the right way and embrace it while protecting ourselves and our clients.

Copyright NZ has courses on copyright, contracts and agreements including AI. www.copyright.co.nz

Watch Rebecca’s interview with Sam Irvine