Welcome to our AI writer overlords
And enjoy your 3 week wait for a pizza
Hello Gobbledeers,
First, thanks to everyone who sent me photos of themselves air-frying stuff over the past week. I appreciate it.
This week I wanted to start with a follow up to the vegan dog brunch I discussed a few weeks back. (Sometimes this newsletter is about software marketing, and sometimes it’s about restaurants. Get used to it.)
Restaurant website The Infatuation recently had a piece about new places to eat in New York City and mentioned Chrissy’s Pizza, describing it thusly:
In case you weren’t following all that, I thought it would be helpful to outline the steps:
Find yourself with a desire for pizza
Follow this particular place on Instagram.
Wait for them to release pizza pickup dates.
Turn on your notifications (though they mention this should take place after you wait for them to release pizza pickup dates, I’d suggest moving this to #3 and moving #3 to be #4, as you would like to have your notifications on when they release the pizza pickup dates.
4a) “Slide into their DMs” the moment orders go live.
Wait up to 3 weeks to eat the pizza.
5a) Drag your ass to Bushwick to pickup the cold pizza. Or, if you live in Bushwick, saunter over to pickup the hot pizza.
That’s it!
Also, I need to know the brand of the “off-brand” home oven. Because “home oven” wasn’t quirky enough. Does the brand of the oven matter? If the oven is at 500 degrees to cook the pizza, does it matter if it’s a Viking or a GE or a JB’s Oven-Like Appliance?
I, for one, welcome our new AI writer overlords
I wrote a little while back about Dall-E, the image generation tool that kept me occupied for hours. It was a fascinating and terrifying experience showing what AI can do for art.
Well, as a writer I’m now fascinated and terrified of the new AI-based writing tools that are out there, and I wanted to share a quick demo of one called Lex. You can watch it here (it’s 5 minutes…I’ll wait).
Fascinating, no? (If you were too lazy to watch that, it’s a tool that looks like Google Docs where once you’ve written some content, you can ask it to come up with the next paragraphs based on what you’ve written). You can have your 5 minutes back.
A reader sent me to another product called Jasper.ai that also uses AI for writing, but there’s an interesting difference - Jasper is positioning itself as a replacement for writers (“Create amazing content 10x faster with AI”). In that demo for Lex, I was struck that he’s positioning it as a tool to help writers get unstuck - in other words, it takes what you have (uh, writers) and makes them better.
Typically it’s a more difficult sell to position yourself as a replacement for something people have than it is to position yourself as something that makes what you have even better.
I like where Lex is going - it’s not pretending that AI is good enough on its own to write full pieces - I think most people aren’t comfortable with just letting our robot overlords write their marketing content.
But getting stuck while writing is a common problem for human writers. And helping them get unstuck is a great use of AI (and a great way to gain AI adoption), as we’re not asking it to replace a product, only augment a product we already have.
I tried using Jasper for this column, and I’ll share the results below…
If you’ve used an AI writing tool, I’d love to hear about the experience in the comments.
TikTok’s $1 billion marketing, uh, investment
If you have not subscribed to the newsletter called Margins, written by two gentlemen named Ranjan and Can, you should go do that now.
Margins is a newsletter about digital businesses and economics (and other things) and while they write infrequently, I’m always oddly excited when it arrives in my inbox.
(To get a taste of what I’m talking about, you should read this iconic piece called Doordash and Pizza Arbitrage that explains the economic insanity of delivery businesses.).
All of that aside, they recently wrote about TikTok parent company ByteDance and included this bit:
We do know that armed with cash, ByteDance undertook an advertising campaign that was reported by the Wall Street Journal to be more than $1 billion dollars in 2018. That breaks down to nearly $3 million dollars in advertising daily. This included nearly $300 million on Google advertisements for TikTok in 2018. There were so many ads that you could find threads on Reddit and other places on how to get rid of the TikTok ads users were seeing (and a lot of offline ads too).
For one moment let’s put aside the fact that ByteDance spent $1 billion in one year to advertise TikTok. That’s $1,000,000,000.
I had a conversation recently with someone I know in Softwareville (tm) who said that he was getting pressure from his CEO to “focus more on brand” and “building the top of the funnel.”
He thought about it and went back and proposed bringing in an agency (“a real agency, not one of the ones we use now”) and pulled together a budget for media and told the CEO he was looking at a ballpark of $1.5-$2 million.
The CEO vomited (metaphorically) and asked what could be done for $500,000.
Putting aside the question of what is the “right” amount to spend on brand marketing/awareness building for a company of this size, you are competing for media and your prospects’ attention with the sum total of all other companies out there.
I’m not suggesting you’re (necessarily) competing with TikTok. But I’m saying that whatever you’re spending (and I’m absolutely certain whatever your CEO thinks you should be spending) isn’t anywhere close what it takes to make an impact.
I didn’t think it was possible to spend $1 billion in 12 months on marketing. $20 million every single week. It’s hard to get my head around that. But that’s what it takes to build a brand like TikTok.
So fine, you aren’t TikTok. Would you like to be 1/10th as successful as TikTok? Would you settle for 1/100th as successful as TikTok? Are you prepared to spend $10 million?
Again - I know TikTok is a mass consumer brand. Fine. No matter how you cut it, the somewhere between $0 and $500k that you’re spending (begrudgingly!) on awareness is nowhere near the right amount.
We’re not in an environment where spending with reckless abandon is in vogue (or even possible.) Which will make the next 12-18 months a long, long, frustrating slog for software marketers. You’ll under-invest in awareness/top of funnel, which will make new business sales cycles longer. Buyers are reluctant to plop down cash while they’re hearing about belt-tightening, which will make new business sales cycles longer. You’ve no doubt noticed that 10-20% of your marketing team no longer works there, which means you’re doing less marketing activity overall, so sales cycles will extend. It’s ugly.
Oh, and because of all that you’ll get more and more pressure to generate more and more leads, which you won’t be able to do because of everything above.
Given all that, a modest marketing proposal: do fewer things.
From what I’ve seen, most marketing orgs continue to do 5 (10?) things, but they’re only able to invest 1/3rd of the optimal amount in any of them. (Choosing what NOT to do is hard!) The result is underperformance from every channel.
Belt-tightening doesn’t mean cutting back on your marketing budget - it means cutting back on your marketing budget in some channels and re-allocating some of that to other channels. Sure, your total budget is cut (life sucks sometimes). But cutting channels that work makes little sense.
The difficult decision is getting the team comfortable with not doing something you were doing before - B2B marketers tend to under-invest in a lot of things. It’s time to believe in a couple of key strategies and triple down there.
Jasper.ai would like a (few) word(s) with you…
So I fed the first half of the section above into Jasper.ai (up until and including the paragraph starting with “I’m not suggesting you’re (necessarily) competing with TikTok…” and I wanted to see how Jasper would complete that section. Here ya go:
These arguments don't hold up because they look at ByteDance's spending as an absolute number, rather than as a percentage of revenue or resources. There are many factors that can influence how companies spend their budgets, and the fact that TikTok is spending $1 billion on advertising does not mean that all tech companies should do the same.
One possible argument against spending more on brand marketing is that it might not be efficient or effective in terms of ROI. There may be other channels or strategies that can deliver better results for less money. Additionally, there may be other priorities or opportunities that are more important for your company, and spending on brand marketing might not align with these goals.
Another argument against increasing brand marketing spending is the risk of putting too much emphasis on top-of-funnel activities at the expense of bottom-of-funnel metrics like conversions or sales. If your goal is to grow your user base or increase awareness, then top-of-funnel activities like brand marketing are important. However, if your primary focus is on driving revenue and growing profits, then investing too much in brand marketing could actually hurt your bottom line.
Ultimately, the decision about how to allocate your company's budget depends on a variety of factors, including your goals and priorities as a business. Some companies may find that brand marketing is a crucial investment, while others may find it more effective to focus on other areas. Ultimately, it is up to you to decide what works best for your company's unique situation.
90% of the way there, no? I can absolutely see how that would help a writer who’s stuck and wants a prompt about where to go next. And if I were running content for a software company I’d absolutely dump a bunch of prompts into it to see how many blog posts it can spit out and use humans to modify the language to optimize for SEO. But damn, that’s impressive.
Thank you as always for spending a few minutes with Gobbledy. I love speaking to groups about how to improve their marketing writing - if you’ve got an event coming up where that would be useful, I’m available :). Probably.
That article made me hungry for pizza and funnel cake. And as appealing as it sounds to wait three weeks for pizza, I think I'll call jets and wait 20 minutes.