Technical Copywriting:What Engineers Can Learn from Apple’s M1 MacBook Marketing
quixoticnomad.blog
quixoticnomad.blog
Hmm, I thought this was a cleverly veiled joke. Free of charge no charger, wink wink chuckle.
> However, even without a fan, there is no tradeoff with performance.
There most certainly is a trade off with performance. The Macbook Air throttles after around 10m of extensive work as far as I know. The marketing claims no such thing, either. It just claims that the Macbook Air is thin and fan-less.
> real problem is that most people hate the sound of their laptop’s fan cranking away while they are trying to get work done
This is a bit of a generous interpretation. Instead, I'd argue that Apple thought how they can differentiate, and came up with the idea of reducing fan noise, which can be considered a "cool feature". The fact that their faulty keyboards on the previous 3 (?) generations of Macbooks were insanely loud and obnoxious somehow didn't matter in their quest for silence.
The article strikes me as looking too deep into something it idolizes. Yes, Apple is great at marketing. Can you get away with the same writing if you're a small or a medium-sized company? Depends on who's your target audience (and I'm sure it also depends on your brand).
Funnily enough, the article doesn't mention target audience. It doesn't mention why certain tactics work (or when would they not work). It takes Apple's claims at face value, without deeper thought.
I think a fellow commenter summary of "sell benefits, not features" is fair to the article, as the article doesn't offer that much more.
I don't think you can learn much from this presentation. It's not that special compared to how other companies present performance differences.
"Performance vs. power" is also confusing, speaking as someone who looks at benchmark results for things all the time and who counts as a hardware enthusiast. What is "latest laptop PC chip?" A Wintel laptop on 14nm++? Duh. I don't understand it, and I know someone who doesn't focus on hardware won't understand it either.
The presentation is generally pretty deceptive. There are other much less confusing charts showing relative performance of the M1 in laptop benchmarks that reflect realistic workloads. And this new set of hardware releases has been very well received even among the typically Mac-critical.
The ML slide he's writing about undermines what he's talking about in terms of clarity and 'talking like a human.' It is just gibberish, equivalent to Star Trek nonsense about dilithium crystals transfobulating the warp core. All it means is "has the same features that have been standard for a decade or more on every operating system except a little better, working on more cores than before."
This is not Apple's best marketing work, but it is their best actual laptop product in a while: go figure.
I beg to differ. Other companies can't even be bothered to come up with proper names for their products, and you get the nth "XV356E-52 mark II" style product in their catalogues...
I think this sentence automatically counts you out of the target for this marketing. Most people don't even know what a "mobile chip" is, except maybe something you find in phone. They certainly won't be looking at the M1 and thinking "mobile chip".
You not being in target for the marketing also explains why you find the "Performance vs. power" graph confusing. Again most people don't know what a "Wintel" even is, never mind 14nm++ (is a big number or a small number?). "latest laptop PC chip?" is perfect, exactly because most people have no idea what the latest laptop PC chip is. What they will know is how fast their current laptop "feels", how new it is, and can then extrapolate how fast they think a "latest chip" should be. By extension, they can figure out what Apple is trying to tell them.
> Duh. I don't understand it, and I know someone who doesn't focus on hardware won't understand it either.
Well I know a few people who don't focus on hardware. When talking to them about the new MacBooks they definitely had an idea of how fast a new computer might run, or at least they new it was faster than their current computer. That all you need to understand Apple's graph.
> There are other much less confusing charts showing relative performance of the M1 in laptop benchmarks that reflect realistic workloads
Again depends on your audience. Most people don't really know what a benchmark is (or even workloads) never mind how to interpret them. As someone who in theory know about benchmarks etc, even I struggle to interpret the graphs and it always takes several minutes to get to grips with the data. Even at that point, the only conclusion I come away with is that thing A is 2x faster than thing B. Which is exactly what Apple's graphs show, but without me needing to do mental gymnastics.
> The ML slide he's writing about undermines what he's talking about in terms of clarity and 'talking like a human.' It is just gibberish, equivalent to Star Trek nonsense about dilithium crystals transfobulating the warp core. All it means is "has the same features that have been standard for a decade or more on every operating system except a little better, working on more cores than before."
I don't ML slide is targeted at the general population. Rather it's for people who know a little bit about tech, have heard about this ML magic, maybe read a couple of articles, and the blurb helps them feel a bit smart because they can understand it, then explain it to their less techy friends.
> This is not Apple's best marketing work, but it is their best actual laptop product in a while: go figure.
The laptops are selling like hot cakes. Clearly the marketing works. There's an almost infinite supply of examples showing that building something good just isn't enough for it to sell. You actually need to market it properly. The evidence show that Apple have nailed it for the new Macs.
"Latest Laptop PC Chip" is just deceptive -- and it's confusing, too. How would it look compared to laptops with the Ryzen 4000 mobile series? Not quite as good. They're not obligated to add that, but if a (presumably smaller) business is trying to emulate that marketing strategy, it's not going to work quite so well because smaller businesses have to market in the teeth of skepticism while companies like Apple do not due to their massive brand equity.
Apple has seen a massive boost in sales this year for Mac laptops. You could attribute this to Covid and you could partly attribute it to product improvement. I don't think it has much to do with the marketing, which is just Apple boilerplate.
How to market CPUs in the contemporary marketplace is really hard in general because they are so much less comprehensible than in the era of "more mhz means more faster / 2 cores better than 1 core / 4 core better than 2 core"). For the huge majority of users, it's nearly incomprehensible. It's not an easy thing to communicate about to the normal market. Talking about battery life and general performance is probably the best that you could do. So I'm not saying that this is an easy thing to present in a Powerpoint, just that I didn't think that Mac's sales strength this year has much to do with their marketing and has more to do with the circumstances of 2020 and the strength of the product itself.
I'm curious how you'd describe external mechanical keyboards with cherry reds or blues. "Ludicrously", perhaps? Even my Logitech keyboard was louder.
> The article strikes me as looking too deep into something it idolizes.
Isn't this the author's first Apple product? Or at least his first purchase with less feelings of disdain?
Idolizes...insanely loud. I see your points, but let's not get carried away here.
I imagine there's a counterexample somewhere, but I don't generally see those types of keyboards on thin-and-light class laptops. They're used in big gaming/workstation machines which have other sources of noise than the keyboard!
The point wasn't to say that the keyboard was "insanely loud" in and of itself. The point was to say that to think Apple cares for quiet sound is to disregard the comparatively amazingly loud (and generally faulty) keyboard.
So it's not really the quietness Apple it after. It's the cool factor, a differentiation feature. Otherwise, if it was a genuine desire for quiet environment, the keyboard wouldn't have made sense.
Tangentially, I do own a number of these "ludicrously loud" mechanical keyboards (with various switches). I wouldn't take them with me to a library, or a lecture hall though.
I'm going to hate using a fanless laptop because I know it's throttling all the time...
It absolutely is. Apple has been doing these for years, it's their style.
It now looks like that isn't true, and the M1 really is revolutionary, but that certainly isn't what I took away from Apple's advertising.
Precisely my thinking. However, I don't think we're the target audience for this Apple generation. If my mother saw it, she'd probably think "great, finally something faster than my computer."
Despite the article's praise of the graphs, I think it's a highly risky move that looks highly suspicious, and can pay off only when you know your target audience, and you know you're 100% right.
If I saw a cloud provider, let's say Oracle Cloud, having a similar graph against the leader, like AWS, they'd lose credibility in my mind. Unless I'd read later they were right, in which case I'd think someone in their marketing just brain farted when making the graphs.
That's because you aren't the target audience for the graphs.
The targed audience (the average user) would be confused and bored for with labels, and axis, and statistical minituae.
Things like lots of floating point ability for JavaScript (which has no integer types) or optimizing ref counting (obj-c and swift do this a lot).
I imagine those things wouldn’t impact traditional benchmarks.
If they'd gone for ref-counting benchmarks, that would have shown something more like 5-10x; it's really good at that. The graphs they showed seem to be reflective of performance in general-purpose benchmarks (SpecINT et al).
Let's not play too fast and loose with the distinction between technical writing and ad copy for a technology.
This article wastes times talking around the point: advertisements and sales pitches are not technical writing. Technical writing is about clear and precise communication of complicated and/or detailed ideas. Ads are the opposite, ads are about persuasion. Full stop. End of. If you want to learn ads, learn ads. Do it properly and don't pretend it's something else.
Technical writing can be an input to ad copywriting, but the presence of a graph or chart in an ad does not make it technical writing.
I wanted to only speak about technical copywriting and not technical writing (if you feel like that distinction wasn't clear I'd love to hear your thoughts!). However, I feel like most engineers do use technical writing in their website as technical copy and wanted to use Apple as an example how it can be done properly.
1) Use clear, concise language that your audience will understand.
2) Focus on tangible benefits, not just pure numbers.
3) Have a conversational tone.
However, Apple copywriting isn't meant for those people. Rather, they niched down to people who use their mac for browsing or video editing (which can be pretty resource intensive). Since they niched down to focus on only those kinds of customer archetypes, they knew what to build and made sure they were able to deliver. I know that's tangential to copywriting but for a solid product launch you need a bunch of different parts to go right.
Part of the issue is (for example): Do most people care about how many cores there are? That speaks to your point about benefits vs features.
But the other issue is: Do most people know what a core is?
This is often referred to as dumbing things down, but that's pretty condescending. The person you're "dumbing things down" for may be a chief surgeon at a major hospital, someone who is probably smarter and more accomplished than most of us. They just might not care to spend their mental energy on understanding computer technology, in much the same way that I don't much care to understand biology or medicine.
So when I say it's about "speaking the same language as your audience", I mean speaking in terms that relate to what's important to them. If you're selling to a mass audience, you have to speak in such a way that doesn't assume any expertise or interest in technology. You have to give them a reason to care.
Most people don't care about the thermal performance of their laptop. But they can probably relate to being annoyed by how loud the fan gets or how hot the machine gets on their lap. They can relate to having the battery die, waiting for apps to load, closing apps because the machine can't handle having so many things open, etc.
But you also have to provide justification for those benefits. It's not enough to say "it's quiet" or "never gets hot", because for a sizeable portion of your audience, they need to understand WHY it's quiet. Which means you still need to talk about the features.
All of which is to say that a better axiom than "sell the benefits, not the features", in my arrogant opinion, is "Meet people where they're at."
That's what makes good copywriting so hard. You have to write in such a way that meets your audience where they're at, without actually knowing where they're at, and without being so simplistic that you insult their intelligence.
Also note there are ways to circumvent the "without actually knowing where they're at" problem, that's why marketing funnels have been invented.
That's not true, the only reason technical people can overlook “sell the benefits, not the features” is that people (including, usually, the tech people that overlook it) do care about “artifacts”. Sometimes because they understand the domain well enough to understand the impact on function and benefit from the “artifact” and don't need someone else spinning a narrative about benefit for them, sometimes because the “artifact” itself is a status/aspirational item for them.
The actual lesson of “sell the benefits, not the features” is not “people care about results not artifacts”, but that motivating action in others who aren't inherently motivated to that action without effort on your part requires understanding their individual interests and providing a narrative connecting the action you desire with their interests. It is not a lesson about what people care about, but about what you need to do to get what you want once you understand what they care about.
That doesn't contradict, but actually supports the fact that people don't care about artifacts in themselves.
Either they understand the function/benefit and that's what they're looking for, or they want them for status/aspiration (which is just a special case of the former, since status and aspiration are other functions among many).
> The actual lesson of “sell the benefits, not the features” is not “people care about results not artifacts”, but that motivating action in others who aren't inherently motivated to that action without effort on your part requires understanding their individual interests
Yes, and their interests are the function and the results people get from the artifact.
>It is not a lesson about what people care about, but about what you need to do to get what you want once you understand what they care about.
Erm, yes. "Sell benefits, not features" is a piece of marketing advice telling you what to do to get results (like any piece of advice). That's missing the point.
The question is why does it work. And the reason it works is for the aforementioned reasons: people care about results and functions, not artifacts.
On the other hand, it's very easy to talk about benefits that no one really cares about. Or to talk about them in such a way that they don't see how it applies to them, or why it might be important or impactful in their lives.
So that's why I prefer my way of thinking over "benefits > features". To me, it's a more complete way of looking at it.
With regards to funnels, you're 100% correct of course. But they're not always available, and many different groups of people with different needs, interests, and beliefs can (and do) end up on the same funnel path. You can keep breaking your funnel into every more refined steps, to be as specific as possible, but then you run into the issue of people giving up because you've asked too much of them before asking for the sale.
For example, the landing pages for those M1 Macs. What would you do to make sure you know more about your audience before telling them about the product? Are you going to put a series of questionnaires ahead of the content? The content there has to work for pretty much everybody.
> On the other hand, it's very easy to talk about benefits that no one really cares about. Or to talk about them in such a way that they don't see how it applies to them, or why it might be important or impactful in their lives.
Can't disagree with that, but it seems to me that's a more general point about precision in communication than the benefit/feature dichotomy.
Note that "a benefit no one cares about" is almost by definition not a benefit. If you're talking about an aspect of the product the person you're talking to doesn't find valuable, it's not a benefit. But that doesn't mean you should talk about features. It means you don't understand where they find value.
> With regards to funnels, you're 100% correct of course. But they're not always available, and many different groups of people with different needs, interests, and beliefs can (and do) end up on the same funnel path. You can keep breaking your funnel into every more refined steps, to be as specific as possible, but then you run into the issue of people giving up because you've asked too much of them before asking for the sale.
> For example, the landing pages for those M1 Macs. What would you do to make sure you know more about your audience before telling them about the product? Are you going to put a series of questionnaires ahead of the content? The content there has to work for pretty much everybody.
Sure. In practice, some benefits will work better than others overall, or will account for most of the persuasive power of a given product. Once you've found that, optimizing for subsegments may or may not be the rational choice.
Note that a lot of that stuff done through marketing automation is invisible to a lot of people. Apple, to the best of my knowledge, doesn't do a lot of it, but they could easily target their users based on all the information they collect on them over time without asking for information directly.
I would really love to see, what ideas Apple discarded. Afterwards it seems obvious, the benefits. ;)
“The new MacBook Air. Even we say it has no fans”?
There are very many situations where numbers and details are not helpful, and can even be harmful as they suck away precious attention from your audience. If you're pitching a product to an exec, sometimes they just want to hear you say "this runs faster than anything else!" rather than "this scored 91% on the latest QIRZ-4 benchmark!"
The article asks us to question how we've all been taught that more details == more better.
I recommend watching this take instead. 10 years old, and still clicks: https://www.youtube.com/watch?v=qp0HIF3SfI4
https://www.amazon.com/Obviously-Awesome-Product-Positioning...
I know I kept that article super short and sparse but hopefully I get someone to think more critically about how they are doing their copywriting.
That's why I bought it
1. Sell benefits, not features 2. Speak like your customer 3. Use graphs to stress relative positioning (don't have it be filled with numbers)
I know the article itself is fairly sparse, but this came from a conversation I had with a Senior Dev at the startup I am currently at (where I'm currently a product manager). Our product is fairly technical (check out vectra.ai if you're curious) so we run into this issue of better copywriting often. This article was written for him and I ended up memorializing it in my blog for better visibility. One of the things I didn't mention as in-depth was making sure you understand who your target audience and who will be digesting the content. So, I thought I'd throw in an example below:
We recently rolled out a new feature set that has more robust support for some external connectors so now people in the organization can use usernames and passwords (SASL) to authenticate rather than needing to manage certs (I'm being intentionally vague given that FireEye just had a major breach). In the marketing materials around this feature, we stressed about how this would help the IT organization lower their time spent on support tickets. Why did we stress this? Because the people we are marketing to/selling to are usually CISOs (or at least a SOC manager). All they see at the end of the day is how effectively their IT organization is operating so they care a lot about minimizing the time they spend on frivolous tickets. When we did our initial customer interviews, we found that most organization we work with spend a lot of time managing cert issues when people not in the IT organization suddenly can’t authenticate into something. This was more pronounced during COVID as people are now working remote. The whole reason we even spent the engineering time to ship this out was because of the deeper emotional need to decrease these tickets.
The Dev I was talking about was confused why we weren’t conveying in our materials about the number of external connectors we rolled out or how easily analysts would configure this on the ground floor. It took a while and showing him this article to get through to him what I meant. And honestly, I totally feel for him. During this sprint he put in a lot of effort into getting this to work with a bunch of external connectors. By not giving the specifics he felt like we were not giving him the credit he deserved. If we were selling to security analysts, I’m sure we would have delved into those topics. However, most managers/CISOs just don’t give a shit about that. They care about how its going to help them work better and by proxy how good is this going to make them look in board meetings.
In general, this is the method I use when thinking about how to write good copy around a new feature. What did we build: SASL integration with external connectors
Why did we build it: To eliminate the need for cert management
Why does the customer care: Decreases total support tickets that come from expired certs or other cert issues
Who cares about this problem: CISOs and SOC managers
Why do they care about the problem: They need analysts to stick to working towards goals set by board of directors/upper management. Analysts cannot work towards these goals if most of their time is spent dealing with useless tickets. If they hit the goals, they look good to upper management and if they don’t they could get fired.
After I fill this out I normally only focus on the why’s and have that be reflected in the copy.
I know technical copywriting is hotly debated topic at times (as I can see from this comment thread) but its honestly a lot of fun, although that may be my product manager self talking. I think a few great examples for great copy writing are the following:
https://basecamp.com/ https://hey.com/ https://www.blinkist.com/
If you have more thoughts, I’d love to hear them! Honestly, I might do a more in-depth Technical Marketing 2: Electric Boogaloo blog post if there is a need.