Mind-spam: do you really need to be at the top of the list?

This morning I received yet another email from a Search Engine Optimisation (SEO) company, offering to get my website to the ‘top of the list’. I’m not generally entirely sure what they mean by ‘top of the list’. is on the first page of Google for Martin Turner, but the bass player of Wishbone Ash is likely to be at the top of that list, except when I’m fighting elections, during which, for about six months, this site rises to the top. If you Google ‘how to write a sword fight scene’, then you will find me at the top of the list. On the other hand, if you Google ‘branding’, my company, Brand Motor Limited, will not be anywhere near the top of the list.

Should it be?

If you are an SEO company, absolutely. I am losing customers right now by the fact that my site isn’t at the top. But am I? And what would I do if I were? Actually, only one commercial company is on  Google’s first page for branding, and, given that it isn’t Wolff Olins or Interbrand, I suspect that some SEO jiggery-pokery has been taking place. Certainly, looking at that particular website (and it may all have changed by the time you Google it, so draw no conclusions), it isn’t exactly what you would call ‘top tier’, though they seem like nice enough people.

I spoke recently with a prospective client who told me that when they started up, they always tried to appear bigger than they were. Now, they are looking to appear only as big as they really are. I like that client: they have mastered the fundamental principle of branding—a brand is not a logo, but a promise, and the strength of your brand is based on your ability to fulfil that promise every single time you do business. Appear bigger than you are, and you will one day disappoint someone.

My first really big client came as a result of word of mouth—by far the best networking tool—and resulted, before we pulled the trigger on the deal, in me sitting with the CEO. His crunch question was: “Is it just you, or are there a string of recent graduates who’ll be doing the work?” As a new start-up business, it would have been terribly tempting to bluster and claim to be bigger than I was. “It’s just me,” I said. He told me how relieved he was. Every business that has hired consultants knows the danger of talking to someone who seems just right at the pitch, and then discovering that they are talking to a string of new-hires for the actual work. We did the deal, and three years later, won brand of the year for that particular industry.

This isn’t to say that a one-person start-up shouldn’t appear completely professional. If you have twenty-five years experience in your field, there is no reason to act otherwise. But, like enormous companies trying to appear much more home-made than they are, trying to be something you’re not only works for a while.

Which brings us back to mind-spam.

I had a call this morning from another client who was worried because their email to me had bounced back. For some reason it had been rejected by SpamCop. I don’t use SpamCop, and neither (she thought) did they. Somewhere along the internet email trail, though, someone did, and SpamCop didn’t like the look of their IP address, or mine, or one of the intermediate servers, or my email address, or theirs, or someone else it was cc-ed to, or the content of the email.

This is something that happens increasingly. I have nothing against SpamCop, and it is just one of the many similar systems that causes bounce backs. For a couple of years, the emails of one of the most trustworthy people I have ever met bounced back because he didn’t have his own name in addition to his email address in his settings, and the spam filters didn’t like the look of his email address. Why, I don’t know. In the end, I had to turn the spam filters off.

However, the work of these spam-filterers is entirely altruistic. They are desperately trying to hold back the tide of phishing, promotional, scam, virus, trojan and inappropriately bulk emails which flood the net every day.

I was at the Liberal Democrat conference this week, and happened to eat my lunch next to Mark Pack. Afterwards, we spent half an hour discussing some survey data. I sent him the link, and he may well use it. Or not. If he does, that data will have far more currency than it ever would on my site. Mark Pack is a top commentator on all things Liberal Democrat. He deserves to be. He writes regular, well-judged pieces that provide information which is otherwise very hard to get.

The SEO people would no doubt like to promise that they can get me above Mark Pack on the Google rankings. Perhaps they could. I don’t intend to invest however many hundred dollars it takes to find out. The thing is, though, if I did appear above Mark Pack, it would just be mind spam. I only occasionally write on political issues. I don’t have access to the data that he does, and I don’t put the work in to analyse it. Perhaps I should, a personal brand advisor might say. But should I? Even if I did do all the things Mark does (assuming I were capable of it), we would then simply have two Mark Packs, competing. If we were selling chocolate bars, this might be good for the market. But we aren’t.

In the mean time, if I paid my $500 to the SEO company, all I would be doing would be making Google rankings less useful. On martin you will find much more about literature, broader notions of art, and the internet’s leading article on writing sword-fight scenes. However, if you were looking for the things Mark writes about, these would not interest you. You would feel cheated, and you would have been. Mark Pack is the best Mark Pack there is.

It’s not just on the internet that we mind-spam.

I’ve been to very many networking meetings where well-meaning graphic designers have stood up and explained that they do branding. I’ve then, when it was my turn, had to do intellectual somersaults to explain what I do without actually saying ‘a brand is not the same as a visual identity’. Graphic design is a crucial element in most brands, and there are some very talented designers who specialise in brand design. However, saying that branding is a kind of graphic design is simply misleading people who are probably already confused.

We all do it, though. When challenged, we talk ourselves up. A logo designer becomes a ‘brand designer’. Amateur musicians described themselves as ‘semi-pro’. People put the word ‘a leading’ in front of everything. The word ‘strategic’ gets used as an adjective meaning ‘particularly good’.

This week I met someone who told me she was a concert pianist. She gave me her card, and I quickly Googled her. She was (and is) indeed a concert pianist, leading exponent of four composers, and performer on Radio 3’s CD of the Week about a year ago. I was absolutely thrilled later to discover that she is married to one of my old college friends. We plan to meet up.

The thing is, if it wasn’t for all the mind-spam that we infest ourselves with these days, I would never have needed to check. People who really are in the top echelon of their profession lose the credibility they deserve because of all the people who aspire to be that, and talk themselves up. As it happens, you don’t see much of that in classical music, for the very simple reason that classical musicians are forced to put in such enormous amounts of practice and face so much rejection that people learn not to speak more highly of themselves than they ought. Plus the fact that, no matter how much someone talks themselves up, other musicians can immediately tell if they really aren’t that good.

If you are an SEO company reading this article, you can save yourself the bother of emailing me. I won’t be buying.

On the other hand, if you, like me, are one of tens of thousands of professionals who live by their reputations, let us agree together to fight against mind-spam not merely on the internet, but also in the way we describe ourselves.

It’s time to call our shovels ‘shovels’.

Google has changed its logo… or has it?

Google has changed its logo… or has it?

Google’s new logo comes hot on the heels of Tokyo’s logo debacle. It’s a good news story that represents the evolution of a successful brand, done in the opposite way and for the opposite reasons as Tokyo’s. That’s a good thing.

But is this a rebrand, as some websites describe it, a new logo, or merely a new typeface?

Let us explore.

First, the typeface. It’s called Product Sans, and it’s been developed exclusively for Google. Expect to see knock-off versions appearing on free font websites from next week. But is it really a new typeface? Those with the 1988 Letraset book (you know who you are) have probably already gone back to see how much it differs from Avant Garde Gothic Medium, with its character-set alternates that don’t usually make their way into computer fonts.

Avant Garde Gothic in the 1988 Letraset book

Avant Garde Gothic in the 1988 Letraset book

Well, you can see that they aren’t entirely identical, but it’s evident where Google is coming from on this one. I entirely applaud this. Avant Garde enjoyed a spell of popularity in the 1990s with the rise of Postscript printers, but it tended to get used in the wrong places. It is a beautiful font, one of the best of the geometrics, and it is a fine thing that its heritage is now in the Google logo.

But what of the logo itself?

Just quickly Google ‘evolution of Google logo’, will you? Seriously, you owe it to yourself to do this via Google rather than taking my word for it.

What you’ll see is that from 1999 onwards, the logo has been entirely consistent. The early 1998 version had a green initial G. The 1998-99 version had an exclamation mark. From the 1999 version, the colours have stayed the same, only the typeface and treatment have changed.

As Daniel Kahneman pointed out in Thinking Fast and Slow, what we see first is the colour of a logo, then its shape, and finally its content. Distinctive colour is hugely important, which is why Cadbury was not happy when a competitor wanted to use purple to sell its own chocolates.

As a general rule, we discourage multi-coloured logos. Apple’s original logo was multi-coloured, and there were Apple devices which may have cost more to emboss and print that logo on than to construct the functional part of the device. Now the world’s biggest brand, Apple has opted for a purely monochrome device. Very nice.

The reason for avoiding multi-coloured logos, though, is mainly to do with consistent reproduction and efficient reproduction. Every colour on the original Apple logo was a separate print process. A multi-coloured logo means you always have to print in CMYK separations, or print four separate Pantone plates. When printing in CMYK, the colours are never quite right anyway.

The exception, of course, is for a logo which exists only in the online world. Yes, you do occasionally see ‘Google’ printed, but you know that the ‘real’ Google logo is the one on your screen. Not only can it be as many colours as Google likes, but it’s more appropriate if it is, especially for a search engine dedicated to producing the most varied results possible, based on your query.

Logo designers (quite correctly) like to fix the form of the logo absolutely, even down to the white space surrounding it. When an organisation receives its branding instructions, they come with stern warnings not to rotate, squeeze, distort, layer, shadow or do anything else with it that is not explicitly allowed in the Corporate Identity Guide.

Google has never played by those rules. The common points are: always the word Google (it was originally called ‘Backrub, but we’ll forget about that), and always, since 1998, in Blue-Red-Yellow-Blue-Green-Red. For the rest, Google plays with its logo on an almost daily basis. Christmas versions, Halloween versions, commemoration of a famous person versions, and much more.

The core of Google’s logo is the distinctive name, and the colours. That’s it. Unless brand-fascists have taken over, we will probably continue to see the playful transformations to suit the days and seasons.

So, rebrand, new logo, or just a tweak?

For my money, it’s a tweak. The Google logo is the name and the colours. I would be willing to accept that it genuinely is a new logo, though, especially to avoid a fight (someone threatened to defriend my on Facebook this week just because he didn’t like my views on spacing).

A rebrand, though? No, not at all. The essence of the Google brand remains the same: the easiest and most trusted search-engine, coupled with a variety of useful and interesting extensions, not all of which are guaranteed to work. Google-plus? Never really caught on. Google Maps? Indispensable to many. Google Mail? Lifeblood for some, anathema to others. Google Hangouts? To be honest, I’m not even really sure what they are, though I do remember trying them. Googling is an officially accepted word (if you think the Oxford English Dictionary makes things official). The Google name is inextricably linked with ‘to Google’. Not bad for something which began as a spelling mistake.

Your brand is a combination of what you promise, what you deliver, and how you present it. We’ve seen a minor change to Google’s presentation: a different typeface, but less different from the ones it tries out during a typical year. What it promises (successful Googling) and what it delivers (well, the same as that) are unchanged. It is the same brand. And long may it be so.

In the meantime, let us continue to reflect on the fact that almost all great brands simplify their logos as time goes on. Food for thought for anyone contemplating making theirs more complicated.


The strange case of the Tokyo logo

Tokyo has abandoned its 2020 Olympic logo, following complaints that it was plagiarised from the Théâtre de Liège.

According to news sources, the Tokyo Olympics team initially brushed off claims from the Belgian designer Olivier Debie on the grounds that it wasn’t a registered trade mark. Whether Debie pursues them in court now is something we will see. However, the result is that the Théâtre de Liège is now world-famous, and the Tokyo Olympics is ever more deeply mired.

The whole thing came to a head when it turned out the Kenjiro Sano, the Japanese designer, had been accused of plagiarism before. At that point the Tokyo Olympic Committee evidently lost its nerve.

Probably a good thing.

If you look at the two logos the first thing that becomes clear is that the Théâtre de Liège design makes a good deal of sense. It incorporates the T and the L in a distinctive way, and you can even find the D if you look at the negative space. Whether it has quite what it takes to become famous without the controversy is another matter. By contrast, the Tokyo logo makes no sense at all: it is the T L of Théâtre de Liège, with an additional circle added to symbolise Japan, and the colours changed. But why? What relevance does the L have to the Tokyo Olympics?

The Tokyo Olympic Committee should have rejected the logo long before there were any doubts about its provenance. It just didn’t make any sense. So, how is it that a multi-billion of whatever currency project didn’t spot that its logo was silly?

This is a problem I face as a brand consultant. Prospective clients quite often show me their existing logo, and ask what I think of it. Generally that means that they recognise there is a problem. (Let me say now that a logo is not a brand, and a brand is not a logo!) How, though, did they approve it in the first place, if it never looked right?

As a general rule, I would suggest that the initial brand of a new business is never quite right. It takes three years or £.25 million in turnover before most businesses really get the idea of why their clients buy from them. When talking to established businesses, I’m generally very interested in their customer feedback: why do the customers love them, why do they keep coming back, what is memorable about the business?

If you’re a 0-3 year old business, please relax. Your brand and logo could probably do with some tweaking, but that isn’t your fault. It’s inevitable.

Rather more worryingly, there are plenty of businesses out there which went for an expensive rebrand, did all their new signage and stationery, and only subsequently realised it wasn’t quite right.

This isn’t necessarily a huge problem. If you look at the Shell logo over the last few decades, you will see that it gradually evolved as it was simplified. Simple is almost always better.

The very worst situation is the one Tokyo finds itself in: after a rebrand, a company discovers that it has the same logo as a competitor.

How can this be?

To paraphrase William Goldman’s comments about Hollywood, very few people commissioning logos have much idea about what they are looking for. The process often goes something like this. A company wants a new logo, and they tender out to a number of design agencies. The brief is something on the lines of ‘to construct a new logo for [name of company], a world leader in our chosen field’. Rather fewer designers than they were hoping for respond to the tender. At the tender interviews, one company asks a lot of questions and explains the process they will go through. Another company shows three logos they’ve designed. In most cases, the business chooses the company offering three logos.

Having sat on both sides of these kinds of tenders, let me say this: never take the company offering three logos. Almost certainly two of them will be reworkings of logos previously rejected by other clients, and the third will be a couple of hours of work. No design company can really afford to spend the weeks required understanding your business and your brand for a purely speculative tender. The result will either be a logo that looks initially attractive but later is discovered to be inappropriate for some reason (for example, because it doesn’t reproduce well on your packaging), or, worse, one which is very similar to something already out there.

No designer is intending to plagiarise (I say that more in hope than confidence), but we are all influenced by what we see. If someone reads about your company and instantly has an idea of what your logo should look like, chances are that they are remembering rather than imagining. Of course, it won’t be quite the same, because your organisation won’t have quite the same name. However, when someone points it out, the similarities are painfully obvious.

There’s a bloke in Birmingham who offers logos for £50, and provides a range of them for you to choose yourself. One of them looks eerily like the World Wildlife Fund’s panda. Probably many of the others look like other things as well. Again, I’ve no doubt that he’s drawing ideas that come into his head, but what comes into our heads most often is what we’ve seen.

You can’t guarantee that the design that you come up with will be dissimilar from all other designs in the world, but you can guarantee that you won’t be plagiarising. To clarify the difference (for those other readers), if your process leads to something which is similar in design to something of which you had no awareness, this is an accidental similarity. You still won’t be allowed to register it as a trademark in the same category as that other brand, but you won’t be plagiarising. On the other hand, if you just get a load of designs and pick the one you like, there’s quite a high risk that you’re half-remembering in the same way that the designer was half-remembering. If it later turns out that your designer’s web history includes that logo, or is on reference material she or he possesses, and you have no history of earlier drafts that led to that design, then you may well be in trouble. This trouble will be worse if it’s then shown that what appears to be an odd ornament, a quirk, in your logo was a actually a key functional element in the original — exactly like the bottom of the L in the Tokyo logo, which has nothing to do with Tokyo, but everything to do with Théâtre de Liège.

I could be a bit self-serving here and say: don’t start by going to a designer, go to a brand consultant first and let them engage the designer. However, that’s for another time.

If you are evaluating a logo, but you have no idea how to other than ‘I like it’ ‘It feels like us’ ‘It feels like what we want to be’ or ‘twenty people picked it in a survey’, then here’s a useful acronym that can help you: MADE. Great logos are MADE, rather than half-remembered. There is a making process, which you can document, and should, just to be safe.

  • Memorable
  • Appropriate
  • Distinctive
  • Efficient

A logo needs to be memorable, and that means memorable in relation to your business. A single red circle may be the greatest colour/shape combination ever invented, but unless your business is selling red circles (in which case it’s not trademarkable, as being illustration), it doesn’t connect with what you do. That’s why most effective logos incorporate the name of the organisation, either as text, or, with something as famous as Apple, by obvious visual association. Even Apple, though, kept Apple Computer, and then the word Apple for years before it was able to go to the symbol only.

The keys to memorability are: something recognisable, one thing which is different, and tip of the tongue pronunciation. Tip of the tongue (t, d, p, b, f, s, not g, k, r, j, n, l) is to do with naming — again, the logo is not actually the right place to start. The recognisable and different, though, is the key to great logos. Gillette is Futura extra bold italic, with the dot over the i change from a circle into a razor shape, and a razor added to the e. SONY is such a distinctive word that it doesn’t need additional distinction. The loop of short term memory is 2-3 seconds. If the eye hasn’t ‘got it’ in that time, your logo is not memorable.

A logo needs to be appropriate. By far the biggest problem I encounter is that organisations don’t know what appropriate is, because they have no clear notion of what their promise is, what the brand’s values are (i.e., what makes it valuable to their customers), how it is distinctive in its market, and what its personality or style is. That’s another reason to not start with the logo. If you have a strong understanding of what your organisation is, it’s relatively easy to just run down a check list and say ‘no, no, yes, yes’ and so on. I was once branding a children’s programme. We went through a personality checklist and agreed what it was. I came back a week later with a logo which I though might do (I was being rather lazy). Having done the personality work, the commissioner was immediately able to point out i) that it wasn’t appropriate ii) why it wasn’t appropriate and iii) what we needed to do to make it appropriate. A day later I was able to come back with the right thing, we agreed it, and, four years later, it is still working hard doing its job. When you know what it is the logo is representing, it’s fairly easy to say exactly how it needs to be improved. Otherwise, you are doing little more than saying to a bewildered designer ‘it’s not quite right—can’t you just make it a bit better?’ The result will usually actually be worse.

Distinctive is, of course, what the Tokyo logo wasn’t. Aside from the fact that Tokyo has a T in it, it didn’t particularly distinguish it from any other monogram. Most importantly, it didn’t distinguish it from a rather smaller, but still important, cultural competitor: Théâtre de Liège. If your logo involves your organisation’s name, and your organisation’s name is not like one of your competitors, then it probably will be distinctive. You can search the trademark register much more easily for names than for designs, and have confidence that your name is distinctive (if not, change it before you change your logo) and that the design is therefore also distinctive. Check what colours your main competitors are using. Pick either the complementary colour, if there’s really only one competitor, or the colour furthest from the other two. I had the privilege a couple of years ago of being the consultant for a brand that subsequently won Brand of the Year in its industry. It had two main competitors, who used respectively red and green as their key colours. The brand’s previous logo was green. One of the more difficult decisions (but absolutely the right one) was rotating the colours so that the new logo was blue, though a blue with a similar quality as the old green. Of course, there were other distinctive, but colour is the first thing we see.

Efficient is something you learn about after you’ve been doing branding for a while. Can your logo, and its corporate colour, and its typeface, be efficiently reproduced on all of your core products, packaging, and promotional materials? You would expect a designer to ask what your materials are, but often they don’t. Twenty-six years ago, when I was starting out, design was a physical process involving pens, prototypes and cutting out. Today, many designers work almost exclusively on computer. There are plenty of things you can show on a website which can’t be done in the physical world. This even goes as far as sending a copy of your draft logo to your packaging printer and asking them if it will be easy to reproduce: if they are printing in Flexo, or using heat-responsive paper, you may discover your marvellous new look comes out as a smudge.

If you—or your designer—can show how you set up your criteria for MADE, in relation to your industry, your organisation’s distinctness, your services or products and your methods of distribution, then you will be fairly immune to any charges of plagiarism.

If it’s just something that popped into your head, or your designer’s head, then there is a real risk that you are simply appropriating someone else’s look.

The lesson of the Tokyo Olympics is this: no one is too big to get away with it.

Ten tips to make your text look better

Ten tips to make your text look better

Sending out a newsletter? Doing a poster? Formatting a thesis? Producing a leaflet? Thirty years ago, you would have been in a tiny minority, working with Letraset and/or a typewriter, with a ruler, T-square, Rotring pen and a cup of instant coffee (unless you were European). Today, everyone with a computer has at least experimented with the idea of doing one of these. Instead of Letraset, you have thirty fonts that came pre-installed, and thousands of others a Google away. Gone are the Rotring and ruler. If you’re drinking coffee, it may even be cappuccino.

And yet, somehow, most of the newsletters, posters, leaflets and theses produced today look inexpressibly ‘not right’. Compare them with a commercially produced magazine, advert or flyer, and there is something which is…

These ten tips will show you how to achieve the missing polish. Whether what you have to say is worth saying is an entirely different matter…

#1 Turn kerning on

The most basic reason why a page of text printed from Word looks less good than the same page printed by a designer from (say) QuarkXpress or InDesign, is because Word turns the kerning off by default.

Kerning? You say. What is kerning?

Kerning example shows the LY gap

In the unkerned letters (top), the gap between L and Y is too great

Typographic fonts (as opposed to typewriter fonts) have proportional letters, which means that the w and m are wider than i and l. Each letter is given the correct width and space around it in the font’s definition (i.e., the font file). By contrast, typewriter letters, which are monospaced, assign the same width to every letter, because that was, mechanically, how typewriters worked. You can use this to your advantage if you like: picking a monospaced font, such as Courier, gives a strong typewritten feel.

However, the standard spacing for the letters does not work for particular pairs of letters, such as L and Y (see illustration). A gap opens up which makes the text jolt as you read it, and gives a more amateurish impression. The kerning tables are built into the font files as well. However, in the early days of personal computers, many printers either could not print properly kerned fonts, or took much longer to do so. For this reason, kerning was not a feature of early word processors. For some inexplicable reason, Microsoft has chosen to keep kerning turned off by default in Word, even though every modern printer can print kerned text without problems, and every modern font comes with the correct kerning tables built in.


Dialogue in Word showing kerning

The Advanced font dialogue — turn kerning on for all text sizes


Here’s how to fix it:

In Format/Fonts (or whatever your version of Word does for this) go to Advanced and turn it on.

Ideally you should do this in your Normal template. Kerning should be on for all fonts, all of the time, for every size, so set a small size such as 6 for ‘Points and above’.

For some fonts, this won’t make much difference. For others, such as Calibri which Microsoft tries to make the default headings font on my version of Word, it will make a substantial difference, when looked at large. It’s not the substantial difference that makes the difference, though—rather, it’s the smoother, more regular, less gappy text which is consistent on every page.

There is no downside whatsoever to having kerning turned on: it isn’t a ‘type trick’, but something which is part of the font, ignored by Word for legacy reasons.



#2 Don’t pick the wrong font

Getting the fonts right is not so much a question of picking the right font, as not picking the wrong one. Essentially, any font which draws attention to itself is the wrong font for text, and is probably the wrong font for titles.

Again, Microsoft doesn’t really help here, because the standard font for Word always used to be Times Roman, which is silly for sending someone a letter (it was the newspaper font created for the Times to be legible at small sizes when printed on newsprint). These days, Microsoft seems to want to offer Arial 12 point, though that may be the result of something that happened to my Normal template years ago.

Arial is not necessarily a terrible choice, and even Times Roman will not let you down. Where things start to go horribly wrong is when you click the font menu, and you are offered a staggering choice of every font on your system. How many do you have? A hundred? Two hundred? My system has 6,000 fonts on it, though I use a font manager to ensure that only a few of them are installed at any one time. Most of these fonts appear to look very much like each other. You’ve got Helvetica, Arial, Franklin Gothic, Gill Sans, Univers, Myriad and Lucida Sans that all look like they went to school together, and you’ve got Palatino, Times, Garamond, Baskerville, New Century Schoolbook and Bookman that look like they all buy their clothes at the same department store. Actually, none of these fonts, used consistently, are bad choices. But those aren’t the fonts that look really interesting.

You probably already know that Comic Sans is a bad choice, and you may have heard that Helvetica is the ‘right’ choice, whatever the situation. So far so good. But what about all these nice curly, fat, extended, artistic, handwriting and other fonts. Surely they have some place on your page?

In the words penned by Meat Loaf, Stop Right There.

If you are a designer, designing a page from the bottom up with a very clear vision in mind, and you are choosing a font that matches that, then, go right ahead, choose that font. Except, if you’re a designer, you’re probably using TypeKit and maybe Linotype’s library to choose absolutely the best font, even if it costs £250 (which is what fonts cost when you buy them). If you’re setting text to be read as text, than pretty much any font which says ‘look at me’ is wrong. This goes for Avant Garde Gothic, Avenir, Ad Pro, and thousands of others all the way to Zapfino and Zapf Chancery.

Stick with the basics, and it will be fine. Wander from the path (in the words of Gandalf) and you are lost.

By the way, if you are part of an organisation, there is probably an official corporate font. Use that, and never use anything else.

#3 Don’t pick many fonts

If the text itself should be in one plain, basic font, doesn’t that give us free rein to set the titles in something nice?

The old rule is: no more than four fonts on a page.

However, this rule is misleading, because ‘font’ referred originally to one weight at one size. What people call fonts these days are actually typefaces.


Lead letters at the Plantin Museum, Antwerp

Let me quickly explain. Between Caxton and the creation of typesetting machines, letters were made in lead at a type foundry, and racked onto the press with a collection of tools, before being locked into place. A typeface would be a family of different sizes and weights (literally, weights) of letters. A font or fount would be where all of the sorts of a particular size, weight and style were kept together. Provided no one had put them back in the wrong place, a printer could lift them out by feel and set them into the press.

The four font rule essentially allows you Regular and Bold of your body text size, a larger heading, and a title heading, or a larger heading and a footnote size.

The most common problem with home-made print is that the result contains many changes of size, bold, italics, headlines in different sizes and different fonts. Stick to the four per page rule. Also, if you are going to have the headers in a different font from the body text — for example, Palatino text and Helvetica Bold headers — make sure that they offer a significant contrast, and make sure that the styles go together. You don’t have to think too long about this: if they don’t ‘obviously’ go together, then they don’t go together.

#4 Use tracking (no, not that kind of tracking)

If you can only have four fonts to a page, and you now know that this doesn’t mean ‘four typefaces’, how do you make the titles really stand out? One of the most common reasons why people jump for fancy fonts for titles is because the titles somehow don’t stand out enough. The typeface Impact is a popular one to make things stand out. After all, it does create an impact. However, it also looks like you’re shouting, even when not in capital letters.

When tracking or letter spacing is reduced, the font appears bolder

Top: Impact
Middle: Helvetica Bold
Bottom: Helvetica Bold with reduced tracking

The answer is tracking. By this we don’t mean the tedious underlines and redlines of ‘turn document tracking on’.

Tracking is the space between the letters. In computer typefaces, the space is automatically set to the most pleasing, and this is defined in the font file. You don’t have to do anything about it, as long as you have kerning turned on. In lead type, spaces had to be inserted between each letter. These were very narrow, and putting none in created text which was hard to read at any length, but made a huge impression in titles.

In the example, you can see the impact that Impact makes. Underneath it is Helvetica Bold, which generally just doesn’t look bold enough for titles. Below that is Helvetica Bold with the letter spacing, aka tracking, reduced. You do this in the same dialogue as the kerning (above) if using Microsoft Word. It’s best to define this as a style, because you want all your titles to look the same. You have to experiment a bit on this, because Word asks you to enter it in ‘Points’, which means the correct amount will change depending on the point size in your text. If you were using QuarkXpress, you could enter it in 1/20 Ems, which would then change proportionately with the size of the font. To get it right, reduce the spacing to something that looks really good, and then halve what you’ve done: it’s easy to overdo things. When you check the page, if it looks wrong, go back and halve it again.

#5 Learn the space rule

Even if this has never bothered you, you are probably aware of the never-ending controversy between the ‘one space after a full stop’ people and the ‘two space after a full stop’ people. This is like the Android versus iPhone and Mac versus PC war, except everyone can play, even if they aren’t sure whether their device is an Android or an Etch-a-sketch.

For a little context, when I learned to type, which was from a proper typing course before the invention of the Personal Computer (I was nine), the rule was: two spaces after a full stop. However, if you look at any book on your shelves, you will see that there is one space after a full stop.

Which is correct?

Proportional fonts such as Times Roman take one space. Mono fonts take two spaces.

Proportional fonts such as Times Roman take one space. Mono fonts take two spaces.

Both, as it happens. Typewriters use mono-spaced fonts, as noted under ‘kerning’ above. Starting in the 1890s, typewriters revolutionised offices, replacing the hundreds of clerks whose job it was to hand-copy (in copperplate handwriting) all records and correspondence with a few, highly skilled, typists who could rattle away all day at forty-five words per minute on a manual typewriter. A mechanical typewriter shifts the carriage along by exactly one space per letter, so every letter must occupy the same amount of space. This is why wide letters, such as M and W, look compressed on typewriter fonts, and narrow letters, such as i and l have heavy ‘slab’ serifs to broaden them. Typewriter

However, a full stop cannot be made broader (if it is, it becomes a dash). What’s more, the full stop has to be in the centre of the typewriter’s hammer, because otherwise it will skew the lever over time. The levers are very narrow on a traditional typewriter, and they take a considerable pounding, so, mechanically, it is essential that they balance correctly. Over many years, particular letters would wear on most typewriters, which is why true typewriter emulation fonts, often scanned from typescripts, generally have variation in the letter position or weight.

Returning to the full stop, the proportions do not look correct when a mono-spaced full stop is followed by a single space, because of where the dot is positioned, so the typist rule, taught in every secretarial college and typist course since the 1920s, is two spaces after a full stop.

Printers have numerous types of spaces.

Printers have numerous types of spaces.

Notwithstanding this, the typographer’s rule, taught to every apprentice since the 1440s, is one space after a full stop.

Okay, this is a slight oversimplification. Printers have a number of different kinds of spaces: en spaces, the width of a letter n, em spaces, the width of a letter m, thin spaces, hair spaces, figure spaces, and several others. QuarkXpress (dialogue pictured left) has twelve kinds of spaces, most derived from old-fashioned typesetting. The printer would insert just one space, equivalent to the punctuation space, after each full stop.

Electronic typefaces are designed so that the width of the ordinary space provides the correct spacing when one space is used after a full stop.

So, when using a proportional font, set with one space, not two.

This leaves the question: which fonts are proportional? The answer is, all of them, except ones deliberately designed to look like typewriter fonts, such as Courier. Even American Typewriter, despite the name, is a proportional font.

#6 Dispense with underline

Using underline? Just stop. Underline is another hangover from typewriter days. A mechanical typewriter can only have one style of type per letter, and even the capital letters are created by pressing down the SHIFT key, which, on a mechanical typewriter, actually shifts the keys physically downwards, so that the Capital hits the ribbon rather than the minuscule. I’m sure that someone, somewhere built at least one prototype typewriter that allowed you to second-shift for italic, but that wasn’t how ordinary typewriters were built. However, italics have been in use by printers for hundreds of years. Therefore, as a convention for informing the printer that something had to be in italics, the typist would back-space and go under the letters again with the underscore character, _. Some machines had a red and black selectable ribbon, and this was sometimes used instead of the more time consuming underline.

Underline should not appear on ‘printed’ documents. When it does, it looks scrappy. This is partly because our eyes are accustomed to seeing italics rather than underline in professional documents, and partly because underline interferes with the letter shapes p, q, y, g and j, making the words less legible and also creating visual awkwardness. When designers use underline as part of a design, they break the underlines so that they don’t conflict with the bottoms of the letters. For general typesetting, just don’t use them. Use italics instead, or, better, consider: is it really necessary to add extra emphasis here? Bold or bold with tracking may be better.

#7 Understand leading

Leading tight, 20%, 50% and 100% extra

Leading tight, 20%, 50% and 100% extra

The space between the lines is called leading. That’s ‘leading’ as in ‘ledding’, not as in ‘leeding’. It comes from the metal lead, from which printer’s letters were made, not from the present participle of the verb ‘to lead’. Technically speaking, the leading is the space from baseline to baseline, though the lead which was actually inserted was between the top of the letter blocks and the bottom of the next letter block. In Word and other consumer software, this is often described as ‘inter-line space’ or ‘ line spacing’.

On a typewriter, you could use single line spacing, double line spacing, and, on some machines one and a half line spacing. As with all things, a printer had more options.

Your word processor most likely defaults to a leading of 120% of the type size. The type size itself, given in ‘points’, is not ‘how big the font is’, but the distance from the baseline to the next baseline, taking into account the maximum height of the ascenders (l, L, etc) and maximum depth of the descenders (p, q, etc). This is why the size in points is not a good guide to legibility. Helvetica reads fine at 10 point, Zapfino is still almost illegible at 18 point: fonts with a high x-height (height of the letter x) compared to the t-height (height of the t) are generally more legible than fonts with a small x-height, such as Bodoni. This even varies within fonts that have the same name. ITC Garamond has a higher x-height than other Garamonds, making it more legible at the same point size.

If you want to increase legibility, you can increase the leading. Normally 12 point text will have a leading of 14 or 15 points (+20% on the base-line height), but if you give it 18 points (+50%), it will be much easier to read, though, of course, it takes up more space. Setting it ‘tight’, where the leading is the same as the font size, will make extended reading more difficult. However, like reducing the tracking (see above), it can make two line titles really jump out.

Once again, you can use the Advanced dialogue in Word’s Font function to make these changes.

Something that’s also worth thinking about is that you don’t need a full new line between each paragraph. In the Paragraph dialogue in Word, you can set the inter-paragraph spacing to be, say, two-thirds of the point size. This makes new paragraphs less abrupt, and also wins you back some space, which you can then use for increasing the leading and thereby improving the legibility (but don’t change the leading from page to page, or, if you do, not by anything noticeable).

#8 Respect the typographer’s art

Every font on your computer (apart from the free ones you downloaded from that site) is the fruit of thousands of hours of a designer’s experience and hundreds spent just on that one font. Creating a font can take as much as three years, and often the exact spacing of the letters will then be sent out to an external agency. In ‘book’ fonts — those such as Palatino and Garamond designed to be used in ordinary body text — every curve of every letter has been balanced with the negative space to create the optimum combination of legibility with personality.

And then someone uses WordArt to skew, bend, twist or reshape this work of art to make something more eye-catching.

The result is always awful.

Most fonts used in logos have been subtly reshaped to match exactly the needs of that logo word. However, this is done with as much care as the original designer took over the font itself. People go for years without noticing that Gillette has the dot over the i revised to create a razor (and the e reshaped as well). However, this subtle reshaping is not done by applying a ‘Create Logo’ function to a font.

The only effect that should ever be applied to a font is dropped shadow, and even that should be used judiciously, and never in body text. If you have the time and expertise to carefully reshape the letters to meet an exact design vision, then by all means go ahead (but, in that case, why do you need to read this article?). If you’re just looking for a way of freshening up an otherwise dull page, then consider rewriting the headlines, improving the layout (generally by including more white-space), copy-editing the text by deleting verbiage, or re-shooting the photographs. Putting an effect, especially a WordArt effect, just makes it look sloppy.

#9 Stick to one colour for text

Got a colour printer? Why not put the text in different colours? Well, simply, because it looks confusing. The most legible colour combination is black text on a yellow background. Actually, that is over-legible, and should only be used for warning signs on dangerous equipment. For most applications, black on white is best. Black on red is the least legible, though any colour combination that reduces contrast creates a problem. Some visual impairments make it easier to read white on black. For most readers, black on white is easier (and for some visual impairments, white on black is illegible). If you absolutely have to set text in any other colour than black, then check two things: first, is there enough contrast for this to be genuinely legible when you print it out (what’s on your screen is an entirely different thing) and, is it still legible when it’s been reproduced on CMYK print (from a commercial printer), Flexo (a packaging printer), Risograph (a political printer), black and white photocopying (someone’s idea of saving money) or whatever means you are actually going to use. Particular colours separate badly, giving text that looks out of focus when printed on particular processes. Whatever you might have gained by setting it in Plum, has now been lost by the fact that it looks like you got it printed in the 1930s.

Stick to one colour for headings, one colour for text, and leave it at that.

#10 Check it by eye

By far the best advice I can give anyone is the advice that was given to me in 1988: check everything by eye. What is on your computer screen is not necessarily what comes out of your printer, and absolutely not necessarily what your eventual readers will get. Screen technology, by definition, is different from print technology. Light is coming through the screen, whereas it only reflects from the paper.

When you actually have something in your hand, and start putting it near other things, the fancy ideas you thought would spruce it up often start to look humdrum. What’s more, many faults will begin to emerge: a line of text with lots of capitals in it will look too close to the line above it. The same goes for numbers. Hyphenation which looked fine on screen may suddenly be disturbing. Things look like they aren’t quite in the right place.

All this is normal: the computer sets things according to the instructions that the software gave it, based on whatever styles and fonts you’ve chosen. None of these things actually understand the way that the letters, words, lines and shapes fit together on your page.

Again, in the advanced dialogue in Word, you can move lines fractionally up or down which look like they are in the wrong place. You can extend or contract spacing, modify where a letter starts, and do many other things. You shouldn’t play with these if you don’t have to, but, if it looks wrong when you print it out, it will look wrong to everyone who picks it up.

Back to Top