Skip to main content

Ampersands, date ranges and contractions: style guidance

Posted by: , Posted on: - Categories: Best practice

The information in this blogpost may now be out of date. See the current GOV.UK content and publishing guidance.

With a lot of new organisations coming onto GOV.UK, I thought I would clear up a couple of style points.

Ampersands: why we don’t have them

We’re getting requests to use ampersands in organisation names that are written in full. We have ampersands in logos - the pictorial logo at the top of the page - but not in body copy.

The reason is that ‘and’ is easier to read and easier to skim. Some people with lower literacy levels also find ampersands harder to understand. As government, I know we would not want to exclude users in any way, so that’s why we have ampersands in logos only - not in copy. Don't worry - we're going to practise what we preach and remove ampersands from those areas of the site that we're responsible for.

Date ranges

We’re also getting requests for different types of date ranges. Take a look at this example:

'This information relates to 2013-14'

What does that mean to you?

For a start, the dash isn’t as easy to see as the word ‘to’ - so we remove the dash in date ranges. But when exactly do we mean here?

It could be the academic year, calendar year or tax year. That’s why we insist date ranges are very, very clear.

Use ‘tax year 2013 to 2014’ or ‘September 2013 to July 2014’. Or if you are comparing statistics from 2 different tax or financial years then use 'Comparing the financial year ending 2011 and that ending 2012, there was a 9% decrease' etc.

Remember: you’re on a site with over 21 million visits a month. We can’t guarantee your audience. We can’t guarantee everyone will understand what you mean. But we can make sure we get as close to accessible for everyone as we possibly can, simply by being very, very clear.


You can use contractions. Some new organisations are reluctant to use them. All I can say is that we have never encountered a problem with understanding when testing with users. Sometimes, lots of ‘cannot’, ‘should not’ etc can seem archaic and formal. That’s a tone I think we can move away from without jeopardising the overall tone of information coming from government.

Have you noticed…

…. this is a sort of frequently asked questions (FAQs) page? These questions are frequently asked, they are just structured so everyone can read them quickly and easily. Did it make any difference to your understanding of the page because there’s no actual questions?

FAQs: why we don’t have them

Keep in touch. Sign up to email updates from this blog, or follow Sarah on Twitter 

Sharing and comments

Share this page


  1. Comment by Richard Heaton (CO perm sec) posted on

    Thanks Sarah. Good stuff.
    Another reason for avoiding FAQs is their generally untruthful premise. FAQs are usually questions the writer would find it convenient to be frequently asked, and rarely the questions anyone has ever actually put to them.

    • Replies to Richard Heaton (CO perm sec)>

      Comment by Sarah Richards posted on

      Thanks Richard. We totally agree.

      I think the hardest thing here is culture change. People have been using ampersands, FAQs etc for a long time and because they haven't had a mass of complaints - they don't change. We challenge everything - every assumption we have ever made - to see if there is a better way of doing it.
      We hope all perm secs would agree.

  2. Comment by NoraB posted on

    Great post - all of it good common sense to make content easy (or accessible if you like) for users. I too don't understand the reasoning behind FAQs; who exactly asks these questions and how frequently? Particularly puzzling when they're published at the same time as a content page!

  3. Comment by Sam Brierley posted on

    FAQs are known here as 'Fondly Anticipated Questions'

  4. Comment by Robert Morgan posted on

    I disagree about ampersands in organisation names - you say "and" is easier to skim, but ampersands are easier to skim over, and that can be part of good communication too. They're a semantic clue that something is just a formal name and doesn't need to be understood.

    So while I'd never write "our Research & Development department" (for all the reasons you give), I'd argue that "Ernst & Young" is easier both to understand *and* to skim than "Ernst and Young". That's because the second version invites you to read and unpack it, perhaps to wonder about two people with those names; but the first essentially says "here is a proper name, it doesn't mean anything, carry on".

    I suspect that point works for non-native and uncertain speakers too. I don't think they're scared at a symbol in a proper name any more than I'm scared by an umlaut in someone's name. But an uncertain speaker reading a full "and" is likely to worry they don't understand the words around it ("I know what young means, now what does ernst mean?")

    (just my view, of course, I don't have the research to back this up)

    • Replies to Robert Morgan>

      Comment by Sarah Richards posted on

      The ampersand thing is a contentious one. And you might be quite right.

      I have no opinion on ampersands in names in logos or headers or posters. I have no evidence/research so I don't have an opinion.

      In this instance, this decision was based on a bunch of stuff:
      * research tells us people are 75% more likely to skip a small word after a long one. Perfect if the word doesn't change/aid comprehension because it speeds reading. Not so good if the word is 'not' etc, obviously.
      * the word 'and' is part of the common word set - from all the lists I've seen it's usually in the first 5. This also means people can take in the word, probably in the third fixation zone (anyone got empirical research on that?) and jump it.

      I don't think people are 'scared' either. I think it's one of those things - people are used to it and deal with it but a tiny change can actually help with few people noticing.

      I've always said: " I don't want anyone to read a single word on GOV.UK. I want them to understand and leave without any unnecessary fuss".

      We can help people skip. It doesn't stop comprehension, so why not?

      Most people come to this with an opinion but no evidence. If there's any evidence at all that comprehension is impaired, I'd change it immediately. I am not wedded to any part of the style guide - if we need to change it, I will callously throw it away without a second thought. I'm positively heartless in that regard 🙂

      I am more than happy to change this as soon as I find comprehension is impaired.

      • Replies to Sarah Richards>

        Comment by E.A. Brown posted on

        Sarah said:
        In this instance, this decision was based on a bunch of stuff:
        * research tells us people are 75% more likely to skip a small word after a long one. Perfect if the word doesn't change/aid comprehension because it speeds reading. Not so good if the word is 'not' etc, obviously.
        * the word 'and' is part of the common word set – from all the lists I've seen it's usually in the first 5. This also means people can take in the word, probably in the third fixation zone (anyone got empirical research on that?) and jump it.
        Most people come to this with an opinion but no evidence. If there's any evidence at all that comprehension is impaired, I'd change it immediately.


        Hi Sarah,

        I'm glad to hear your style guide can accommodate change.

        You maintain that your style choices are strictly based on evidence. Is it possible to see any of that evidence and research?

        As someone who consults the style guide regularly,
        I need access to your evidence base
        so I can contribute to, or contest, your style decisions.

        Without knowing what constitutes 'evidence' it's difficult to explain your decisions to my colleagues, or to suggest improvements and amendments.

        I've asked in the past for the evidence for style choices that are apparently related to accessibility. The reply was:

        ' I'm afraid I don't have any studies on the effects of screen readers, just the word of our experts. But please do work with a few different screen readers, and let us know what you find out. I imagine the technology changes greatly every few years, so we would love to get your conclusions.'

        How is 'the word of our experts' any better than any other editorial decision, if I can't see the evidence for myself?

        • Replies to E.A. Brown>

          Comment by David Read posted on

          The state of government communications has been so contemptibly bad and inconsistent for so long, I think it is great that a decent modern style guide has been produced to up standards across the board. Sure, lots of people have opinions or might write dissertations on every last decision if they wanted to. However, I don't think anyone is arguing that GDS has made an utter hash of this, so I think it is a waste of tax payers' money to discuss the minutiae of these decisions. Give it a couple of years to bed-in before investing in a time-boxed review and discussion.

          • Replies to David Read>

            Comment by E.A. Brown posted on

            Hi David,
            I have no issue with the style guide's intent: plain English is a project after my own heart.

            I do have questions about the evidence base used to make decisions, and what kind of evidence would influence changes. I have a vested interest in its enforcement too because it affects my daily work.

            I'd also prefer to remedy any poorly-thought-out decisons earlier rather than later, to avoid inconsistencies in GOV.UK content over time. As it is, I've seen about-face changes in the style guide (from 'must do this' to 'must not do this'); I'd greatly prefer to keep these to a minimum.

        • Replies to E.A. Brown>

          Comment by Sarah Richards posted on

          Hi E.A. Brown,
          The evidence I have stated is just desk research so if you search for psychology of reading or scholarly articles - you'll find a heap of it.

          Also, sometimes a style guide is just about consistency. A site benefits from consistency and perhaps there is no evidence. This is not the case here and generally, we do go looking for research before making a decision.

          I'm afraid I don't agree about changes to the style guide. Whilst I understand it may be a chore to change content from an existing style to a new one, we will change the guide as soon as we have new research or a better understanding of how to best fulfil our users' needs. This means the style will change regularly - at least I hope it does. If it doesn't, we will become stagnant and we won't be learning. Something I hope GOV.UK never stops doing.

    • Replies to Robert Morgan>

      Comment by Joshua Marshall posted on

      Interesting comment Robert.

      One of the reasons why I've pushed to not use them is that, for users of assistive technologies such as screen readers, depending on how they're set up by a user, typographic symbols aren't always read out as intended.

      Accessibility company Deque posted a fantastic post about this a few days ago:

      While it's typically classed as a "safe" symbol to use, we can't guarantee that it will be read out, so we advise against using them to make sure we're communicating clearly to everyone.

      Hope that helps.

  5. Comment by Tim Blackwell posted on

    I wouldn't want to be too dogmatic about FAQs. I maintain the help system for a product with about 100,000 words of text over 750 or so pages. We have one FAQ of a few hundred words. It covers questions we have been asked frequently.

    Nonetheless, the FAQ has shrunk over time. This is because the FAQ has a secondary benefit: at least in part it represents a list of problems we should be trying to engineer away.

  6. Comment by Johanna Lakeman posted on

    I often see ampersands corrupted to '&AMP' when the original content is shared on social media sites or data downloaded/shared. Another reason not to use them maybe?

    • Replies to Johanna Lakeman>

      Comment by Sarah Richards posted on

      Yes, another good reason. if comprehension isn't actually impaired, I don't see why we would continue to use them.

  7. Comment by Graham Lee posted on

    Good point about date ranges. This is a real issue in publication titles - a quick search brings up 2012/13, 2012-13 and 2012 to 2013.

    Using the preposition and spelling out that it's a tax year - or whatever - in each case might be too much in the title, but it would be good to have more consistency - perhaps enforcing an oblique '/' for financial or academic years (2012/13), as against a dash '-' for calendar ones (2012-13), with a note in the summary explaining the date period covered?

    Not convinced by ampersand argument - especially if they're deemed ok for logos, and we abbreviate other things like 'Her Majesty's' in 'HM Revenue and Customs'.

    '&' is two characters shorter, and is used in traffic signs, A&E departments and M&S...

  8. Comment by Adrie van der Luijt posted on

    Having worked on Universal Credit and now CAP Services at the Rural Payments Agency, I have mixed feelings about contractions. Used in moderation they can enhance a conversation tone of voice, but I fear text with many contractions will soon start to feel badly dated too.

    I do, however, draw the line in content where the word 'not' is crucial, for example to distinguish between two important options. Using 'don't' instead of 'do not' feels inappropriate in such cases and doesn't enhance clarity.

  9. Comment by Ross Orange posted on

    You mention in your style guide 'Don’t use Americanisms' yet the date format you use is that of an Americanised style, 14 June 2014 (yes I know that 'over the pond' it would be June 14 2014). In the UK it is standard practice to write this as 14th June 2014, apart when using Microsoft Word where Americanisms are being foisted on us but we can amend.

    So to be consistant may you not continue to use 'our' version rather than a twisted Americanism?