Main points or Main changes

All articles include a section at the start that summarises the main trends, analysis or changes detailed in the article. This section is called “Main points” for analysis articles and “Main changes” for information articles. It should include up to six bullet points each no longer than one sentence.

Best practice for web writing is to use an inverted pyramid style of writing. The pyramid means placing information in order of importance, so your main findings and conclusions should come first.

Split your main content into sections

The main body of your article should provide more detailed analysis or information on the trends or changes summarised in the Main points or Main changes section. 

You can either create a new section for each bullet point in the Main points or Main changes section, or group the information into topics of interest and create a new section for each topic.

Keep your sections short; avoid putting all the information under one section as users will not be able to find what they are looking for in the table of contents.

Use clear section headings

Use clearly labelled section headings so that users can find what they need quickly. They should be short and concise with the most important information first, and reflect the topics users are interested in. 

These section headings will appear in the table of contents and will help users find what they are looking for.

  1. Knife crime
  2. Deaths by local area
  3. Use of administrative data
  4. Challenges of data collection

There are standard section headings that you can use to structure your content. Some section headings should be avoided to meet best practice for web writing.

The content design team can help you structure your content and choose the most suitable section headings – email

Use subheadings to break up your text

Use a new subheading within your section every time you discuss a new subject, trend or change. Put the most important point at the start. Subheadings should be a maximum of 75 characters, including spaces, to prevent the text wrapping over too many lines, particularly on mobile devices.

Avoid numbering subheadings (for example, 3.1 or 3.1.2) as these cannot be linked to in the text. They also slow down online users who often use subheadings and scan the left side of a page to find the topic they are looking for (see the F pattern). A short and clear subheading is more useful and will help users navigate by headings.

Write clearly and in plain language

Avoid using jargon when a simpler term or explanation can be used. UK Statistics Authority standards and UK accessibility legislation require the use of plain English. If you need to include more technical language, make sure that it is clearly explained for users who may be less familiar with the term. Include a brief summary in your analysis and link to a more detailed definition in the Glossary section.

Keep your paragraphs short and concise, covering one topic or idea. Aim for a maximum of four sentences per paragraph.

Using charts and tables

Analysis articles

Simple charts and tables can often convey a trend in the analysis more clearly than a dense paragraph of text. Your surrounding text should add to your visuals, not just report what they show. It is not necessary to comment on every trend shown in a chart or table.

Use descriptive chart titles

Chart titles and subtitles should be concise, sentence case and in plain English. The main title should be descriptive, and tell the trend of the data or highlight the main story. Our chart title guidance provides more detailed information on how to write descriptive chart titles.

Keep your tables focused

Tables can be used to present trends in the data alongside your analysis and allow users to easily compare data. Column headings should be clear, concise and written in house style.

Avoid including too much information in your tables as this is likely to overwhelm users. Limit the number of columns to be fewer than nine as this is the maximum number of columns that we can create.

Information articles

All charts should be interactive where possible but occasionally you may need to include a diagram  in an information article to help visually explain a process.

Before deciding to use a diagram, identify the user need. Consider if the user need can be met solely by clearly written text. If you have too much information to communicate, a diagram is unlikely to be useful to users. If you can describe the diagram in a short paragraph, use text instead.

Any diagrams included in your article must have a clear user need and be created by our in-house Design team. This is to make sure they are accessible and in line with our design principles. More detailed guidance on the use of diagrams is available.


Warning patterns are used to highlight essential quality limitations to the user. They can be used to highlight if:

  • Experimental Statistics, a type of official statistics that are going through development and evaluation, have been used in the article
  • the research or methods being presented are in the early stages of development (and not yet official statistics)

Use a grey warning box at the end of the Main points or Main changes section to make users aware of any quality issues around the statistics or methods.

For Experimental Statistics, use:

“These are Experimental Statistics. The [method/data source/estimates is/are] currently under [review/development], which means [brief detail about how this affects estimates or data quality]. We advise caution when using the data.”

Experimental Statistics should always be primarily published as a bulletin where they are the first release of new data. An article using the data can be published alongside the bulletin if more detailed analysis is needed.

For statistical research and methods, use:

“These are not official statistics and should not be used for policy- or decision-making. They are published as research into [a new/an alternative] method for producing [topic] statistics. We advise caution when using the data.”

Change the second sentence to describe your release. More information on the methods used and quality limitations of the data can be included in the Data sources and quality section.

Next: View the article types and how to structure them


The data section provides links to the most important datasets used in the article. It helps users find the information they need and brings data from different topics or themes together in one section.

This section is only included in analysis articles that contain data, and it follows a similar structure to the Data section in bulletins

Provide links to up to five datasets that users are most likely to be interested in. If your article references more than five datasets, choose the most relevant. Use the standard format for each link.

Migration data
Provisional Long-Term International Migration estimates
Dataset | Released 29 November 2018
Migration flows to and from the UK, quarterly tables and charts.

You can also include a sentence at the end of the Data section to help users access any other datasets used in the analysis. Use the following standard text to link to the related data page:

“View all data used in this article on the Related data page.”


Articles can include a Glossary section to provide short, understandable definitions for users who may not be familiar with the terms or concepts described on the page.

More information on how to structure and format this section is available in the bulletin guidance.

If the main purpose of the article is to provide a glossary of terms and definitions for a topic, use a methodology page instead of an article.

Data sources and quality

Provide a summary of important quality and methodology information about your article in a Data sources and quality section. This section should only be included in analysis articles, as it is unlikely that information articles will contain data. Using a standard heading for this section will create consistency for users and help identify articles from other content types, such as statistical bulletins.

Highlight any important information about the data source and collection method, as well as any caveats about the quality of the data. Use clear subheadings to break up the content and guide users through the information by topic. For example, use subheadings such as Data collection, Sample size, and Comparability with other sources.

Keep this section brief and link to any existing methodology information using clearly written link text rather than duplicating the content and increasing the word count for your article.

Future developments

Our article audit identified that articles are often used to explain recent or upcoming changes to users and how these may affect the data. 

Include any upcoming changes or developments in a “Future developments” section. Using this standard section heading will make sure users know where to find this type of information in the different types of articles. It will also ensure consistency on the website.

Use clear subheadings within the section to help guide users to the information they need.

Related links

Research tells us that users have two separate needs from related links: to go into more detail or to find broader but related content. All articles include a Related links section to help users access this content.

Include between three and six links in this section. These links should be to:

  • any related bulletins or articles (either published on the same day or previously)
  • any methodology pages that provide further detail on the data source, method or project
  • recent ONS publications that also reference this topic 
  • relevant articles that are published by other official organisations

Links should help users get directly to relevant content. Do not link to: 

The content design team can use analytics to help you choose which links to include – email

The Related links section should be formatted in the same way as the Related links section of a bulletin. More information on how to format your links is available.

How much to write

Analytics show that users spend an average of four minutes looking at an article page on the website. A typical person could read around 900 words during this time. Aim to limit the amount of analysis on the page to be shorter than this.

We also know that users only read the first 25% of a page, with most only reading the Main points. Put your most important or interesting information first to make sure users see it; think of the inverted pyramid when writing your article.

If you need to write more than this, consider who you are writing for, and how much value they would get from this extra content. We know that even expert users value getting information from the ONS website quickly and clearly. Consider splitting your content into separate articles and linking between them. This will give your articles a clearer focus.

Cite this article

Writers and academics use citations to tell users that certain material in their work came from another source.  

We need to include a “Cite this article “section in our articles. This will help writers and academics to find the information they need, and cite our releases accurately and consistently. 

The “Cite this release” section will be the final section of the article and included in the table of contents. It needs to be formatted as:  

X. Cite this article 

Office for National Statistics (ONS), released XX Month 20XX, ONS website, content type, Title: edition with link embedded 

Do not include a citation in a digital content article, as these are aimed at inquiring citizens rather than academics and expert users.

Next section: What to avoid in your article