Step 5. Check your data quality

Test whether your 360Giving data is all in the correct format – also known as ‘valid’ data.

Test your data before you publish

Once you have prepared your file of grants data, the next step is to check that it is valid 360Giving data by using our Data Quality Checker. 

In order to be ‘valid’, the file must include the 10 core fields, and all the data must be correctly formatted. The file must either be a spreadsheet in OpenDocument Spreadsheet (.ods), Excel Workbook (.xlsx) or CSV (.csv) format, or it must be a JSON file that complies with the 360Giving JSON schema (.json).

Only valid 360Giving data can be combined with other published data and be included in 360Giving tools, such as GrantNav and GrantVis. 

If the data isn’t valid, it means there are problems that need to be fixed before it’s ready to publish.

Name your 360Giving data file

This is a good point to think about the name you’ll use for your data file.

• Include your organisation’s name in the file name to make it easier for people who download the file to know who published the data.
• If you split your grant files by programme or year, you can include this in the file name, if it doesn’t make the title too long.
• If you’re planning to publish all your grants data in a single file, and will be adding more data over time, you should choose a general name that covers everything.
• Avoid including file version numbers or dates that will need to be changed if you update the file.

How to use the 360Giving Data Quality Checker 

You can use our Data Quality Checker by uploading your file directly, or by copying and pasting a link to your file if it is already saved or ‘hosted’ somewhere online. 

If you see an error message saying that the tool can’t process your data, or that it doesn’t recognise the file type, try again using a different file format.

Understand your Data Quality Checker feedback

Once you have uploaded your file, you’ll see feedback on some key points. 

Straight away, you’ll see three cards that will give you a quick look at the quality of your data. If your data isn’t valid, you’ll see a question mark instead of a number on the ‘Potential accuracy issues’ and ‘Usefulness opportunities’ cards.

There are two icons that you’ll see in the feedback:

• A red exclamation mark means there are serious accuracy issues. You should fix these issues before you publish your data.
• A yellow exclamation mark means there are probably some things you can do to improve the accuracy or usefulness of the data.

You may have feedback in up to six different areas: 

1. Summary
2. Additional Fields
3. Conversion errors
4. Validity
5. Accuracy
6. Usefulness. 

Click on the tab for each area to see the feedback in detail.

Summary tab

The summary tab shows basic information about what’s in your file – how many grants, funders, recipients, and the date range and total amount of the grants given, broken down by currency.

A green or red box at the top shows whether the data uses the 360Giving Data Standard correctly. Green means the data file is valid, and red means it’s not valid.

The feedback in the other tabs will help you investigate what is causing any issues.

Check that the summary looks right before you go any further. Even if everything in the summary tab is correct, there may be issues highlighted in the other tabs that you need to look at.

• This file contains. A preview of your data in a table. Plus, the number of grant, funding organisation and recipient identifiers. You can click these links to see a list of each.
• Share. Here, you can find a unique URL for your feedback, which you can share with other people in your team and with our Helpdesk as a link. Each tab also has its own unique URL. The links will expire after seven days.
• Download. Here, you can download a version of your spreadsheet file converted into JSON format. If the file you have been testing is already in JSON format, you can change it to a spreadsheet in the ‘Data Conversion’ section. 

Additional Fields tab

This tab will appear if your data includes additional fields which don’t use 360Giving Data Standard field names. You can check that they are supposed to be there,  and they’re not the result of spelling or formatting mistakes in the Standard field names.

Conversion Errors tab

To test the data in a spreadsheet file, the Data Quality Checker converts the data into JSON, a different, more technical format.

If you see a message which says ‘Data conversion unsuccessful’, it means there are some issues with the data that stopped it from being converted into JSON.

One of the most common issues is duplicate identifiers. All grant Identifiers must be unique so if two grant records have the same Identifier, the Data Quality Checker will try to merge the two records, thinking it’s the same grant! If there are any differences between the two grant records (because they are actually different grants), this will cause an error for each field which is different.

• If the same grant is included in your file more than once, delete all the duplicates until only one record for each grant is left.
• If different grants have the same identifier, you’ll need to find out if that’s the case only in your data file, or also in the system that you exported the data from originally. You might need to add an extra letter or number to make all your grant identifiers unique. For example 360G-ExampleFnd-001A and 360G-ExampleFnd-001B

Validity tab

You’ll see one of two messages:

• ‘This data uses the 360Giving Data Standard correctly’ means the data is valid 360Giving data.
• ‘This data does not use the 360Giving Data Standard correctly’ means the data isn’t yet valid – there are problems which need to be fixed.

If there are errors, you’ll see them grouped under different types of issues, with an Error Count to show how many grant records are affected by the problem. If there are three or fewer grant records with the same error, the location of the issue will appear in a preview, with the sheet name and row number for the grant.

If there are more than three grant records with the same error, there’ll be a hyperlink to a full list of their locations in the Error count.

There are four types of validity errors:

Missing fields

This means that data is missing in the 10 core fields of required information. Look at the feedback to find what’s missing. If there’s no data available for the missing fields in your system, you may need to remove the grant from your data until you can add the rest of the required information. Tip: Watch out for spelling mistakes in field names! 

Incorrect Formats

Some of the data formatting isn’t right. For example, it could be because:

• A date is not in the correct format.
• An invalid ‘URL’ has been found, which means there’s a problem with a website address in your data.

Codelist Errors

Some fields in the 360Giving Data Standard use codelists, which means you must use one of the standard values from one of our codelists when you use these fields. If you put something else in one of these fields, it will be flagged as an error in this tab, with a link to the codelist you need to use. You’ll also get an error in the ‘Other validation errors’ section: Invalid code found in [field name].

Other

Other errors include text in number fields, numbers in text fields, or the wrong dates under grant duration. The error descriptions link to more information about each error.

Accuracy tab

This tab highlights areas where your data may need to be corrected or need further attention. 

The feedback under this tab does not mean your data is invalid. It can be ignored when not relevant or until you can improve your data quality. However, it is important that any feedback marked by a red exclamation mark is checked and corrected whenever possible.

Organisations

More than one funding organisation

You’ll see this if your data file has more than one unique Funding Org:Identifier. Unless you are publishing on behalf of multiple funders, you need to fix this.

Charity and company numbers

If your charity or company number fields include data other than the numbers (like filler text) or extra or missing numbers, you will see this error message: ‘XX grants have a value provided in the Recipient Org:Charity Number column that doesn’t look like a UK charity number’.

You’ll get a similar message about your Recipient Org:Identifier data, because these are created based on the charity or company numbers in your data.

To fix these issues, correct the data in your system’s charity or company number field, and export your 360Giving data again.

Funding or Recipient Org:Identifier that does not draw from a recognised register

This means an organisation identifier has an unrecognised prefix. See our guidance on organisation identifiers for further help with this.

Grants

If any grants have a zero Award Amount, the feedback will highlight them. Check these and remove them if needed.

Dates

If you have a ‘minus duration’ error, it’s possible that the grant start and end dates are the wrong way around.

Data protection

The tool flags text that resembles email addresses, postcodes, and demographic information associated with individuals, in case it’s personal data which shouldn’t be published.

Usefulness tab

This tab highlights ways that your data could be made more useful. It doesn’t mean your data is invalid. If the feedback isn’t relevant, you can ignore it, or you can make a note of things you can do in the future to improve your data quality.

Further resources

• About identifiers in the 360Giving Data Standard
• Read our guidance on data protection
• See our detailed guidance on the Usefulness tab feedback

Resolving data quality issues

The Data Quality Checker will show you a few things that might be wrong, but it will also highlight opportunities to make your data more useful and usable.

You may find that there is some information you can’t include because it isn’t available in your data, or the data quality is currently too poor. Publish what you can, and make a note of any recommendations, so that you can improve your data quality in the future.

If your data is poor quality, you may want to look into providing more training or guidance for people who are collecting or entering data into your systems. You might be able to add validation to some fields so that any errors will be flagged immediately.

It’s also a good idea to schedule routine data cleaning, if you can. Good quality data will be more useful for your organisation’s needs, as well as for GrantNav and other tools.

If you would like help to get your data in order, we can offer you consultancy support.

Data Quality Checker security

When you upload your data to the Data Quality Checker, you will get a feedback report with a private link or ‘URL’, which means only those with the link can access the page and the data. 

If your data isn’t ready to be shared publicly, then you should only share the link with people who are allowed access to the data. As long as you do not put the URL on a public webpage, then it won’t be possible for people to find it by accident.

Files are deleted automatically from the Data Quality Checker after seven days. 

Checklist 

You’ve reached the end of the Stage 2: Prepare! Now complete the checklist:

Have you…

• Chosen what information to publish, and picked which fields you will use?
• Decided how often you will update your 360Giving data once it’s published?
• Extracted the data you want to publish from your grants management system?
• Cleaned your data? Checked for gaps and errors?
• Converted your data to the right format for the 360Giving Data Standard?
• Removed any personal or sensitive data?
• Completed the privacy checklist?
• Informed grantees about your plans, if you choose to tell them?
• Checked your data using our Data Quality Checker and corrected any errors?