Guide

How to get CVs into a shortlist spreadsheet

You advertised one role and forty CVs came back, as PDFs, Word files and the odd phone photo. You want one row per person so you can sort and shortlist. Every page currently ranking for this is a resume parser selling itself, and none of them start with the cheapest fix, which is not to extract anything at all.

Short answer

For your next ad, ask for the details you shortlist on in the application itself, through screening questions or a form that writes straight to a spreadsheet, so there is nothing to extract. For a pile you already have, typing six or seven columns by hand is fine for a few dozen CVs, and a free parser such as Affinda's handles bigger piles, as long as you are comfortable uploading applicants' personal details to it.

Decide your columns before you open a single CV

Most of the time lost here is not typing. It is reading each CV without knowing what you are looking for, then reading them all again once you do.

Write down the handful of things that actually decide whether someone gets a call. For most small business roles that is a short list: name, email, phone, suburb or whether they can get to you, the licence or qualification the job requires, relevant experience, availability, and a column for your own note.

Resume parsers will happily give you a full work history, every skill keyword and every school. That is a wide spreadsheet you will never read, and more personal information than you need to hold.

Seven columns you will use beat thirty you will scroll past. Decide them first, and every method below gets faster.

Method 1: collect the data in the application, so there is nothing to extract

This is free, it fixes the problem for every future role, and no extraction tool can beat it, because the applicant types their own details into the right box.

Screening questions on the job board
SEEK and Indeed both let you add questions to an ad, such as whether the applicant holds a particular licence or has the right to work in Australia. Their answers sit alongside the application, so you can filter before you open a CV. If the job board is where your applicants come from, set these up first.
Google Forms into Google Sheets
Build a short application form, link it to a sheet, and every application becomes a row as it arrives. One catch: if the form asks for a file upload, Google makes every applicant sign in to a Google account, which will put some people off. Asking for a link to their CV instead avoids that.
Microsoft Forms into Excel
Responses export to Excel in the same way. But Microsoft only allows a file upload question when the form is limited to people inside your own organisation, so it cannot collect CVs from the public. Use it for the questions and take the CV by email or a OneDrive file request.

If you hire more than once a year, this is the real answer. It costs an hour to set up, and the next shortlist arrives already in a spreadsheet.

Method 2: type it, for a pile you already have

Nobody likes hearing this, but for a few dozen CVs, typing seven columns is often faster than setting up and checking a tool. You are reading each CV anyway to decide who to call. The typing is the small part.

  1. 1Set up the sheet with your columns from the first section, plus a column for the file name so you can find the CV again.
  2. 2Rename each file to the applicant's name as you go, so the folder and the sheet match.
  3. 3Fill in a row for each CV in one sitting. Use a dropdown for yes or no columns like licence held, so you can filter on them later.
  4. 4Add your own score or note last, after you have seen the whole pile, not as you go. The tenth CV changes what you think of the second.
  5. 5Sort or filter, and you have your shortlist.

Free and reliable, and you end up knowing the candidates better than any tool would. It stops being sensible somewhere past the point where you are filling rows rather than reading people.

Method 3: a free or cheap resume parser, for large piles

Resume parsers read a batch of CVs and return one row per person. They are built for recruitment agencies, which is why they are good at it.

Affinda has a free parser on its website that takes PDF, Word and image files, with no account, and exports a CSV that opens in Excel or Google Sheets. At the time of writing it allows up to 100 CVs at once and a daily cap of 150 without an account. Of the free options we looked at, it is the most generous, and if you have a large pile and no privacy concerns, it is the one to try first.

Datanem charges per document after some free ones, and says it stores files in the EU and deletes them after a set period. Several other sites offer free bulk conversion with much smaller limits.

We have not measured the accuracy of any of these, so we are not repeating anyone's numbers. Check the first ten rows against the CVs before you trust the rest. Dates, job titles and phone numbers from unusual layouts are where they usually slip.

The right tool for a big pile, as long as you have thought about the next section first.

The part the parser pages leave out: these are other people's personal details

A CV holds a name, address, phone number, work history and sometimes a date of birth. Uploading forty of them to a free website is sharing forty strangers' personal information with a company you have not checked.

Many Australian small businesses are outside the Privacy Act, because the general small business exemption covers most businesses with annual turnover of $3 million or less, although there are exceptions, such as businesses that provide a health service. If your business is covered by the Act, note that the employee records exemption does not apply to applicants you do not hire. The OAIC says so directly in its guidance on that exemption.

Whether or not the law applies to you, the sensible habits are the same.

Read where the files go
Before uploading, find the parser's statement on where files are processed and how long they are kept. Affinda says its free tool does not keep documents after parsing. Datanem says files are stored in the EU for a set period. If you cannot find a statement, do not upload.
Do not keep the pile forever
Once the role is filled, delete the CVs and rows of the people you did not hire, unless they have agreed to be kept on file for future roles. A folder of old applications is a liability, not an asset.
Keep shortlist columns about the job
Leave out age, photos, marital status and the like. You do not need them to decide who to call, and holding them invites the question of whether you used them.

When an applicant tracking system is the better answer

If you are hiring every month, or several people are reviewing the same applicants, a spreadsheet becomes the wrong tool. Applicant tracking systems such as Workable, JobAdder and Breezy HR collect applications, read CVs, track who is at which stage and send the rejection emails.

They are subscriptions, and they are more than a business hiring once or twice a year needs. But if you find yourself emailing a spreadsheet of candidates between three people, it is time.

One role a year: a form and a spreadsheet. A role a month with a team deciding: an applicant tracking system.

Where Once fits, honestly

Once, our own tool, reads PDFs, photos and Word files and writes the details you point at into a spreadsheet, and it runs on your own computer, so the CVs are not uploaded anywhere. That last part is the only reason to choose it over a free parser for this job.

It was built for invoices and similar documents. We have not tested it on a large batch of CVs, and CV layouts vary far more than invoice layouts do. If you try it on CVs, check the first rows as you would with any tool, and tell us how it went.

Common questions

How do I extract data from CVs into Excel for free?
For a few dozen, type the six or seven columns you shortlist on by hand. For a larger pile, Affinda's free resume parser accepts PDF, Word and image files and exports a CSV that opens in Excel. Check the first rows against the CVs before relying on the rest.
Can Excel read CVs on its own?
No. Excel can import tables from a PDF, but a CV is a layout, not a table, so the import gives you fragments. You need to type it, use a parser, or collect the details through a form so they arrive as rows.
Is it safe to upload CVs to a free resume parser?
It depends on the parser. CVs are other people's personal information, so read where the tool processes files and how long it keeps them before uploading. If there is no clear statement, do not upload, and use a manual method or a tool that runs on your own computer.
How long should I keep CVs from unsuccessful applicants?
Only as long as you need them for that role, unless the applicant has agreed to be kept on file for future roles. If your business is covered by the Privacy Act, the employee records exemption does not apply to people you did not hire, so the normal rules on holding personal information do.
Can Google Forms collect CVs?
Yes, with a file upload question, but Google then requires every applicant to sign in to a Google account. Asking for a link to the CV avoids that. Microsoft Forms only allows file uploads from people inside your own organisation.

Related

Solven builds custom software and automation for Australian businesses. If you have a job like this that no tool quite fits, that is the work we do.

Get a free teardown