Using DeepL as a free tool to correct your writing

Photo by Waldemar Brandt on Unsplash

If you’re a non-native English writer, you might be feeling a bit left behind in the language tech revolution.

These days, Grammarly and Hemingway are the go-to tools for writing correction, but they’re a bit disappointing when it comes to dealing with typical challenges when English isn’t your first language.

Questions like this:

  • Which word sounds more natural? Is it widen your outlook, deepen your outlook, or broaden your outlook?
  • Do you need present perfect (they’ve given), past simple (they gave), or past perfect (they’d given)?
  • What’s the right word order or structure? Is it What you think should I do? or What do you think should I do? or something else?
  • What’s the right phrase? Is it spend time, spend your time, take time, or take your time?

There are paid tools such as Instatext and Trinka that can help with things like this, but… well, they’re paid.

Here’s a quick hack that gets you an alternative draft of a short piece of writing for free. This is great if you want your English to sound more natural and flow better. (If you’re looking for help with structure or shaping your ideas in the first place, unfortunately this won’t help.)

All you need is:

  1. A piece of writing up to 5000 characters long, which averages about 1000 words. If your piece is longer, you’ll have to break it up into sections.
  2. (Important!) A reasonable grasp of English, so you can feel which of the two versions (yours or the suggested one) sounds better.

Here are the steps.

1: Hop over to DeepL’s online translation engine (deepl.com).

This hack uses reverse translation as a tool to correct your writing. I like the DeepL engine as it gives natural results, but you can use a different translation site if you prefer.

2: Paste your text (up to 5000 characters) in the left-hand window.

DeepL translation interface with text translated from English to French

DeepL should detect that your text is in English and will automatically produce a translation, usually in French or German. I prefer French, but you can experiment. Basically, you need a widespread language that will have had plenty of investment in natural language research.

3: Click the two-way arrow button to re-translate your writing into English.

DeepL translation interface with the same text translated from French to English

This time, DeepL translates the foreign-language version of your writing back into English.

4: Check your original writing against DeepL’s reverse translation.

Most of the time, the process will have smoothed out tense issues and given you some good ideas for alternative sentence structure and wording. It might also have misunderstood some things, so it’s up to you which version you choose for your final piece.

An example of this hack in action

For this example, I’ve taken part of an article about bug testing in software development. It has some problems that make it sound unnatural. You can see the original on the left and DeepL’s reverse translation on the right. DeepL’s changes are shown in purple.

Original version:

In IT, the demand for agile application development and faster delivery increases day by day. A decade ago, manufacturing downtime for a day might not be catastrophic event, but in today’s world, everything is connected to everything else. Hours and even minutes until a bugfix reaches production can cost thousands of dollars.

Early finding and resolving bugs in the development process and limiting the needlessly done work due to misunderstood requirements is crucial to limiting development costs.

DeepL reverse translation:

In IT, the demand for agile application development and faster delivery is growing by the day. Ten years ago, a one-day downtime might not be catastrophic, but in today’s world, everything is connected to everything else. The hours, or even minutes, that pass before a bug fix reaches production can cost thousands of dollars.

Finding and fixing bugs early in the development process and limiting unnecessary work due to misunderstood requirements is critical to keeping development costs down.

The DeepL version flows a lot better, though it still has some problems. If it was your text, maybe you wouldn’t like some things and you’d prefer to keep your original. This isn’t a perfect process, but it’s free!

In comparison, here’s the DeepL version alongside Grammarly‘s recommendations (in blue). I ran the text through Grammarly’s paid version and accepted all the suggestions.

DeepL reverse translation of original:

In IT, the demand for agile application development and faster delivery is growing by the day. Ten years ago, a one-day downtime might not be catastrophic, but in today’s world, everything is connected to everything else. The hours, or even minutes, that pass before a bug fix reaches production can cost thousands of dollars.

Finding and fixing bugs early in the development process and limiting unnecessary work due to misunderstood requirements is critical to keeping development costs down.

Grammarly correction of original:

In IT, the demand for agile application development and faster delivery increases day by day. A decade ago, manufacturing downtime for a day might not be a catastrophic event, but everything is connected to everything else in today’s world. Hours and even minutes until a bugfix reaches production can cost thousands of dollars.

Early finding and resolving bugs in the development process and limiting the needlessly done work due to misunderstood requirements is crucial to restrict development costs.

And here’s the same comparison with Instatext (a paid tool):

DeepL reverse translation of original:

In IT, the demand for agile application development and faster delivery is growing by the day. Ten years ago, a one-day downtime might not be catastrophic, but in today’s world, everything is connected to everything else. The hours, or even minutes, that pass before a bug fix reaches production can cost thousands of dollars.

Finding and fixing bugs early in the development process and limiting unnecessary work due to misunderstood requirements is critical to keeping development costs down.

Instatext correction of original

In IT, the demand for agile application development and faster delivery is increasing day by day. A decade ago, a one-day production outage might not have been a catastrophic event, but in today’s world, everything is connected to everything else. Hours and even minutes for a bug fix to reach production can cost thousands of dollars.

Finding and fixing bugs early in the development process and limiting unnecessary work done due to misunderstood requirements is critical to limiting development costs.

And we’re done

As you can see, Grammarly helped the least with the text in this example. Instatext and DeepL changed similar things, but as DeepL doesn’t cost you, it’s definitely worth a look as a free tool to correct your writing.

I hope this quick hack has given you a new way to look at your writing, and I encourage you to try it out for yourself. The only challenges are that you have to decide which version sounds better, and machine translation isn’t always perfect. Don’t forget that there’s also plenty more a human editor can do to help you make the right impression!

I’d love to hear your comments about the tools you use to check your work. Good luck with your next piece!