Nanonets OCR Reviews

This software is really impactful, easy to adopt, easy to make a custom model, flexible to deploy

limited output when implemented on-premise

Probably one of the easiest softwares to use in the industry. Anyone without prior experience can upload data and extract it with their solution.

There should be better documentation on the Cancellation Policy, but besides this none.

Nanonets is one of the best OCR software I have ever used. I like this platform the most because of its high accuracy models. I am very happy for using this platform into our product.

I didn't find any reason to dislike the platform. The features this platform provides are excellent, and I would recommend this platform to everyone who needs to automate the data extraction process from various documents.

comprehensible technical documentation for all products
easy to use and easy to train the models in Invoice OCR product

the design part in "Invoice OCR" product.

Nononet is easy to use. I am in the teaching profession. I had to demonstrate how the procurement department can make use of Nanonet to MBA students. I provided them with some scanned bills to train the model and later another set to test. It worked like a magic. My 54 students were happy and satisfied.

Probably, Nononet can provide some sample datasets so that learners will feel comfortable. Also use-cases! They may consider providing more facilities to instructors so that they can teach this powerful tool even better.

High accuracy without standard limits, that allowed to process almost any image or document
User-friendly output formats directly or via modern API
Possibility to train own model or use pretrained one

Just one tip to add some SQL extension as output formats

easy to use, UI friendly, non-developer can use this apps, can integrate into many API

not compatible to mark sample when it's rotating 20º, and can't do flexible shape of marking

I really liked the interface and how it was straightforward to get my documents into the system, and get my data out of the uploaded document. I also really liked the options for massaging the extracted data before finalizing it. It is clear that Nanonets has considered a lot of the pain points around extracting data from PDF files ( for example ).

As I mentioned before, the price point is very high. I understand that this is a tool mainly aimed at businesses for regular text extraction, but I am a researcher and have less money and don't need to use the software as much, would be good to have an affordable paid options.

Helped us save time by automating data extraction at scale

Sometimes you need to have calls with the team to help them annotate unusual parts of a document

The solution is able to detect multiple languages at the same time with fantastic results.

I wish it could automatically extract documents from images, like extracting id cards from image.

The OCR annotation was accessible, quick and accurate 👌

Their customer service was also proactive in assisting me with the initial annotation for the training models, with a speedy turnaround that enabled me to extract and export my data within 24 hours of registering my account.

I didn't like that the table annotation (list items) has opaque column headers, which means that you have to
zoom in and readjust the table size when the column headers happen to be floating over a valid (legible) data row.

Nevertheless, I pinged the team about this and they were aware of the issue - they have it on their roadmap
for UX improvements.

The system can be calibrated to recognise information from any document type! It takes about 60 minutes to calibrate and you can start reading these documents automatically!

I was not sure, whether Nanonets provides cloud services in european data centres?

We have many different data feeds we intake and perform ELT processes, but could only get PDF files for a new custom program. Nanonets software approach made it easy for me to fully automate the conversion from PDF to normalized data we can quickly ingest.

Integration was very easy. Once we customized the export with Nanonets it only takes a few short minutes daily to process. Once this is out of pilot, we will fully automate with no human touch needed.

Nanonets it's the most powerful OCR software.

Probably nothing. I think it is also useful and smarty.

Ease of use, ability to intake large volumes of documents, reasonable price point

No issues faced so far, the experience has been excellent.

What stood out most to me was how user friendly it is. It's an excellent user experience from start to finish. If you have a game plan for what and how you want to process something it is as easy as 1 2 3. It is very easy to customize to our own forms and it gets better at it every time we use the software. Editing mistakes within the software is also very easy and way faster than when we would normally process the forms by hand.

Nanonets is based on AI that needs you to select the text to extract as closely as possible, whereas I feel for our handwritten forms it would be great to set markers for fields and after that, have the AI read all the text within that field. Just to avoid it mixing things up.
What would also be good is to be able to set field properties e.g.: numbers only, text, yes or no, like you would when using tables.

Had the setup in place for collecting info from financial docs.

Not much to dislike works pretty well from the beginning

High accuracy, easy to implement, excellent support team! Awesome annotation team!

We need post-processing the invoice data for creating structured data.

Agile ability to meet customers needs with speed.

None.

the best part is OCR boxing to train models by labeling. I don't know any other product offering this kind of service. UI is great too.

Sometimes it just stops working suddenly. Not facing these issues for some time.

This software was the best I could find for reading data from blurry PDF tables. It was able to infer the values of blurred or partial numbers fairly well, and after proper training it was effective at picking out the table structure. The turnaround time was quick - I was able to read and output hundreds of tables in a relatively short period.

The model training process is a little tedious and involved delineating the table structure by hand. This would have been a quicker process with a smoother user interface.

Easly to create a model and implementation where ever we can.

I think there is no least about this software. Everything is good.

Very high reliability when using documents of a fairly uniform template.

Annotation is quite time consuming but I cannot think of a way this could be avoided other than using your paid annotation service.
I think it is a necessity associated with the process.

Our use case was performing OCR on around 750 pdfs a month of the same format but each with their own unique data. We use a zapier integration to populate the data and upload to our marketing platform of choice.

Does exactly what it is supposed to do.

The support team is amazing. Helped us get setup the whole way and checks in regularly to see how things are working or if we need anything.

A small learning curve but the support team trained me well and now I am able to manage myself.

We have an issue in our firm, where bad data can cost us more than a lost receipt. Nanonets helped put a stop to that problem, by allowing our receipts to be OCR and searchable. No longer are we losing or missing data / bad data inputs. The computer does it for us

Unfortunately the price is a little bit on the steep end, but if it wasnt saving me money I wouldent be writing this review.