GITNUXSOFTWARE ADVICE
Data Science AnalyticsTop 10 Best Online OCR Software of 2026
Top 10 list compares online ocr software tools for scanning, text extraction, and file handling, focusing on practical tradeoffs for teams.
How we ranked these tools
Vendors set the topic, but our editors set the inclusion criteria and the evaluation framework before research starts.
Search agents gather the source set and the review record, including transcriptions of video reviews and podcasts.
Framework: Features 40% · Ease 30% · Value 30%
Gitnux may earn a commission through links on this page — this does not influence rankings. Editorial policy
OCR.space is the best pick for teams automating OCR because its confidence scoring with bounding boxes helps flag the exact uncertain characters for review, whereas OnlineOCR is the better fit for SMBs that want quick per-page scan-to-text checks in a simple web workflow.
Editor’s top 3 picks
Three quick recommendations before you dive into the full comparison below — each one leads on a different dimension.
OCR.space
Confidence scoring paired with bounding boxes makes it practical to flag specific uncertain characters for review.
Best for: Fits when teams need automated OCR for scans with confidence-based exception handling.
OnlineOCR
Editor pickPer-page OCR output for multi-page documents simplifies spot-checking and reprocessing only failed pages.
Best for: Fits when teams need fast, web-based OCR text extraction without building an OCR API pipeline.
Convertio
Editor pickSearchable text layer downloads produced from uploaded scanned files.
Best for: Fits when teams need scan-to-text outputs quickly for manual review and indexing.
Comparison Table
OCR.space
API-firstFree OCR API and web interface provided by OCR4U.
Confidence scoring paired with bounding boxes makes it practical to flag specific uncertain characters for review.
OCR.space is practical for teams that need fast OCR conversions from image formats like JPG, PNG, TIFF, and multi-page uploads that OCR jobs can process as a batch. It includes character-level confidence scoring and exposes bounding coordinates so downstream systems can route low-confidence spans to manual review or apply regex post-processing. The API surface supports job-based processing with status checks, and the output options include text and PDF with a text layer suitable for search.
A tradeoff appears in governance and control depth compared with enterprise OCR stacks that offer tenant isolation features and audit logging, since OCR.space is centered on service calls rather than admin-grade RBAC and audit trails. OCR.space fits best for document ingestion pipelines that need straight-through extraction for clean scans and a manual exception queue for noisy pages.
- +Returns bounding coordinates and confidence values for low-confidence routing
- +Provides searchable PDF output with an embedded text layer
- +Supports batch processing for multi-image ingestion workflows
- +Job-based OCR API supports automation with status polling
- –Limited enterprise governance features compared with managed cloud OCR services
- –Handwriting and degraded-document accuracy can lag for highly noisy scans
- –Table structure extraction is less complete than form-first extraction platforms
- –Concurrency and throughput depend on service limits and file sizing constraints
Customer support operations
Convert ticket screenshots to searchable text
Faster retrieval from document archives
Document ingestion engineers
Automate OCR API extraction jobs
Lower manual processing volume
Show 2 more scenarios
KYC operations teams
Extract fields from ID image scans
Fewer transcription errors
Language selection and confidence values support routing uncertain characters to review.
Accounts payable teams
OCR invoices for text layer search
Improved document findability
Invoice images are converted into searchable PDF for quick internal document search.
Best for: Fits when teams need automated OCR for scans with confidence-based exception handling.
OnlineOCR
SMBFree web-based OCR tool supporting 46 languages and multiple output formats.
Per-page OCR output for multi-page documents simplifies spot-checking and reprocessing only failed pages.
OnlineOCR targets text extraction workflows that need quick turnarounds from image inputs such as JPG and multi-page PDF files. The main workflow is upload, OCR run, and download of extracted text, which reduces the need to build a document ingestion pipeline. Language selection helps recognition on documents that contain accented Latin characters and non-Latin scripts, and it supports exporting results for downstream cleanup.
A tradeoff appears in automation depth compared with major cloud OCR APIs, because OnlineOCR exposes a web-based workflow rather than a rich REST endpoint with job controls. The best fit is a manual or semi-automated flow where staff upload document scans, review the output, and then copy text into business systems.
- +Straightforward upload-to-text workflow for scanned PDFs and image files
- +Language selection improves recognition for non-English document scans
- +Per-page processing makes multi-page review faster
- +Multiple export options simplify copy into editors and systems
- –Limited integration surface for enterprise automation and workflow orchestration
- –OCR results can require manual cleanup for low-quality or noisy scans
Operations teams
Convert scanned SOP PDFs to text
Reduced manual transcription time
Customer support
Extract text from uploaded proof scans
Faster ticket resolution
Show 2 more scenarios
Legal teams
Extract citations from image-only filings
Improved search and reuse
Runs OCR on multi-page scans and exports text for review and copying into drafting tools.
Research analysts
Digitize printed tables from scans
More data captured from scans
Converts scanned document pages into text for downstream sorting and verification workflows.
Best for: Fits when teams need fast, web-based OCR text extraction without building an OCR API pipeline.
Convertio
SMBOnline file converter that includes an OCR module for images and PDFs.
Searchable text layer downloads produced from uploaded scanned files.
Convertio’s workflow centers on uploading scan files, running OCR, and downloading a result that can include a searchable text layer for document-like outputs. It is designed for straightforward operation where the primary variable is input file quality and language selection for recognition. The UI also fits teams that want human review after OCR because it returns outputs that can be checked page by page.
A tradeoff is that Convertio’s OCR is primarily mediated through its web workflow instead of a deep automation surface for high-volume ingestion and concurrent processing. It fits situations like small to mid-volume document batches for back-office teams that need text extraction from scans and quick turnaround for manual verification.
- +Web workflow keeps OCR and output download in one place
- +Batch-style job handling supports repeated document processing
- +Searchable text layer outputs help downstream document lookup
- +Multi-format input reduces friction from mixed scan sources
- –Limited visibility into recognition tuning and low-confidence routing
- –Automation and API integration are not the primary workflow path
Accounts payable teams
Extract invoice text from scans
Faster invoice data validation
Legal ops teams
Make discovery scans searchable
Quicker document finding
Show 1 more scenario
Customer support teams
Digitize uploaded forms and letters
Reduced manual transcription
OCR turns photo or scan inputs into usable text for case notes.
Best for: Fits when teams need scan-to-text outputs quickly for manual review and indexing.
i2OCR
SMBFree online OCR service supporting over 100 languages.
Per-character confidence indicators for low-confidence flagging and faster manual correction routing.
i2OCR serves as an online OCR workflow for turning image and scan files into extracted text, with outputs meant to fit document ingestion pipelines. Recognition support includes multi-language processing and confidence-style outputs that help route low-confidence content to review.
- +Provides multi-language OCR to reduce manual reprocessing
- +Returns structured OCR outputs that support automated downstream parsing
- +Supports batch-style processing for repeated document ingestion
- +Includes per-character confidence indicators to help triage errors
- –Accuracy can drop sharply on highly noisy or skewed scans
- –Layout handling is weaker for complex forms with dense tables
Document processing operations
Batch convert scans into parsed text
Faster indexing and retrieval
Records and compliance teams
Extract text from multi-language scans
Searchable record copies
Show 2 more scenarios
Content review routing teams
Flag low-confidence OCR outputs
Lower manual rework
Route low-confidence content for review using confidence-style results for downstream workflows.
Document ingestion pipeline owners
Return machine-readable OCR results
Cleaner downstream ingestion
Feed OCR outputs into parsing and indexing steps using structured, machine-readable results.
Best for: Fits when teams need recurring OCR for mixed document batches and automated extraction with review for low confidence.
PDF24 Tools
SMBSuite of free PDF tools including an online OCR function.
Automatic embedding of recognized text into a searchable PDF output within the same web workflow.
PDF24 Tools, accessed through tools.pdf24.org, provides a browser-based OCR workflow focused on turning image-based documents into text and searchable PDF outputs. The tool supports multi-page inputs and embeds the recognized text layer so the result remains usable in document viewers and search.
- +Searchable PDF text-layer embedding from scanned documents
- +Web-based batch handling for multi-page files without client software
- +Built-in image cleanup steps like deskew and rotation correction
- +Layout-aware output that preserves reading order better than basic text OCR
- –No documented REST OCR API for job automation and deep integration
- –Limited structured extraction for receipts and invoices compared with capture-specific tools
Archive and records teams
Search scanned documents in PDFs
Searchable digital documents
Office and administrative teams
Package invoices into a single PDF
Consolidated document package
Show 2 more scenarios
Small business operations teams
Standardize mixed files into PDFs
Consistent formats for teams
Convert documents and images so files become standardized for sharing and downstream use.
Small business teams
Create searchable PDFs with controls
More secure shareable PDFs
Create PDFs from printable sources, then apply OCR so the output is shareable with access protection.
Best for: Fits when teams need fast searchable PDFs from scanned docs without building an OCR ingestion service.
Tesseract.js
API-firstPure JavaScript port of the Tesseract OCR engine running in the browser.
HOCR output that preserves word bounding boxes, enabling custom text-to-region mapping for indexing and review tooling.
Tesseract.js brings the Tesseract OCR engine to browser and Node.js environments through a JavaScript SDK, which is distinct from cloud-only OCR APIs. It supports multi-language recognition via language data packs and returns results with word and character level bounding boxes, enabling downstream layout handling.
- +Runs in browser or Node.js for local OCR without an external OCR service
- +Supports multiple language packs with separate models for better script coverage
- +Returns layout coordinates suitable for custom zonal post-processing
- +Can emit HOCR markup to keep word boxes tied to recognized text
- –No built-in layout reconstruction for tables or forms beyond coordinate-based post-processing
- –Accuracy depends heavily on caller image preprocessing like deskewing and binarization
- –High-throughput batch processing requires custom concurrency control in application code
- –Complex workflows like human review queues must be built outside the library
Web app engineers
OCR in-browser from user uploads
No server OCR needed
Node.js document pipelines
Convert batches with HOCR output
Better page-aligned indexing
Show 2 more scenarios
Knowledge base indexing teams
OCR scanned PDFs for search
Searchable scanned content
Produce full text output for search indexing and coordinate data for retrieving snippets by location.
Multi-language OCR teams
Recognize multilingual documents
More accurate multilingual reads
Load language data packs and use bounding boxes to process word and character layout.
Best for: Fits when teams need embedded OCR in web apps and want custom preprocessing and layout logic without an OCR API.
Smallpdf OCR
SMBOnline PDF toolkit that offers an OCR feature for scanned files.
Searchable PDF output with an embedded text layer from uploaded scans directly in a web workflow.
Smallpdf OCR converts uploaded images and PDFs into searchable text and document files with embedded text layers. The workflow focuses on browser-based ingestion, page-level processing, and output formats that include searchable PDFs.
- +Browser workflow converts scans into searchable PDFs without separate tooling
- +Page-by-page OCR keeps multi-page documents manageable in the editing flow
- +Supports multiple languages for mixed-language document sets
- +Preprocessing improves OCR readability for skewed and noisy scans
- –OCR output options prioritize text layers over structured extraction fields
- –No clear controls for confidence thresholds or low-confidence routing
- –Limited automation surface compared with OCR API-first providers
- –Throughput is tied to interactive use rather than queue-based batch processing
Operations teams
Turn scanned PDFs into searchable files
Searchable document exports
Customer support teams
Extract text from incoming scans
Faster case triage
Show 2 more scenarios
Content management coordinators
Process multilingual documents for storage
More reliable archived text
Select multi-language recognition so archived documents get consistent OCR outputs for downstream use.
Small teams without OCR tooling
Avoid building an OCR integration
Less setup overhead
Use a browser workflow to produce readable outputs without setting up a local OCR pipeline.
Best for: Fits when teams need fast, browser-based searchable text from scans without OCR API buildout.
NewOCR
SMBFree online OCR tool that converts images to text without registration.
API job handling for asynchronous OCR batches with retrieval of extracted results for pipeline automation.
NewOCR is an online OCR service that converts scanned images and document files into extracted text and selectable output artifacts. Batch processing supports multi-page documents and workflow-friendly ingestion of common image and document formats.
- +API-first job flow fits document ingestion pipelines without manual file handling
- +Batch processing supports multi-page OCR runs for higher throughput
- +Multi-format input coverage supports mixed scan sources in the same queue
- +Output choices support searchable text layer building and downstream parsing
- –Complex layout reconstruction can degrade on low-contrast, noisy scans
- –Handwriting recognition quality needs tuned preprocessing for mixed-quality captures
Document ingestion engineering
Job-submit OCR for pipeline automation
Automated OCR processing
Back-office operations
Convert scanned documents into text
Reusable OCR text
Show 2 more scenarios
Data enrichment teams
Index OCR outputs for search
Improved search coverage
Use workflow-friendly OCR results to support downstream indexing and document retrieval.
Manual review coordinators
Generate selectable OCR artifacts
Quicker validation cycles
Create OCR outputs that support human review and correction after batch processing.
Best for: Fits when teams need API-driven OCR for batch document ingestion and downstream indexing with occasional review of low-confidence results.
Sejda OCR
SMBOnline PDF editor with a dedicated OCR feature for scanned documents.
Job queue style web processing that supports iterative reruns and review of extracted text per document batch.
Sejda OCR converts scanned documents into editable text using a browser-based workflow that emphasizes repeatable input-to-output processing. It supports common file inputs like PDF and image formats and can return OCR results in multiple text and document output forms.
- +Browser workflow fits ad-hoc OCR jobs without project setup
- +Provides multiple output formats for OCR text reuse
- +Handles multi-page documents for batch-style turnaround
- +Quick reruns support iterative cleanup of low-quality scans
- –Limited team controls for governance and tenant-level administration
- –No first-party OCR API surface for automated document ingestion pipelines
- –Handwriting recognition is not a focus area for long-form scripts
- –Large-volume throughput needs external job orchestration
Operations analysts
Convert PDFs to editable text
Editable text for reuse
Content managers
Export OCR for document searching
Better searchable archives
Show 2 more scenarios
Manual document reviewers
Process a repeatable OCR job set
Repeatable OCR outputs
Handle quick turnaround OCR tasks using a job-based web flow for consistent input to output processing.
Team coordinators
OCR mixed document inputs
Standardized OCR deliverables
Convert scanned documents from common formats into multiple OCR output forms for downstream work.
Best for: Fits when teams need fast, browser-driven OCR on batches and value human review over custom integrations.
ILovePDF OCR
SMBPopular online PDF toolset featuring an OCR conversion module.
Searchable PDF text output that preserves the page structure for direct review without re-OCR.
ILovePDF OCR converts image and scanned PDFs into searchable text and editable output, which fits document workflows that start in PDF. It provides page-level recognition with language selection for better results on multilingual material.
- +Browser workflow keeps OCR in the same file-centric toolchain as PDF editing
- +Language selection improves recognition on mixed-language documents
- +Exports searchable PDF text so downstream search works without re-OCR
- +Handles multi-page documents with consistent per-page processing
- –Limited control over confidence thresholds and low-confidence routing for review
- –Batch throughput and concurrency ceilings constrain high-volume OCR jobs
Document coordinators
OCR scanned PDFs for searchable text
Searchable PDF outputs
Review and reflow teams
Export editable OCR text
Faster document edits
Show 2 more scenarios
Multilingual processing teams
OCR multilingual page content
More accurate recognition
Select a language to improve OCR on multilingual material that appears across scanned PDFs.
Small business teams
Run browser batch OCR on documents
Efficient batch conversions
Use page-level OCR in a browser batch workflow when code-driven integration is not available.
Best for: Fits when teams need quick browser-based OCR on scanned PDFs and must return searchable text fast.
Conclusion
After evaluating 10 online ocr software, OCR.space stands out as our overall top pick — it scored highest across our combined criteria of features, ease of use, and value, which is why it sits at #1 in the rankings above.
Use the comparison table and detailed reviews above to validate the fit against your own requirements before committing to a tool.
How to Choose the Right Online Ocr Software
Online OCR software turns scanned PDFs and image files into machine-readable text through a browser workflow or an OCR API, and this guide covers OnlineOCR, Convertio, OCR.space, i2OCR, NewOCR, PDF24 Tools, Sejda OCR, ILovePDF OCR, Smallpdf OCR, and Tesseract.js. The shortlist emphasizes how each option handles multi-page uploads, outputs searchable text layers, and supports exception handling for low-confidence characters.
The buying criteria focus on integration depth and automation surface, including whether the tool is built for pipeline orchestration with asynchronous batches like NewOCR and OCR.space, or for file-centric upload-to-text workflows like OnlineOCR and PDF24 Tools.
Online OCR Software for turning scans and images into searchable text
Online OCR software accepts scanned PDFs and image files and produces OCR output such as plain text or searchable PDFs with embedded text layers. Tools like Convertio and PDF24 Tools keep OCR and output download in the same web workflow, which fits teams that need text extraction for indexing or review without building an OCR ingestion service.
OCR.space and NewOCR support automation-friendly batch flows, with OCR.space returning bounding coordinates and character-level confidence values and NewOCR providing API job handling for asynchronous multi-page runs. Tesseract.js differs by running in the caller environment, producing HOCR with word bounding boxes so teams can implement their own layout logic and table or form handling around coordinate data.
Online OCR evaluation features that affect throughput and control
Online OCR choices differ most in how reliably the output supports review and downstream automation. The gap shows up in multi-page handling, confidence and bounding data, and whether the workflow stays file-centric or becomes API-driven.
This guide prioritizes exception handling and workflow shape because teams usually fail OCR projects at the handoff points, not at the text recognition step. Tools that return per-character signals and structured outputs reduce manual cleanup time and lower rerun costs on low-confidence regions.
Multi-page OCR output handling for document batches
OnlineOCR simplifies spot-checking by providing per-page OCR output for multi-page documents. Sejda OCR and Smallpdf OCR keep multi-page work in a browser file workflow that returns searchable PDFs for direct review.
Confidence scoring plus bounding coordinates for targeted review
OCR.space pairs confidence scoring with bounding boxes so teams can flag specific uncertain characters for review. i2OCR adds per-character confidence indicators to speed manual correction routing when batches contain mixed-quality scans.
Searchable PDF text-layer embedding for immediate indexing
Convertio and OCR.space return searchable text layers from uploaded scanned files and extracted outputs. PDF24 Tools embeds recognized text into a searchable PDF within the same web workflow.
Automation surface for batch ingestion pipelines
NewOCR provides API job handling for asynchronous OCR batches with retrieval of extracted results for pipeline automation. OCR.space also supports automation-friendly batch flows, while OnlineOCR stays closer to upload-to-text without an OCR API pipeline.
HOCR and coordinate data for custom layout logic
Tesseract.js returns HOCR with word bounding boxes so teams can implement custom text-to-region mapping for indexing and review tooling. OnlineOCR and ILovePDF OCR focus more on file-based text extraction and searchable PDF outputs than on coordinate-first integration.
Layout and form handling quality on dense tables
OCR.space supports searchable PDFs with embedded text layers, and its bounding data helps exception handling on uncertain characters. i2OCR can struggle with layout handling for complex forms with dense tables, which can increase rework when field extraction depends on grid accuracy.
How to choose online OCR based on workflow shape and exception control
The first choice is workflow shape. Some tools stay file-centric and return downloadable outputs right after an upload, while others expose an automation surface designed for queued OCR jobs in document ingestion pipelines.
The second choice is exception control. Tools that expose bounding data and confidence signals reduce manual review cost by routing only low-confidence regions, while tools that focus on plain text or a basic searchable PDF layer can require more cleanup time downstream.
Pick a workflow shape that matches how work enters the system
If the process starts with occasional browser uploads and ends with searchable PDFs for indexing, Convertio, PDF24 Tools, and Smallpdf OCR keep OCR and download in the same file flow. If the process starts in a queue and needs automated job execution with later result retrieval, NewOCR provides API job handling for asynchronous OCR batches.
Decide whether review needs per-character bounding signals
If the workflow includes confidence-based exception handling that routes only uncertain characters, OCR.space and i2OCR provide bounding coordinates and per-character confidence signals for targeted review. If review mainly checks extracted text in a single output per document, OnlineOCR’s per-page output can reduce reprocessing scope without requiring coordinate-driven review.
Match output needs to indexing and downstream parsing requirements
If the priority is immediate search over scanned PDFs, prioritize searchable PDF text-layer embedding like Convertio, OCR.space, PDF24 Tools, and ILovePDF OCR. If the priority is custom post-processing, Tesseract.js provides HOCR with word bounding boxes so custom mapping and alignment logic can live in the calling application.
Validate layout behavior on your hardest document type
For dense tables and complex forms, test i2OCR because layout handling is weaker on dense table structures and can degrade field-level usefulness. For low-contrast or noisy scans, validate OCR.space and OnlineOCR on representative samples because OCR accuracy can vary based on scan quality and preprocessing.
Use rerun strategy to manage iterative OCR on noisy batches
If iterative reruns and human review per document batch are core to the process, Sejda OCR uses a job queue style workflow that supports repeated reruns and review cycles. If the process needs controlled pagination during review, OnlineOCR’s per-page output helps isolate failed pages instead of rerunning entire documents.
Who should use which online OCR approach
Teams that extract text for search, indexing, and human review benefit from tools that generate searchable PDFs quickly and keep the workflow file-centric. Teams that automate ingestion and exception handling need an automation surface and output signals for low-confidence routing.
The right fit depends on whether the organization owns the downstream parsing logic. When layout reconstruction and coordinate-based alignment matter, coordinate-first tools fit better than file-centric download tools.
Operations teams doing manual review of scanned batches
OnlineOCR supports per-page OCR output so reviewers can reprocess only failed pages instead of handling full documents. Sejda OCR supports job queue style reruns so teams can iterate on extracted text inside a browser workflow.
Engineering teams building ingestion pipelines for asynchronous OCR
NewOCR provides API job handling for asynchronous OCR batches with result retrieval, which matches queue-based ingestion patterns. OCR.space also supports automation-friendly batch flows with confidence scoring and bounding data for exception routing.
Indexing workflows that rely on searchable PDFs
Convertio and OCR.space return searchable text-layer outputs from uploaded scans that support immediate search without building an OCR API pipeline. PDF24 Tools and ILovePDF OCR embed recognized text into searchable PDFs within the same web file flow.
Apps that need custom OCR-to-layout mapping for indexing and QA tooling
Tesseract.js returns HOCR with word bounding boxes so engineering can implement custom text-to-region mapping in the calling environment. This approach fits when table or form handling must be driven by caller logic rather than a vendor layout model.
Teams extracting text from low-quality scans with confidence-based triage
OCR.space returns confidence scoring plus bounding coordinates to flag specific uncertain characters for review. i2OCR also returns per-character confidence indicators to speed routing when batches contain mixed scan quality.
How We Selected and Ranked These Tools
We evaluated OnlineOCR, Convertio, OCR.space, i2OCR, NewOCR, PDF24 Tools, Sejda OCR, ILovePDF OCR, Smallpdf OCR, and Tesseract.js across features, ease/value, and how each option supports real OCR workflows. Features accounted for 40% of the score because multi-page handling, confidence and bounding signals, searchable text-layer outputs, and automation surfaces directly affect reprocessing effort.
Ease/value each accounted for 30% because upload-to-text workflows, job flow clarity, and the amount of cleanup needed after extraction determine day-to-day throughput. OnlineOCR separated itself by combining per-page OCR output for multi-page documents with a straightforward web upload-to-text workflow.
Frequently Asked Questions About Online Ocr Software
How do Google Cloud Vision, Azure AI, and Textract compare with OCR.space for confidence-based exception handling?
Which workflow is best for teams that need asynchronous OCR API jobs rather than synchronous text extraction?
How should teams handle multi-page TIFF and other scanned document batches across different tools?
When does a searchable PDF output matter more than editable plain text?
What breaks if an organization replaces a cloud OCR API like Textract with Tesseract.js in production?
How do HOCR and bounding boxes change downstream indexing compared with plain text outputs?
Which tool fits document ingestion pipelines that need structured, machine-readable extraction instead of user review?
How do teams migrate from a browser OCR workflow to an automation pipeline without reworking their data model?
What security controls differ when OCR runs in a managed online service versus within Tesseract.js in an application?
Tools reviewed
Primary sources checked during evaluation.
Referenced in the comparison table and product reviews above.
Funding and competing interests
This study was co-funded by several of the vendors assessed in it. No single vendor funded it. Funders could propose the market studied.
They had no role in selecting vendors, setting or weighting criteria, analysing data or drawing conclusions, and they received the same factual-accuracy check as every other vendor assessed.
Market definition, inclusion rules and criteria weights were fixed by an editor before research began and did not change afterwards. Editorial process →
Keep exploring
Comparing two specific tools?
Software Alternatives
See head-to-head software comparisons with feature breakdowns, pricing, and our recommendation for each use case.
Explore software alternatives→In this category
Data Science Analytics alternatives
See side-by-side comparisons of data science analytics tools and pick the right one for your stack.
Compare data science analytics tools→