Finders and zones

A finder is a named rule that finds one value on a document of a class: a date next to a label, a number in a zone, a barcode, a table, or a value built from other finders. Fields then take their values from finders. If you know Kofax/Tungsten Capture, finders are the same idea as locators. See Kofax locators and DocView finders.

The zone designer is a simpler tool for a single field: you draw a zone on a sample page and save it to the field.

Both work on sample documents that you upload, so you can see the result straight away. Samples are kept with the Batch Setup. They come from the Samples & testing panel on the Finders tab, from the zone designer, or from an import, and they open automatically in the zone designer, the Finders tab and the field editor.

Why use finders

A field can have its own extraction method (see Fields and extraction). Finders go further:

• Search near another finder. Read the amount to the right of where the "Total" finder found its label.
• Search only part of the document. Look only on some pages, or only inside an area of the page.
• Fall back. If the label-and-value finder finds nothing, use a fixed zone instead.
• Combine. Build an address from street and city finders.
• Match master data. Find the supplier by fuzzy matching the page against your supplier list.
• Read a whole form. One Advanced zones finder reads every zone, tick box and signature of a form.
• Script. Read several values at once with JavaScript.
• Test on many samples. See the hit rate of every finder across a set of documents.

Open the Finders tab

1. Open the Batch Setup and click the Finders tab.
2. Choose the Document type at the top left. Finders belong to one document type.
3. Add samples if you have none: drag one or more PDF, TIFF, PNG or JPEG files onto the Samples & testing panel under the page, or click Add files there. Each file shows its upload progress; the cross cancels it.

The Finders page

Drag the bar between the sample page and the finder to make either wider (or click it and use the arrow keys). On narrow screens the parts sit one below the other.

The Finder tab

Under the finder type, How to use explains when to use this type of finder and how to set it up, step by step, with a worked example. Fill in this example puts the example's settings in the finder for you to adjust; until you save, the undo button takes them out again. Close How to use once you know the type: DocView remembers that in your browser. When no finder is selected, the right side shows Getting started with finders.

The Name, the Finder type and a Description, then the type's main settings. The type's other settings are in sections that open and close; the fine-tuning ones (Evaluation Settings, Advanced, Search Masks, Registration) start closed:

Every finder also has Values and fields (every value the finder gives and the fields that take it, with Create & assign fields for those no field takes yet) and its Time limit (see Status, time limits and switching off).

Related settings sit together in a box, side by side: for example Zone on the page holds the page, left, top, width and height of a zone, in % of the page (drag on the sample page to set them), and Scoring weights holds the weights of a Label & value finder. Click the ⓘ beside a setting to read what it does.

When a setting is wrong (for example two finders with the same name), the message under the tabs says what to fix and Save waits until it is fixed.

Note: Uploaded samples are cleaned up and read with OCR using this version's settings. They are also used for classifier training.

Add a finder

1. Click Add finder under the list and choose the finder type.
2. Give it a Name on the Finder tab: letters, digits and underscores, starting with a letter, for example InvoiceDate. Other finders and fields refer to it by this name.
3. Add a Description if it helps.
4. Fill in its settings.
5. Click Run test in the Samples & testing panel under the page (or press Ctrl+Enter) to see what it finds.
6. Click Save (or press Ctrl+S).

Duplicate makes a copy to start from; Delete removes a finder (fields that use it must take another finder first).

The sample page helps you fill in the settings:

• For Label and value and Zone relative to a label, click a label on the page to use it. Shift+drag picks several words.
• For Fixed zone and Handwriting, drag on the page to draw the zone; for Ask the vision model, drag to ask about one part of the page only.
• For Pattern (regular expression), click a word to start a pattern from it.
• For Tick boxes (OMR), drag around each tick box and type the value it stands for.
• For Advanced zones (form), select a zone and drag on the page to place it. See Advanced zones (form).
• For every search method, the Search areas tab limits where the finder looks. See Search areas.

The quickest way: New finder from selection

1. Click the value on the sample page (Shift+drag for several words).
2. Click New finder from selection under the list and choose what to make:
   • Best guess: a Label & value (smart) finder when there is a label on the same line or just above (with that label, where it is, and the value's shape: a date, an amount, a reference…), otherwise a Fixed zone;
   • Label & value: always the label finder (DocView says so when there is no label nearby);
   • Fixed zone from this rectangle;
   • Pattern from this text: a pattern written from the selected text (digits become \d, capitals [A-Z], other characters stay), for example XMA-3301 gives ([A-Z]{3}-\d{4}); adjust it under Format Definitions;
   • Barcode in this area: a Barcode finder with a region around the selection;
   • Tick box here: a Tick boxes finder with one box (Y when ticked, N when not).

Test it on your samples and adjust.

Switch on Clicks select values under the list to pick several values in a row without changing the selected finder.

Finder methods

The settings of Label and value, Pattern, Fixed zone, Zone relative to a label, Barcode, Table / line items, Cloud key-value and Tick boxes (OMR) are the same as the field methods in Fields and extraction.

Pattern

A Pattern finder searches the text of the document (or of its regions) with format definitions.

Under Format Definitions:

Patterns are tried in order (use the arrows to move one). Insert from library… adds a ready-made pattern: dates (31/12/2026, 2026-12-31, 31 December 2026, December 31, 2026), amounts (1,234.56, 1.234,56, negative amounts), IBAN, UK sort code and account number, UK and EU VAT numbers, UK National Insurance number, UK postcode, e-mail address, UK phone number, and invoice, credit note, purchase order and reference numbers. Type sample text under Test expression to see what each pattern matches.

Under Evaluation Settings:

Older Pattern finders with a single Pattern and More patterns keep working; those settings are under Format Definitions too.

Fixed zone

A Fixed zone reads the words the full-page OCR found inside a rectangle. Draw it on the page, or type Left, Top, Width and Height (fractions of the page).

Use Advanced zones (form) when a zone needs its own recognition or clean-up (hand printing, tick boxes).

Ask the vision model

The vision model (DocView OmniRead) reads the page image and answers questions, for values that are hard to find with rules, or for any document type at all. It must be allowed under Settings → DocView OmniRead.

• Question asks for one value, for example What is the invoice number? Answer with the number only.
• Questions (under Questions) asks several at once, one per line: Name | question | optional pattern, for example InvoiceDate | What is the invoice date? Answer as DD/MM/YYYY. | (\d{1,2}/\d{1,2}/\d{4}). All are asked in one request, which is far quicker than one finder per question, and each answer is a value of the finder (Ask.InvoiceDate) that a field can take. Create & assign fields makes a field for each question.

The model answers with text only, so answers get a fixed confidence of 85%; a pattern keeps just the part you want. On a computer without a graphics card each request takes one to three minutes, so vision finders always get a time limit of at least 300 seconds and run one at a time.

Label & value (smart)

The smart label finder works the way a person reads a document: find a label, look where its value should be, and check that the value looks right. Settings:

Every value the finder considers gets a score from 0 to 100%: how well the label matched and how specific it is, whether the value has the right shape, how near it is, how well it lines up with the label, and how well OCR read it. The best (or first or last) is the finder's value, with its score as the field's confidence, so an unsure value is flagged for checking like any low-confidence value.

When you test the finder, Candidates on this sample lists what it considered, best first, with each score and why (for example label "Invoice date" read as "lnvoice dale", value right of it, 27% of the distance, shape matches, OCR 97%). Click a candidate to see it and its label on the page. Pattern finders list their matches the same way.

Combine it with Search areas (for example the top half of the first page) and with another finder to search near.

Search areas

The quickest way: on the Finder tab, the Search area strip shows where the finder looks. Drag a rectangle on the sample page — for any finder that does not read a fixed zone, that rectangle becomes its search area on that page, outlined on the page as Search area. Draw on the page in the strip does the same, Whole document removes the areas again, and More… opens the Search areas tab for pages, ready-made parts of the page and several areas. Drag again (also starting inside the outline) to draw it anew. Shift+drag still picks words, and a click on a label still uses it.

The Search areas tab of a finder holds its search areas. A finder that searches the document can be told where to look. This works like the search region of a Kofax locator. A finder can have several areas, and it searches all of them together, for example the top of the first page and the bottom of the last page.

Open the Search areas tab of the finder. Each area is a card:

1. Click Add area.

2. Choose its Pages:

   Pages	Meaning
   All pages	Every page
   First page	The first page
   Middle page	The page in the middle of the document
   Last page	The last page
   First pages…	The first so many pages; type how many
   Last pages…	The last so many pages; type how many
   Specific pages…	Pages you type, such as 1, 2-3 (pages 2 and 3), 1,3 (pages 1 and 3), 3-last (page 3 to the end) or -2 (the second last page)

   Page where another finder found… searches only the page where the chosen finder found its value (for example the totals on the page where the invoice number is), and makes that finder run first.

3. Choose the Part of the page — Whole page, Header (top 15%), Top third, Top half, Top right quarter, Top left quarter, Middle third, Bottom half, Bottom third, Footer (bottom 15%), Left half or Right half — or click Draw on the page and drag a rectangle on the sample page. The button then reads Redraw on the page. Whole page clears the rectangle again, so the area covers its whole pages.

4. Add more areas the same way. First + last page adds two areas at once, the first page and the last page, each with its own part of the page (for example the header of the first page and the footer of the last). The X on a card removes that area.

The finder then only looks at those pages, and only at words and barcodes whose centre is inside one of the rectangles. If any area is All pages without a rectangle, or the finder has no areas at all, it searches the whole document. The areas are shown on the sample page, saved with the finder and exported with the Batch Setup. The field editor's sample preview and the zone designer show them too.

Let DocView suggest areas

Click Suggest areas from the page. DocView outlines the blocks of text it found on the current sample page with dashed boxes. Click a block to add it as an area of that page. Click the button again to stop.

Which finders use search areas

Search areas are used by finders that search the document: Pattern, Label & value (smart), Label and value (simple), Zone relative to a label, Barcode, Table / line items, Cloud key-value, Database lookup and Script. Methods that read fixed places have their own positions, so search areas don't apply to them: Fixed zone, Tick boxes (OMR), Advanced zones (form) and Document type (classification).

Tip: Search areas make a finder faster and stop it picking up the same label elsewhere, for example a "Total" in the line items when you want the one at the bottom of the last page.

Barcode

Every barcode that passes the checks is listed as a candidate with its type and page.

Table / line items

A Table finder fills a table field: one row per line item, each cell with its value and its place on the page. Its settings are in the sections Settings · Columns · Cells · Advanced of the Finder tab; it also has Search areas and Test.

On the Test tab, the rows found are shown as a grid. Click a row, or Show the cells on the page, to see every cell marked on the sample.

Tick boxes (OMR)

1. Choose Tick boxes (OMR).
2. Under Boxes, click Auto-detect boxes in an area and drag around the group on the sample page: DocView finds the printed boxes and takes the words beside each as its value. Or drag around each box on the page.
3. Check the value of each box, for example Mr, Mrs and Ms. Remove box takes one off.
4. On General, choose the Answers: One answer (the ticked box's value; Value when several are ticked otherwise) or Several answers (the ticked boxes' values joined with Join several answers with), and the Value when none is ticked. A group of one box is a yes/no box.
5. On Settings: Ticked from and Empty below (between the two a box is uncertain and checked by a person), Leave out the border, and how far to Look for each box.
6. Click Test.

See Forms, tick boxes and signatures for how the boxes are read.

Advanced zones (form)

An Advanced zones finder reads a whole form in one go. It is like the Advanced Zone Locator of Kofax Transformation. The finder holds many zones, and each zone is a value that fields use as <finder>.<zone>, for example AZL.Surname. Each zone is read on its own from the scan (it does not reuse the page's OCR), after the page is lined up.

Its Finder tab has the sections Registration · Zones · Anchors · Values and fields (each zone and the field it fills); its results are on the Test tab.

General

• Line pages up by: Anchor words and shapes together (recommended), Anchor words only, Anchor shapes (printed boxes) only, or Do not line pages up.
• Read text zones with: the Batch Setup's OCR engine, or DocView OmniRead handwriting, which is far more accurate on hand-printed forms.

Add the zones

1. Add a finder, for example AZL, and choose Advanced zones (form).
2. On the Zones tab, click Zone wizard (or Add zone). The wizard takes you through the steps: Draw the zone on the page, its Type, its Recognition (characters, boxes, pattern), its Clean-up, and a Test on the sample.
3. Repeat for each zone, then click Apply or OK.

The list shows each zone's type and page, with a check when it is placed. Click a zone in the list or on the page to select it. Use Filter zones to find one in a long list, and Remove zone to delete one.

Types of zone

How the recognition options help:

• Characters: a digits-only zone turns letters OCR reads for digits into digits (O → 0, I → 1, S → 5…).
• Must match (pattern): when the value does not match, the OCR look-alikes are tried; if it still does not match, the value is kept but marked for checking.
• Ticked from and Empty below: a box filled more than "empty" but less than "ticked" is uncertain and marked for checking.

For hand printing in comb boxes, see Hand printing in comb boxes.

Clean-up profiles

Under Clean-up profiles under Zones, add named settings that several zones share: Drop-out (read only the pen, not the printed boxes), Margin (mm) added around the zone, and One or Several lines. Choose the profile on a text zone; the zone's own settings win.

Line the page up (anchors)

Scans are often shifted by a few millimetres, slightly turned, scaled, or stretched more down the page than across (scanners feed paper unevenly). Like Kofax's Advanced Zone Locator, DocView lines each page up with anchors before reading its zones:

• Anchor words: captions printed on every copy of the form, such as Surname or Date of birth.
• Anchor shapes: printed boxes on every copy (for example tick boxes).

The quickest way: open a sample and click Find anchors on every page under Anchors. DocView picks, on every page, words printed only once and well read, and boxes with no look-alike nearby, spread over the page — never inside the zones that are read, so nothing written by hand is used. The table under Anchors shows per page how many zones, words and shapes it has, and after a test how the page lined up; pages with zones but no anchors are marked. Click a page in the table to see it; the anchors are outlined on the page.

By hand: select a printed caption on the page and click Add the selected words, or click Draw a shape and drag around a printed box. The × on a page's row removes that page's anchors.

How it works on each scan:

1. The anchor words are found in the page's OCR, then the shapes close to where the words put them (so a row of identical boxes cannot be mistaken for one another).
2. From the anchors found, DocView works out how the page moved: a shift with one anchor; also a turn and a scale with two; and with three or more spread over the page, also a different stretch across and down.
3. Anchors that disagree with the others (a caption printed twice, a misread, a wrong box) are left out.
4. Every zone moves with the page, and then also by the anchors nearest to it (Also move each zone by the anchors nearest to it, on by default, up to about 3 mm) — paper that stretched in one part of the page still lines up there.

Implausible results (turned more than 5°, scaled more than 4 %, or stretched unevenly by more than 2.5 %) are not used.

Under Registration: At least this many anchors must be found (when anchor words were expected but fewer are found, the zones are read where they were drawn and every value of that page is marked for checking), Look for each anchor within (a share of the page), Look for anchor shapes within (mm) (0 switches shapes off) and the local correction. How it ran on the Test tab says for every page how many anchors were found and how far it moved, turned and stretched.

The zone on the page

Select a zone to see where it is: a small map of the page with the zone, its Page (‹ › to move it to another page) and its Left, Top, Width and Height in % of the page, with about how many millimetres that is on A4. Type a value, use ↑ ↓ in the box (0.1 %), or click − and + (Shift: 1 %) to nudge it; or drag on the sample page to draw it again. Fixed zone, Handwriting and Vision model finders show their zone the same way.

Use the zones in fields

Under Values and fields, every zone is listed with the field it fills; Create & assign fields makes fields for the zones that have none. Or open each field and choose the zone under Value from finder, for example AZL.Surname.

Fields filled by advanced zones ignore their confidence threshold: a value is always written, and only zones that failed their pattern, uncertain tick boxes or a page that did not line up are marked for checking. A finder reading many zones may take a while, so advanced zones always get a time limit of at least 300 seconds.

Cloud key-value (Azure)

An optional plug-in for the few documents rules cannot read, and an alternative to Kofax's text content and key-value locators. It sends the pages to Azure AI Document Intelligence and takes the keys or fields Azure finds.

Important: the pages this finder reads leave your premises. The finder is off until an administrator sets Cloud key-value finder (Azure) to true under Settings → System settings and stores the secrets ocr.azure.endpoint and ocr.azure.key (Settings → Secrets). Everything else in DocView runs without it.

Document type (classification)

This finder gives the document's type as a value: its name, or its key (Value). It also gives Confidence (how sure the classification was, in percent) and Alternatives (the other types it scored, best first, for example Credit note 40%), so fields and validation rules can use them; Create & assign fields makes all three. It is like the classification locator of Kofax Transformation.

Database lookup (fuzzy)

This finder matches the document against a Master data list lookup, such as all your suppliers, and tolerates OCR mistakes. Every column of the matching row is a value of the finder (Supplier.VatNumber…). Its settings are in the sections Fields · Search Masks · Advanced of the Finder tab; it also has Search areas and Test.

A row's score is its best column match (weighted) plus a little for every other column that also matches, so a row matching the name, the postcode and the phone beats one that shares only the address. Letters with accents match their plain forms (é and e).

Script

A Script finder runs JavaScript in a sandbox (no files, network or .NET; time and memory limits). Its tabs are General · Field Definition (the Values it returns, each a value of the finder that a field can take as <finder>.<value>) · Script, as sections of the Finder tab; it also has Search areas and Test.

The script can use:

Return a value, or an object whose properties each become a value. A value can also carry its confidence, its place and candidates:

const all = findAll('INV-[0-9]+');
return {
  InvoiceNo: { value: all[0] || null, confidence: all.length === 1 ? 0.95 : 0.6, candidates: all.map(v => ({ value: v })) },
  Total: parseAmount(label(['Total due', 'Total'])),
  InvoiceDate: parseDate(label(['Invoice date', 'Date']))
};

Under Script, Check script says whether the JavaScript is correct (and on which line it is not), and Run on sample runs it and shows what it returned. The script editor suggests the data and helpers as you type; scripts are kept with the Batch Setup version.

Combine other finders

Under Sources, list the finders to combine, in order, with a Weight (0–1, empty = 1) for how much each counts. On Evaluation Settings, choose the Mode:

Every source value is listed as a candidate with where it came from.

Use other finders

• Other finders tab, Search near what another finder found: choose To the right of, Below, To the left of, Above or Around, the other finder, and a distance within so many pixels, mm or % of the page (default 800 pixels). The whole document switches this off.
  • Off its line or column by up to (mm): how far above or below the other finder's line (or left or right of its column) a value may be; empty is half a line.
  • Nearer values score higher (0–1): each candidate's score × (1 − this × its distance as a share of the search distance), so the nearest good value wins.
  • Same page only: untick it to look in the same place on every page.
• Other finders tab, Fallbacks: if nothing is found, use the result of (in this order): click Add a finder… to add a fallback. Fallbacks are tried from the top of the list. Use the arrows to move one up or down, and Remove to take it off. A finder that fails with an error also falls back. Also when the score is below tries the fallbacks for a value that was found but scored lower than that (the value is kept if no fallback finds anything).

Execution order

DocView works out the order itself: a finder runs after every finder it searches near, falls back to, searches the found page of, combines or reads in a script. Finders that do not depend on each other run side by side (up to four at a time; database lookups, scripts, the vision model and cloud finders one at a time), which makes extraction quicker. Execution order (under the list) shows the steps and what each finder needs. DocView stops you if a finder uses itself, uses a finder that does not exist, or finders depend on each other in a circle, and names the circle.

Status, time limits and switching off

Every finder ends with a status: Found, Not found or Error.

• Error means it could not run: a pattern that is not valid, a lookup that failed, the vision model not answering, or it took longer than its Time limit (Finder tab, 30 seconds unless you change it; vision model, cloud and advanced zones finders always get at least 300 seconds). The other finders carry on, and its fallbacks are still tried.
• Switched off finders (the On/Off switch, or right-click → Switch off) do not run and find nothing, but keep their settings. Use it to try a document without one finder.

The Test tab shows the status, the time taken, the error if there was one and How it ran: which pages and regions it searched, what it searched near, which fallbacks it tried, and the value with its confidence.

Create & assign fields

Create & assign fields (under the list, and under Values and fields on a finder's Finder tab) makes a field for every finder value that no field takes yet: the finder's own value, or each value of a finder that gives several (database lookup columns, script values, zones of advanced zones, vision model questions). Each field is named after the value (InvoiceDate → label Invoice date), with a data type guessed from the finder (a date or amount label & value finder, a date or amount pattern, or a name ending in Date, Total, Amount…). Table finders are left out, because a table field needs its columns. Check the new fields on the Batch Setup's Fields tab.

Import and export finders

Project → Export saves the finders of the document type as a JSON file; Project → Import adds the finders of such a file (names already in use get a number). Use it to copy finders between document types or Batch Setups. Import does not save: test the finders, then click Save.

Test finders

Tests run from the Samples & testing panel under the page. Its Samples & results tab lists every sample; click one to show it on the page. A cross removes a sample.

1. Tick the samples to test, or use All, This type or None. With none ticked, the test runs on the sample shown.

2. Next to Test, choose:

   • This finder, the selected finder with every value it returns;
   • This zone, only the selected zone of an Advanced zones finder, for example AZL.Surname;
   • All finders of the document type; or
   • All fields, to run the fields that take their values from finders as well.

   The finders that the tested finder depends on, such as the finder it searches near or its fallbacks, are always run too.

3. Click Run test (or press Ctrl+Enter). The button says what it runs on, for example Run test on 5 selected.

The panel then shows, for every tested sample, a Result such as 3 / 4 (green: everything found, amber: some, red: nothing) and one column per value with what was read, or not found. Hover a value for its confidence and page; click it to open that sample with the value highlighted on the page. The bottom row, Found on, gives each value's hit rate over the tested samples, and the bar above the table the overall result. Clear removes the results and the marks. Use the arrows above the page, or Alt+Left and Alt+Right, to move between the samples.

The finder's Test tab (in the settings on the right) shows its result on the sample shown, how it ran, the rows of a table finder and its Candidates: every value it considered, best first, with its score and why (the first, in green, is the finder's value). Click one to see it on the page. Run on sample on the Finder tab tests only the selected finder on the sample shown.

Benchmark

The benchmark tells you how well the finders read your sample set, field by field, and whether a change made it better or worse.

1. Show a sample and test it with All fields (Ctrl+Enter).
2. Open the Benchmark tab of the Samples & testing panel. Under Expected values on the sample, click Use the values read, correct the values that are wrong (or clear a field the document does not have), and click Save. To change the sample, click it on the Samples & results tab or use the arrows above the page.
3. Repeat for a few samples of each layout you get.
4. Click Run benchmark. Every sample with expected values is read with the finders as they are now (saved or not).

For each field you see how many values were Right, Wrong and Missed, the Precision (of the values read, how many were right) and the Recall (of the expected values, how many were read), with the change since the previous run. Values not as expected lists each miss; click one to open that sample. Values match when they read alike, ignoring case, spaces and punctuation.

All samples and finders at once

Tick All samples, choose All finders and click Run test: the Samples & results tab then shows every document and finder at once, with each finder's hit rate in the Found on row.

You get a grid of documents and finders with every value found, each miss, fallbacks used, and a hit rate per finder. Click a row to open that document with its results marked.

Tip: Aim for a high hit rate on 10 or more real samples before you release. A finder that works on one sample may still fail on the next supplier's layout.

Connect a field to a finder

Open the field and choose the finder in Value from finder, or let Create & assign fields do it. On the Finders tab, a finder's Values and fields shows which fields take its values, and its Fields list shows each field with its finder. To check what every field would get, press Ctrl+Enter and look at the Test tab.

Close candidates and picking another

On the field, Minimum lead over the next candidate (0–1) is an ambiguity check: when the finder's two best different values score closer than this (for example 0.1 = ten points), the value is kept but marked uncertain, so a person checks it. In Index and Validate, a field shows the other values its finder considered under Other reads; point at one to see where it was read, click it to use it.

The zone designer

For a single zonal or anchor field you can use the zone designer instead of finders.

1. Open the field and save it once.
2. Click Open the zone designer.
3. Choose a Sample document, or upload one and click Upload.
4. Choose the Method: Fixed zone (same place on every document), Anchor-relative (relative to a label such as "Total") or Tick boxes (OMR).
5. For Anchor-relative, click Pick anchor and click the label on the page, or type the Anchor text.
6. Click Draw zone and draw a rectangle around the value.
7. Set the options you need: Use the last occurrence, Always read from the last page, and Re-OCR the zone (small print, handwriting-like fonts).
8. Click Test to read the zone. The result shows the value, its confidence and the page.
9. Click Save to field, then Back to field.

For Tick boxes (OMR), drag around each box instead of drawing one zone, and type the value of each box. Set the value when none is ticked and when several are ticked, then click Test and Save to field.

Use Pan, the zoom buttons, Fit width and Fit page to move around the page.

Coming from Kofax Transformation

For each Kofax locator and the DocView finder that replaces it, see Kofax locators and DocView finders.

Released versions

On a released version you can still test finders and zones, but you need a draft to save changes. See Versions and publishing.