How to Use Family Tree Analyzer

What is a GEDCOM file?

GEDCOM is the standard file format used by genealogy software to store family tree data. It contains names, dates, places, relationships and sources — everything that makes up your tree. The file typically has a .ged extension.

How to export a GEDCOM file

Most genealogy websites and software can export your tree as a GEDCOM file. The exact steps vary, but here are pointers for the most popular providers:

• Ancestry — Go to your tree, click the tree name, and choose "Export tree"
• MyHeritage — Open your tree → tree name menu (top left) → "Export GEDCOM"
• findmypast — My tree → Tree settings → "Export tree" (download as GEDCOM)
• Geni — Tree → Settings → "Export" → download GEDCOM (covers your tree, not the whole shared World Family Tree)
• WikiTree — "My Tree" → Tree tools → "Export a GEDCOM"
• Family Tree Maker — File → Export → GEDCOM
• RootsMagic — File → Export → Export GEDCOM file
• Legacy Family Tree — File → Export → GEDCOM
• Family Historian — File → Export/Backup → Export GEDCOM File
• MacFamilyTree / Reunion — File → Export → GEDCOM
• Gramps — Family Trees → Export → GEDCOM

If your provider is not listed, look for an "Export", "Download" or "Share" option in the tree settings menu — nearly all genealogy websites and software support GEDCOM export. FamilySearch is the main exception: its Family Tree is a single tree shared and edited by everyone, so there is no "export my tree" button — see below for how to work with it.

Don't have a tree started yet?

If you have only paper records, old certificates, or family knowledge passed down rather than an existing digital tree, you can still get started for free — no subscription required:

• FamilySearch Family Tree — completely free and online. It's a single tree shared by everyone, so many of your ancestors may already be recorded — you can attach yourself to existing relatives instead of starting from nothing, and FamilySearch will suggest record matches automatically. Because it's shared rather than personally owned, there's no direct "export" button; free partner tools such as RootsMagic (via its FamilySearch integration) can download your portion of the tree as a GEDCOM file.
• Gramps — free, open-source desktop software for Windows, Mac and Linux. Works entirely offline, has no subscription, and exports GEDCOM natively — a good choice if you would rather enter your paper records privately before uploading anywhere.

Either way, once you have entered even a handful of relatives, export a GEDCOM file using the steps above and upload it here — you don't need a complete tree to get useful results.

Uploading your file

Click Upload GEDCOM in the sidebar menu (or the upload button on the home page). You can either drag your file onto the upload area or click to browse and select it. FTAnalyzer accepts .ged files and .zip archives containing a .ged file.

Once selected, processing begins automatically. You will see a progress bar and a log of what is being read. For most trees this takes just a few seconds. Very large trees (25,000+ individuals) may take a little longer. Files over 50 MB cannot be processed in the web version — please use the free FTAnalyzer desktop app from the Microsoft Store for larger files.

Automatic reload

By default, when you return to the site your last GEDCOM file is automatically reloaded — either from your browser's local cache or, if you have tree storage enabled, from the server. This means you can close the tab and come back later without needing to upload again.

If you prefer to start fresh each visit, you can turn this off from Account Settings → Preferences → Automatically reload your tree.

Recommended first steps

If you are new to Family Tree Analyzer, here is the recommended path to get the most out of it quickly:

1. Upload your GEDCOM file — use the Upload GEDCOM link in the sidebar. Processing takes a few seconds for most trees.
2. View the Dashboard — this gives you an instant overview of your tree's size and overall health score.
3. Open Your Improvement Plan — this page analyses your tree and produces a ranked list of the most impactful things you can do right now to improve your research. Start here rather than trying to fix everything at once.
4. Explore the rest of the app — once you have tackled the highest-priority improvements, use the sidebar to explore the Census reports, Colour BMD, Data Errors, Research Suggestions and other tools in more depth.

The Dashboard is your main overview once a tree is loaded. It brings together the most important information in one place.

Summary statistics

At the top you will see cards showing how many individuals, families, sources, locations, occupations and data errors are in your tree. Click any card to go directly to that section.

Tree Health gauge

The arc gauge shows your overall Tree Health score from 0 to 100, with a letter grade from A to F. This score is calculated from ten weighted categories (see the Tree Health section below). A score above 75 generally indicates a well-researched tree; below 40 suggests significant gaps worth investigating.

Health bars

Below the gauge, coloured bars show how each of the ten health categories is performing. Click any bar to see more detail on the Tree Health page.

Top 10 panels

Three panels show the most common surnames, occupations and countries in your tree. These give a quick snapshot of your family's geographic and social history.

Relationship Types filter

At the top of many pages you will see a Relationship Types filter bar. This lets you choose which people are included in the statistics and reports based on how they relate to the root person in your tree. The available types are:

• Direct Ancestors — parents, grandparents, great-grandparents and so on
• Descendant — children, grandchildren and further descendants of direct ancestors
• Blood Relation — anyone related by blood (siblings, cousins, aunts, uncles, etc.)
• Married to Blood/Direct — spouses of blood relatives or direct ancestors
• Related by Marriage — people connected only through marriage (in-laws)
• Unknown — individuals whose relationship to the root person could not be determined

By default all types are selected. Deselecting types lets you focus your analysis — for example, selecting only Direct Ancestors shows how complete your direct line is without the noise of distant connections. The filter carries through to any page you navigate to from the Dashboard.

Deep dive cards

Further down, navigation cards link to every area of the application — Individuals, Families, Surnames, Facts, Occupations, Locations, Sources, Data Errors and more. Click any card to explore that data in detail.

The Improvement Plan page analyses your tree and produces a personalised, ranked list of the most impactful changes you can make. Rather than confronting you with a list of thousands of missing records, it focuses on the areas where fixing a small number of things will move your score the most.

How priorities are calculated

Each of the ten health categories has a weight (how much it contributes to the overall score) and a current score (how well your tree performs in that category). The improvement plan multiplies the two together to produce an impact estimate — how many overall score points you could gain by getting that category to 100%.

Categories are listed highest impact first. If your Date Errors category scores 20% with a 15% weight, fixing it could be worth up to 12 points — far more than polishing a category that already scores 90%. Categories that are already performing well (scoring above 90%) are suppressed entirely so you are not distracted by minor gains.

What the items show

Each category shows up to 10 specific individuals or records to work on, together with an indication of what to do — for example, "Search for a 1851 census record for John Smith (born ~1830, Scotland)" or "Add a birth year for Mary Jones." Click an individual's name to view all their facts. Use the See all in full report link to open the relevant filtered report showing every affected record, not just the top 10.

Impact badges

• High impact — fixing this category could improve your overall score by 10 or more points
• Medium impact — 5–9 points of potential gain
• Low impact — under 5 points, but still worth addressing once higher priorities are done

Relationship filter

The filter at the top of the page works the same way as on the Dashboard — you can focus the plan on direct ancestors only, all blood relatives, or any combination. The individual items and counts update automatically when you change the filter.

The Tree Health page gives you a detailed breakdown of how complete and well-sourced your family tree is, across ten categories. Each category is weighted to reflect its importance to research quality.

The ten categories

Each category has a default weight (shown below) that determines how much it contributes to the overall score. Signed-in users can customise these weights — see Customising weights below.

1. Birth Completeness (default 10%) — How many of your blood and direct relatives have a known birth date recorded. Even an approximate year is far more useful than none. The Loose Births report can help you tighten existing wide date ranges into more precise estimates.
2. Death Completeness (default 5%) — How many of your deceased blood and direct relatives have a known death date recorded. The Loose Deaths report can help narrow down approximate death dates based on other facts in your tree.
3. Date Precision (default 10%) — How precise your recorded dates are, scored on a scale from "Unknown" through to "Exact Date." Approximate dates are better than none, but exact dates unlock more features.
4. Census Coverage (default 15%) — Whether your ancestors appear in the census records you would expect for the years they were alive. Missing census entries may point to unresearched lines.
5. Source Quality (default 15%) — How many facts in your tree have source citations attached. Well-sourced trees are more reliable and easier for others to verify.
6. Family Structure (default 5%) — Whether family groups are properly formed with parents linked to children, and marriages recorded. Ancestors born before 1800 are excluded, since surviving records naturally thin out the further back you go. Click this bar to see the two measures that make it up (parents found and marriage found), each with its own count and a link to the matching list. If someone genuinely never married, add a Custom Fact called "Unmarried" or "Died Single" to them and they'll be left out of the marriage count entirely, instead of counting against it.
7. Location Quality (default 10%) — How complete and well-structured your place names are. Good location data includes a country, region and specific place. Click this bar to see the five measures that make it up (known country, known region, birth location, death location, and geocoding), each with its own count and a link to the matching list.
8. Data Errors (default 10%) — The proportion of your tree that is free from logical errors such as impossible dates or contradictory facts.
9. Date Errors (default 15%) — Dates that could not be understood when your GEDCOM was loaded. Common causes include impossible dates (31 February 2026), free-text notes in date fields ("1923 mum thinks"), and ambiguous numeric formats like 11/9/2001 which means something different in America versus the rest of the world — use unambiguous GEDCOM format (11 SEP 2001) instead. Date errors are scored harshly because an unparseable date is silently discarded, causing the fact to be treated as having no date at all.
10. Fact Coverage (default 5%) — Whether individuals have a reasonable number of recorded life events beyond just birth and death.

Customising weights

The default percentages above reflect a balanced view of tree quality, but everyone's research priorities are different. If you sign in, you can adjust the category weights from Account Settings → Preferences → Adjust weights. If you enjoy regularly being confused about where you found information and never add sources, you can of course set Source Quality to 0% and redistribute that weight to other categories. The total must always add up to 100%.

Filtering

You can filter the health report by surname to focus on a particular family line, and by Relationship Types to see how different parts of your tree compare.

The Data Errors page lists issues that FTAnalyzer has detected in your tree. These are not necessarily mistakes — some may be legitimate unusual situations — but they are worth reviewing.

Common error types

• Facts after death — An event is recorded after the person's death date
• Birth after death — The birth date is later than the death date
• Marriage before birth — A marriage is recorded before either spouse was born
• Unlikely age — A person's calculated age at an event seems unreasonable (e.g. parent at age 8)
• Duplicate facts — The same event appears to be recorded more than once
• Missing data — Key information such as a surname or birth date is absent

How to fix errors

Data errors need to be corrected in your genealogy software (Ancestry, Family Tree Maker, RootsMagic, etc.), not in FTAnalyzer. Once you have made corrections, export a new GEDCOM file and upload it again to see the updated results.

Filtering errors

Use the Error Type dropdown to show only specific kinds of errors. You can select multiple error types at once to focus your review.

Ignoring an error you can't fix

Some errors are legitimate rather than mistakes — for example a 1939 Register entry where the exact birth date genuinely isn't known. If you're signed in, click the eye icon next to any error to ignore it, optionally with a note explaining why. Ignored errors are hidden from the list; tick Show ignored errors to review or restore them later. This isn't available for anonymous sessions, since there's nowhere to remember your choice between visits.

The Census Search page is designed to help you systematically work through a census. It presents family groups as they would appear on the census record — husband, wife and children together — making it easier to locate them. The same individual may appear in multiple family groups if they were a child in one family and a spouse in another.

How to use

1. Select a census date from the dropdown
2. Optionally filter by surname, relationship type, or maximum age
3. Click Show Found to see family groups already recorded on that census
4. Click Show Missing to see family groups not yet found on that census

Working through a census

The "Show Missing" list is particularly useful for focused research sessions. Pick a census and work through the missing families — you can filter by surname to concentrate on one family line at a time. As you find records and update your GEDCOM file, those families will move from the missing list to the found list.

The Census References page checks the quality of census reference data in your tree. Many genealogy programs store census piece, folio and page numbers. FTAnalyzer can validate these references and tell you which ones need attention.

Report types

• Good — census facts with valid, recognised references
• Missing — census facts with no reference recorded at all
• Incomplete — references that are partially entered (e.g. missing page number)
• Unrecognised — references that don't match any known census format

Click any report button to load the matching facts. Use the grid to sort and filter, then fix issues in your genealogy software.

The World Wars page helps you find men in your tree who were of fighting age during World War I or World War II. Click either button to search with the appropriate date ranges.

Date ranges used

• World War I — born between 1869 and 1904, with a death date from 28 July 1914 onwards
• World War II — born between 1894 and 1931, with a death date from 1 September 1939 onwards

Filtering

• Surname — narrow results to a specific family name
• Military facts only — show only individuals who have military or service number facts recorded in the GEDCOM file

Click the article icon next to any individual to see all their facts.

The On This Day page shows events from your family tree that occurred on a specific date. Select a date and click Search to see births, marriages, deaths and other events that happened on that day across all the years in your tree.

Options

• Show whole month — instead of a single day, shows all events that occurred in the selected month across all years

Census and occupation facts are excluded to keep the results focused on life events. Click the article icon next to any event to see all facts for that individual.

World events

If Show world events on the "On This Day" tab is enabled in Settings, a second "On this day in history" panel shows notable historical events for the selected date, sourced from Wikipedia, each with a link to the full article. This is only available for a single day — not whole-month searches.

The Parent Age Report shows the spread of ages at which fathers and mothers in your tree had children, grouped into five-year age bands and split by the child's gender. It's a quick way to spot unusually young or old parents worth double-checking, or just to see the overall pattern across your tree.

Only children with a known birth date, whose parent also has a known birth date, are counted. The chart and the table below it show the same data — the table is there for anyone who prefers exact numbers or can't easily read the chart.

Use the Relationship type filter to narrow the report to, for example, your direct ancestors only.

The Duplicates page helps you find individuals who may have been entered more than once in your tree. FTAnalyzer compares every individual against every other individual using a scoring system that considers name similarity, birth date proximity, location matches, gender, and shared parents.

Using the score slider

The Minimum match score slider controls how aggressive the duplicate detection is. A lower score shows more potential matches (including weaker ones), while a higher score limits results to only the strongest matches. Start with a moderate value (around 40) and adjust as needed.

Settings

• Ignore unnamed twins — When checked, pairs of individuals with no forename who share the same parents and birth date are excluded. These are often unnamed twins or stillbirths that are not true duplicates.

Interpreting results

Each row shows a pair of individuals side by side with their match score. Higher scores indicate stronger matches. Click the article icon next to either individual to view their facts. If you confirm a pair are duplicates, merge them in your genealogy software and re-upload the GEDCOM file.

The Loose Births page lists individuals whose birth date can be narrowed down based on other facts in your tree. For example, if someone was married in 1850 but has no birth date recorded, FTAnalyzer can suggest a likely birth date range based on the minimum parental age and other family relationships.

The Can be updated to column shows the tighter date range that FTAnalyzer has calculated. To apply the suggestion, update the birth date in your genealogy software and re-upload the GEDCOM file.

The Loose Deaths page works the same way as Loose Births but for death dates. It lists individuals whose death date can be narrowed down based on other facts such as census appearances, children's birth dates, or other recorded events.

The Can be updated to column shows the suggested tighter date range for the death.

The Loose Info page is a combined view showing individuals who have either a loose birth date, a loose death date, or both. It provides both suggested updates in a single grid, making it convenient to review all date improvements at once.

The Birth can be updated to and Death can be updated to columns show the suggested tighter date ranges. Click the article icon to view all facts for an individual.

Many pages display your data in a grid (table). The grids on the Individuals, Families, Locations, Sources, Occupations, Surnames, Data Errors and Facts pages all work in the same way.

Sorting

Click any column header to sort by that column. Click the same header again to reverse the sort order. A small arrow indicates the current sort direction.

Filtering

Below each column header is a filter field. Type into it to show only rows that match your text. Filtering is not case-sensitive — typing "smith" will match "Smith", "SMITH" and "Blacksmith".

Paging

Grids show 50 rows per page. Use the page controls at the bottom of the grid to move between pages. The total number of matching rows is shown as a badge next to the page heading.

Resizing columns

Drag the border between column headers to make a column wider or narrower. This can be helpful when names or locations are long.

Exporting to CSV

Where available, a CSV export button appears near the top of the page. Click it to download the current data as a spreadsheet-compatible file that you can open in Excel, Google Sheets or similar software.

The Colour Census report shows, at a glance, which census records you have found for each person in your tree. Each row represents a person and each column represents a census year. The cells are colour-coded to show the status of each census entry.

What the colours mean

Searching for census records

Click any coloured cell to search for that person in the census year using the search provider selected at the top of the page. This uses automated searching to open the search in a new browser tab with the person's name, dates, and age range pre-filled — saving you the time of navigating to the right form and entering the details yourself.

What is Lost Cousins?

Lost Cousins is a free service that matches your ancestors with those researched by other genealogists. By entering your census data on Lost Cousins, you can discover relatives researching the same family lines. The yellow and orange cells in the Colour Census report help you track which entries you have and have not submitted.

You can now view your Lost Cousins reports and upload your census entries to the Lost Cousins website directly from this app — see the Lost Cousins section below.

Filters

Use the Filter dropdown to show only people with a particular census status (e.g. "No census" to focus your research). The Region dropdown narrows results to a specific area. The Relationship Types filter lets you focus on direct ancestors, blood relatives or other groups.

The Lost Cousins pages help you get your census data onto the Lost Cousins website, which matches you with other researchers who share your ancestors. Lost Cousins itself supports nine censuses; matching in Family Tree Analyzer works from seven of them: England & Wales 1841, 1881 and 1911, Scotland 1881, Canada 1881, and the United States 1880 and 1940 (see below for the two not yet supported). The reports below are available to everyone. A Website Sync feature that automates entering your census records on Lost Cousins directly is currently being tested and will be available to everyone soon.

LC Report

The LC Report page shows, for each supported census year, how many of your relatives already have a Lost Cousins fact recorded in your tree and how many have a census entry that has not yet been added to Lost Cousins. It also highlights problems to review, such as duplicate Lost Cousins facts or Lost Cousins facts without a matching census fact. Use the Relationship Types filter to focus on the relatives you research.

Important: these counts come entirely from what's recorded in your own tree, not from the Lost Cousins website itself. If you've been adding entries on your My Ancestors page there directly for a while, the report will still show those people as "missing" until that's reflected in your tree — see below for how to fix this.

Census Entries

The Census Entries page lists the people behind those numbers for any of the eight census years — either those still to enter on Lost Cousins, or those already entered. You can jump straight to it from any row of the LC Report.

Already entered people directly on Lost Cousins?

Until Website Sync is available to everyone (see below), FTAnalyzer has no way to see what's already on your Lost Cousins My Ancestors page — it can only read what's in your GEDCOM file. If you've been entering records on the Lost Cousins website by hand, you can make the reports accurate straight away by adding a custom fact/event in your family tree program for each person already entered:

• Name the custom fact/event type Lost Cousins.
• Date it with just the census year it relates to (e.g. 1881) — one Lost Cousins fact per census year the person is entered for.

FTAnalyzer recognises this fact and will treat the person as already entered for that census year the next time you upload your GEDCOM — no need to wait for Website Sync.

Website Sync

The Website Sync page reads and checks your Lost Cousins My Ancestors page directly, making the manual custom fact described above unnecessary. Uploading the matched records straight to the website (Step 4) is still being tested, so for now that step shows a "coming soon" message — use the earlier steps to check and find records, then add them on the Lost Cousins website by hand in the meantime. It works in four steps:

1. Log in with your Lost Cousins email and password. Your password is sent only to the Lost Cousins website and is never stored by Family Tree Analyzer. (Not a member? Joining is free using the link above.)
2. Check My Ancestors reads the entries already on your Lost Cousins account, so the upload can skip anything already there. This also makes the manual custom fact described above unnecessary — your reports will reflect what's really on Lost Cousins. Each entry found is checked against your tree by census reference and name, and shown with a green tick (linking straight to that person's facts) if matched. If no reference match is found, a second check looks for a same-name, same-birth-year, same-census person in your tree who simply has no usable census reference recorded at all — shown as a blue question mark, also linking to that person's facts. This is a lower-confidence hint, not a confirmed match: it means "this is probably them", so add the reference shown on the Lost Cousins website to their source citation to confirm it (and let it match properly next time). An amber warning icon means neither check found anyone. A summary line above the list totals matched, possible-match, and not-found counts. A "Breakdown by census" button opens a popup splitting the matched/not-found counts by census year, further split into entries you added on the Lost Cousins website yourself versus entries Family Tree Analyzer added for you. An "Export CSV" button above the list downloads the full results (it opens in Excel or Google Sheets like any spreadsheet). A second tab, By person in my tree, shows the same check from the other direction — one row per person in your own tree instead of one row per Lost Cousins entry — which is much quicker to scan than the main list if your loaded GEDCOM file is small compared to your whole Lost Cousins account. It has its own "Breakdown by census" and "Export CSV" buttons, matching the main list. Any link that opens a person's facts from either view opens in a new browser tab, so your results stay on screen.
3. Find records searches your tree for census entries with a valid reference that are not yet on Lost Cousins. Records with incomplete or unrecognised census references are listed separately so you can fix them in your family tree program — each list also has its own "Export CSV" button. If you're signed in, matches confirmed against your My Ancestors page are remembered, so they're restored automatically the next time you load this tree — you won't need to check your My Ancestors page again just to keep your reports accurate.
4. Upload sends the selected records to your My Ancestors page. You must first confirm the root person of your tree is correct, because each entry includes how the person relates to you. Use the practice run option to see exactly what would be uploaded without sending anything. This step is still being tested, so for now it's only available to site admins — everyone else sees a "coming soon" message here instead.

After uploading, visit your My Ancestors page on the Lost Cousins website and click the arrow next to each new household's census reference to check the entry — Lost Cousins requires this step for matching to work.

Entries you upload are remembered against your Family Tree Analyzer account, so the next time you load the same tree the reports will still show them as entered — no need to edit your GEDCOM file.

Census reference formats Lost Cousins needs

Lost Cousins only stores a handful of fields per census year, and they're not always the same fields your source citation captures. If "Check My Ancestors" (Step 2 above) shows an entry as not found in your GEDCOM despite you being sure it's there, checking your citation against the list below is a good first step.

You don't need to write your citation in the exact layout shown for each census below — Family Tree Analyzer reads whatever format your source uses and reformats it into what Lost Cousins needs automatically. What matters is the information: each entry below is the minimum set of numbers Lost Cousins requires for that census, not a specific format to copy. If your citation captures all of them somewhere, matching will work; if one is missing altogether — most often because a hint-based source only captured a page number and dropped a schedule or district number that was never on the same field — no amount of reformatting can recover it, and it needs adding to the citation by hand.

• England & Wales 1841

  Piece, Book and Folio numbers plus a Page number, e.g. "246/13/9/2". If matching still fails, add a note to the fact in this exact form: "HO107/246/13/9/2" (Piece/Book/Folio/Page).

• England & Wales 1881

  Piece and Folio numbers plus a Page number, e.g. "246/13/9".

• England & Wales 1911

  A Piece number and a Schedule number only — there's no field for a page number at all. If your citation only has a page number and no schedule number, it can't be matched; check your source (the schedule number is usually printed on the same page) and add it to your citation.

• Scotland 1881

  The numeric Registration District, an Enumeration District and a Page number — not the parish name. Family Tree Analyzer works out the Registration District from your citation's parish name automatically, so just make sure your citation names the parish clearly. If matching still fails, add a note to the fact in this exact form: "225/7/15 Scotland 1881" (Registration District/Enumeration District/Page).

• US Federal Census 1880

  Just the microfilm Roll number and a Page number, e.g. "3764/246D" — no Enumeration District, even if your citation includes one. A leading zero on the roll number (e.g. "03764") is stripped automatically. If matching still fails, add a note to the fact in this exact form: "3764/246D US 1880" (Roll/Page).

• US Federal Census 1940

  The microfilm Roll number, an Enumeration District and a Page number, e.g. "3764/4-86/16B". As with 1880, a leading zero on the roll number is stripped automatically.

• Canadian Census 1881

  Lost Cousins needs the census District number, a Page number and a Family number, e.g. "186/9/30" — but most citations for this census only capture the microfilm Roll number (e.g. "Roll: C_13283"), a completely different, unrelated number the District can't be worked out from. Where the citation also names the place (e.g. "Census Place: Woodlands, Marquette, Manitoba"), Family Tree Analyzer now works out the District automatically using Library and Archives Canada's own 1881 district finding aid — no action needed. This can't resolve a place name that's genuinely ambiguous between provinces (a handful of township names are reused), or a citation with no place name captured at all. If matching still fails for a Canadian entry, look up the District number for that record on FamilySearch and add a note to the fact in this exact form: "186/9/30 Canada 1881" (District/Page/Family).

• Ireland 1911

  The National Archives of Ireland's own 9-digit reel number, e.g. "002247382". Family Tree Analyzer reads this straight out of a citation URL from either the archive's current viewer (nai.prod.derilinx.com) or its older one (census.nationalarchives.ie) — the number immediately after "nai" in the address. If matching still fails, add a note to the fact in this exact form: "nai002247382". If your citation instead links to the archive's now-retired per-record page (e.g. "census.nationalarchives.ie/pages/1911/...") — an older format that can't be converted automatically — use the 1911 Irish Census Helper under the Census menu to look each one up and get a ready-to-use citation.

Newfoundland 1921 is also supported by Lost Cousins, but automatic matching isn't implemented in Family Tree Analyzer yet.

One more quirk worth knowing about: when a household spans two census pages, Lost Cousins records every household member against the head of household's page number, even for people whose own entry is genuinely on the next page. Family Tree Analyzer accounts for this automatically — matching (and, once uploading is available to everyone, the records it sends to Lost Cousins) always uses the head of household's own citation for the whole family group, not each person's individual page number. This uses the family a person belongs to as a stand-in for "household", so it won't help with an unrelated boarder or lodger who was recorded on the same census page but isn't part of the family in your tree. If a match still doesn't happen after this, Step 2's grid and CSV export show a Lost Cousins Reference (the website's own value) and a GEDCOM Reference (built from your own citation) side by side, to help spot what's different.

The Colour BMD (Birth, Marriage, Death) report shows how precise the dates in your tree are, using colour-coded cells. This helps you identify where you have exact dates, rough estimates or missing information.

Date precision colours

Each Birth, Marriage and Death cell is coloured on a scale from red (unknown) to green (exact):

Marriage status colours

The Marriage column may also use these additional colours:

Using the report

Click any coloured cell to search for that person's record using the selected search provider. This uses automated searching to open a pre-filled search in a new tab, saving you the time of manually entering the person's details.

Use the Filter dropdown to show only people missing certain dates. The Apply to dropdown lets you apply the filter to Birth, Marriage or Death specifically.

The Research Suggestions page is your starting point for finding gaps in your research. It provides quick access to the Colour Census and Colour BMD reports.

Automated searching

The most powerful feature of the Colour Census and Colour BMD reports is automated searching. When you click a coloured cell in either report, FTAnalyzer automatically opens a search on your chosen provider (FamilySearch, Ancestry, Find My Past, FreeCen, or ScotlandsPeople) with the person's name, dates, and age range pre-filled. Instead of manually typing search criteria for every ancestor in every census year, you can simply work down the grid clicking red cells — each click opens a ready-made search in a new tab.

Each automated search typically saves a minute or two compared to manually navigating to the right search form, entering the person's name, selecting the correct date range, and setting the right search options. That adds up quickly — an evening's research session with dozens of searches can save you an hour or more of repetitive form-filling.

Tip: log in to your chosen search provider (e.g. Ancestry, FamilySearch) in a separate browser tab before you start clicking cells. If you aren't already signed in, the new tab will land on the provider's sign-in page instead of your search results, so you'll need to log in and re-run the search.

Census reports

Buttons are provided for census reports by country — UK, Ireland, US and Canada. Each opens the Colour Census report filtered for that country, showing which census years you have covered and where gaps remain.

BMD report

The Colour BMD button opens the Birth/Marriage/Death report, showing how precise your recorded dates are and highlighting where further research could improve them.

Filters

You can narrow results before opening a report by selecting a Relationship Type and/or surname. These filters carry through to whichever report you open.

What are facts?

In genealogy, a "fact" is any recorded event or attribute in someone's life — birth, marriage, death, census, residence, occupation, immigration, military service and many more. Your GEDCOM file stores each of these as a separate fact.

The Facts page

The Facts page lets you search and filter across all facts in your tree. Two filter lists let you:

• Include fact types — Choose which types of facts to display
• Exclude individuals — Hide people who have a particular fact type (useful for finding people who are missing certain events)

You can also choose between All Facts, Preferred Only or Alternate Only. When your genealogy software records multiple versions of the same fact (e.g. two possible birth dates), one is usually marked as "preferred" and the others as "alternate." This filter lets you see each version.

Custom facts

Custom facts are non-standard tags that your genealogy software may have created. These often start with an underscore (e.g. _MILT for military service). The Custom Facts page lists all such tags found in your tree.

You can tick the Ignore checkbox next to any custom fact type to exclude it from the Data Errors list. This is useful when your software creates tags that are valid for your research but are not part of the official GEDCOM standard. Your selections are saved in your browser and remembered across sessions.

A few custom event types are recognised as another name for a standard fact rather than treated as a custom fact — for example "Birth Certificate," "Death Certificate" and "Marriage Certificate" are all folded into the matching standard Birth, Death or Marriage fact. This means they won't appear on the Custom Facts page, and on the Facts page they show as an extra, unlabelled Birth/Death/Marriage row rather than under their original name. If that extra row records the same event as an existing Birth/Death/Marriage fact, it's expected to be flagged as a possible duplicate — that's Data Errors correctly spotting the same event recorded twice under two different GEDCOM tags.

Open Settings by clicking the cog icon (settings) in the top-right corner. Settings are grouped into sections.

Appearance

Choose between System (follows your device preference), Light or Dark theme.

Dates

• Use baptism/christening date — When someone has no birth date, use their baptism or christening date instead. This is helpful for older records where only church records survive.
• Use burial/cremation date — Similarly, use a burial or cremation date when no death date is available.

Census

• Treat residence facts as census facts — Count residence events as census entries. Useful if your software records census data as "Residence" rather than "Census." Ancestry is the worst offender for this sort of bug.
• Tolerate slightly inaccurate census dates — Accept census facts even if the date is not exactly on the official census day.
• Family census facts apply to parents only — When a census fact is attached to a family rather than an individual, apply it to the parents only, not the children.
• Auto-create census events — Automatically detect census information in notes and sources and create census facts from them.
• Convert residences with census references — Change residence facts to census facts when they contain census reference numbers.
• Skip checking for census references — Do not search notes and sources for census reference numbers.
• Use compact census references — Display census references in a shorter format.
• Hide people tagged as missing — Do not show people you have tagged as missing from the census.

Locations

• Allow empty values in locations — Permit blank fields within a location hierarchy (e.g. a country and address but no region).
• Country-first locations — Enable this if your GEDCOM file records locations with the country first (e.g. "England, London, Westminster") rather than the more common address-first format.
• Add auto-created census locations — Include locations generated from census facts in the Locations list.
• Skip fixing location data — Disable the automatic normalisation and correction of place names.

Display

• Show alias in name — Display any recorded alias alongside the person's main name.
• Show multi-line ancestors — When an ancestor appears in multiple direct lines (common with cousin marriages), show them in each line.
• Multiple fact forms — When viewing potential duplicates, show all versions of their facts.
• Show current options in reports — Display the active filter settings within reports.
• Show world events — Show historical events alongside your ancestors on the "On This Day" tab.

Data Processing

• Ignore custom fact type warnings — Suppress warnings about non-standard GEDCOM tags.
• Treat female surnames as unknown — Useful for research where married women's maiden names are not recorded and you want to avoid inheriting the husband's surname.
• Hide ignored duplicates — Do not show duplicate individuals that you have previously dismissed.
• Include alternate facts — Load alternate (non-preferred) versions of facts. Disable this if you only want to see your preferred data.
• Minimum parental age — The youngest age at which someone can be considered a parent. Facts suggesting parenthood below this age are flagged as errors.

Family Tree Analyzer maintains a shared database of geocoded locations (latitude/longitude lookups) used to plot maps across the site. Any logged-in user can help grow this database by supplying their own free Google Geocoding API key on the Geocoding page.

What you're contributing to

When you run geocoding, you are looking up whichever locations in the shared queue need it most — not necessarily the locations from your own GEDCOM file. There is currently no way to prioritise your own tree's locations ahead of everyone else's. Your contribution helps every user of the site, and in turn you benefit from geocoding that other contributors have already done.

Master Geocodes

This is the shared database of locations that have already been geocoded — every map across the site draws its pins from here. Each entry has one of three statuses:

• Matched — Google returned a confident, exact match for the location at the correct level (e.g. a town matched as a town, not a wider region). These pins are placed with full confidence.
• Partial — Google could only find an approximate match, for example a place whose exact name or boundary it isn't sure of. A common case is a street address with a house number — e.g. "10 Downing Street, London, England" — where Google can only confirm the street itself ("Downing Street, London, England") and not the specific house number. The pin is still placed, but may be slightly off or represent a broader area than intended.
• Level Mismatch — Google matched the location, but at a different level than expected (for instance, a sub-region in the GEDCOM came back matched only at the country or region level). The pin is usable but less precise than a full match.

Location Candidates

These are locations awaiting geocoding — not yet in the master database. They move through a pipeline of statuses as they're processed:

• Imported — Newly added from a GEDCOM upload, or backfilled as the missing parent of a candidate (e.g. "Toscana, Italy" is auto-added as the parent of "Pisa, Toscana, Italy"). Imported candidates wait for a scan before they can be geocoded.
• To Geocode — A scan has confirmed this candidate's parent location (its country, region, etc.) is already in the master database, so it's ready to be looked up via the Google API. Only candidates with this status are picked up by "Prepare & Geocode".
• Geocoded — Successfully looked up and promoted into the master database. It no longer appears in the candidates list.
• No Match — Google couldn't find this location, or returned a poor result. It is moved to the ignored locations list rather than retried indefinitely.

Because a candidate is immediately moved out to the master database or the ignored list the moment it's resolved, it never sits in the candidates table with a Geocoded or No Match status — so the Location Candidates stats on the Geocoding page only ever show Imported and To Geocode counts.

Candidates only become eligible for geocoding once their parent location is already known — this avoids wasting lookups on a sub-region before its country or region context exists. That's why "Scan Only" can be useful on its own: it promotes Imported candidates to To Geocode as their parents are resolved, without spending any API lookups.

Step-by-step: getting a free Google Geocoding API key

1. Go to the Google Cloud Console and sign in with (or create) a free Google account.
2. Create a new project — click the project dropdown at the top of the page, then "New Project", give it any name (e.g. "FTAnalyzer Geocoding"), and click "Create".
3. Enable billing for the project. Google requires a billing account even for free-tier usage, but you will not be charged unless you exceed the free monthly allowance. See Billing in the console to add a payment method.
4. Enable the Geocoding API — search for "Geocoding API" in the console's search bar, open it, and click "Enable".
5. Create an API key — go to "APIs & Services" → "Credentials" → "Create Credentials" → "API key". Copy the key that's generated.
6. (Recommended) Restrict the key — click on the new key, under "API restrictions" choose "Restrict key" and select only "Geocoding API". This stops the key being used for anything else if it's ever exposed.
7. Paste the key into the API Key field on the Geocoding page. It is held in memory for your session only — Family Tree Analyzer never saves your key to its database, disk, or configuration. You'll need to re-enter it each time you want to contribute. You don't need to write the key down anywhere — you can always come back and copy it again from the Google Cloud Console credentials page.

Usage limits and cost

Google's Geocoding API includes 10,000 free lookups per month. Beyond that, Google charges $5 per 1,000 lookups. Use the "Max Lookups This Run" field on the Geocoding page to cap how many lookups a single run can make. The Geocoding page also shows your own estimated usage for the current month, but the authoritative count is on the Google Cloud Console API dashboard — check it before increasing your limit, especially once you're close to 10,000 for the month.

There are three map views, each showing your tree's geocoded locations in a different way:

• Locations Map plots every geocoded place in your tree, grouped into clusters when they're close together — click a cluster to zoom in, or click a single marker to see the individuals and facts recorded at that place.
• Timeline Map plots every dated event across your whole tree for one chosen year at a time — drag the year slider to see how the family's locations changed over time.
• Lifeline Map plots one person's life journey — choose someone from the list and see a line connecting every place they're recorded, in date order.

What the markers mean

Marker colour and shape both show the same thing — a location's status in the shared geocoding database — so the information doesn't rely on colour alone:

• Green circle — Matched. A confident, exact match.
• Orange triangle — Partial Match. An approximate match — the pin may be slightly off or cover a wider area than intended.
• Blue square — Level Mismatch. Matched, but at a broader level than expected (e.g. a town matched only down to its region).
• Grey diamond — Other. Not yet matched with confidence.

When a place's exact address hasn't been geocoded, these maps show it at the nearest geocoded region or country instead of leaving it off the map — drawn faded with a dashed outline so you can tell it's approximate. If you'd rather only see exact matches, turn this off under Locations in Settings (the gear icon).

See Helping with Geocoding for a fuller explanation of these statuses, and how you can help improve them.

Switching the background map

Use the "Base map" buttons above each map to switch between OpenStreetMap, an Esri street map, and — if your tree has UK locations — two free historical Ordnance Survey maps of Scotland from the 1840s-1880s, courtesy of the National Library of Scotland. The small i button next to the base map buttons explains more, including how to unlock a third historical map covering all of Great Britain — see Historical Maps Key below.

The map is read-only

You can't edit or correct a location's pin position directly from the map — that protects the shared database that every user relies on. To fix a location, use the Locations page, or help improve the database generally via Help with Geocoding.

Exporting to KML

The Export KML button on the Locations page downloads a KML file listing every geocoded place in your tree, along with the individuals and families recorded there. KML is a standard format you can import into Google Maps, Google Earth, and most other mapping software, to explore your tree's locations outside Family Tree Analyzer. Very large trees produce a compressed .kmz file instead — Google Maps and Google Earth open these the same way.

The free historical Scottish map already shown on the Locations Map needs nothing from you — it just works. This section is only for anyone who wants to unlock additional historical map backgrounds in future, which need a free personal key from a company called MapTiler. There's no obligation to do this — it's entirely optional.

The steps below are written for anyone, even if you've never used a website like this before. Take it one step at a time — none of it requires installing anything on your computer.

What is an "API key" and why do I need one?

An "API key" is just a long password-like code that proves to MapTiler that it's really you asking to see their maps, so they can keep their service free for everyone. You only need to get one once — after that, Family Tree Analyzer remembers it for you (see "Where is my key stored?" below).

Step-by-step: getting your free MapTiler key

1. Open a new browser tab and go to www.maptiler.com/cloud.
2. Look for a button that says something like "Get started for free" or "Sign up" — usually near the top right of the page — and click it.
3. Create your free account. You can normally sign up using your existing Google account (look for a "Continue with Google" button) so you don't need to remember a new password — or use your email address if you prefer.
4. Once you're signed in, you'll land on your MapTiler "Dashboard" — a page that lists your account details. No payment details are required for the free plan.
5. On the left-hand side of the Dashboard, look for a menu option called "API keys" and click it.
6. You should see a key already listed (often called "Default key"), shown as a long string of letters and numbers — something like AbCdEfGh12345.... To copy it: click at the start of the key, then drag your mouse to the end to highlight the whole thing (or triple-click on it to select it all in one go), then right-click the highlighted text and choose "Copy" (or press Ctrl+C, or Cmd+C on a Mac). If no key is listed, look for a "New key" or "+" button to create one first, then copy it the same way.
7. Come back to Family Tree Analyzer. Click your name or profile picture in the top right of the page, choose "Account Settings", then open the "Historical Maps Key (MapTiler)" section — or just go straight to Account Settings. Click into the "MapTiler API Key" box and paste your key in — on most computers this is Ctrl+V (or Cmd+V on a Mac).
8. Click the "Save" button. You're done — you won't need to do this again.

If you'd like to double-check what you typed before saving, click the small eye icon next to the box to reveal the key as plain text instead of dots. There's no copy button on this box either — if you ever need to copy your saved key back out, reveal it the same way, then highlight it and right-click → "Copy" (or Ctrl+C / Cmd+C) as described above.

Where is my key stored?

Unlike the Google key used on the Help with Geocoding page (which has to be re-entered every time, for privacy reasons), this MapTiler key is saved against your Family Tree Analyzer account, so you only need to enter it once. It is only ever used to fetch map images on your behalf, and is never shared with other users. You can remove it at any time from the "Historical Maps Key" section on your Account Settings page.

Will this cost me anything?

No. MapTiler's free plan allows up to 100,000 historical map views per month — far more than you're ever likely to use browsing your own family tree. If that limit is ever reached, historical maps simply stop working until the start of the next month — MapTiler does not charge you automatically, unlike some other mapping services.