Most serious chess still gets recorded the old-fashioned way: a pencil, a paper scoresheet, and a player too deep in the position to think about digitizing anything. Those sheets pile up in a folder and rarely get a second look. That is a shame, because a club game you actually played is far better study material than yet another grandmaster brilliancy.
This guide walks through the full pipeline, from a stack of handwritten sheets to a polished, shareable Lichess study with one chapter per game. It is written for club players reviewing their own games and for coaches assembling material for students.
What You Will Need
- The paper scoresheets you want to digitize.
- A phone or scanner to photograph them.
- A free ScanChess account for the recognition and PGN export.
- A free Lichess account for building the study.
That is the entire toolchain. No special hardware, no manual transcription.
The Workflow, Step by Step
1. Photograph the scoresheet
Lay the sheet flat in even light and shoot it straight on from above. Avoid hard shadows, and fill the frame with the sheet so the move text is large and readable. If a game runs across two sheets, photograph each one clearly. Sharp, well-lit photos are the single biggest factor in clean recognition.
2. Scan it at ScanChess
Open the board recognition scanner and upload the photo of your scoresheet. ScanChess reads the handwritten moves and converts them into standard algebraic notation (SAN), the same e4 e5 Nf3 form you see in books and databases. The recognizer also runs move validation, so it flags entries that are illegal or ambiguous in the position rather than silently passing along a misread. This used to mean ten minutes of squinting and typing; now it is a single upload. For the deeper background on how this handwriting-to-PGN conversion works, see Convert a Handwritten Scoresheet to PGN.
The same scanner also handles board photos: snap a picture of a physical position and ScanChess returns a FEN string. That is handy when a game starts from a set position, but for full games the scoresheet path is what you want.
3. Review and correct the recognized moves
Handwriting is handwriting. A b and an h, or a sloppy 6 and 8, can trip up any reader. Walk through the recognized move list against the paper sheet and fix anything the validator flagged. Because each move is checked for legality in the live position, most errors surface immediately rather than hiding until you replay the game.
Use the built-in replay to step through the game move by move. If the board reaches a position that does not match your memory of the game, you have found the spot to correct. This pass is quick, and it is the difference between a study you trust and one you have to second-guess. For more on checking recognized games, see Analyze Chess Games from Scoresheets.
4. Export the PGN
Once the moves are clean, export the game as a PGN file. PGN (Portable Game Notation) is the universal plain-text format that virtually every chess platform reads, Lichess included; What Is PGN? covers the format in plain language. Repeat steps 1 through 4 for each game you want, and keep the exported files together for the next step.
5. Create a Lichess study
Log in to Lichess and create a new study from the Studies area. A study is a container for one or more games, each held in its own chapter, and it is a natural home for coaching material because chapters carry variations, comments, and drawn shapes on the board.
When you set up the study you choose its visibility, typically private while you build it and a public or unlisted link when you are ready to share. You can also invite specific people as contributors if you want students or teammates to add their own notes.
6. Import the PGN, one chapter per game
Add a chapter and choose the option to import a PGN, then paste or upload the file you exported from ScanChess. Lichess parses the movetext and builds a navigable chapter you can click through from the first move to the result. Add a separate chapter for each game and give each a clear title, such as opponent and event, so the study reads like a tidy table of contents rather than a pile of unlabeled positions.
If you would like a closer look at the import side specifically, Import PGN to Lichess walks through the details.
7. Annotate with arrows and comments
This is where a study earns its keep. On any move you can add a text comment explaining the idea, the plan, or the mistake. Hold the right mouse button and drag on the board to draw arrows and highlight squares, the fastest way to point out a tactic or a weak pawn without writing a paragraph. Build out the critical variations as branches so a student can explore "what if I had played this instead" inside the chapter. Keep annotations focused: a few well-placed arrows and two or three sharp comments per game teach more than a wall of text.
8. Share the study link
When the study is ready, set it to public or unlisted and copy the link. Send it to a student, post it to your team channel, or work through it live in a lesson. Anyone with the link can replay every game, read your comments, and step through the variations at their own pace, with nothing to install.
Why This Beats Manual Entry
Typing a game by hand is slow and error-prone, and the friction means most paper games never get studied at all. Scanning collapses the tedious part into a single upload and a quick review, making it realistic for a whole stack of sheets. Coaches can turn a weekend tournament into a ready-to-teach study; club players can finally mine their own games for patterns. For the broader options, the convert chess scoresheet to PGN and digitize chess games overviews lay out the approaches.
Start with One Game
You do not need to digitize a whole season tonight. Pick your most interesting recent game, photograph the sheet, and run it through the scanner to see how fast clean PGN comes out. From there, building a Lichess study is just paste, annotate, and share.
Open the scoresheet scanner and turn your first paper game into a study chapter. New accounts get free starter credits, so you can try the full workflow first.