Why your spreadsheet will outlive your job, your company, and your byline, and what a 1982 home computer already knew about it
An article gets read once. A spreadsheet gets saved, renamed, emailed to a colleague, uploaded to a shared drive, and opened again in 2031 by somebody who has never heard of you but is reading a tab with your name on it. That is a different kind of publishing, and almost everyone in professional services is doing it by accident. What follows is the case for doing it on purpose.
1. What is the real difference between publishing an article and publishing a file?
Reach and half-life. An article’s audience is the people who read it during its news window. A file’s audience is everyone who ever receives a forward of it, forever, and that group grows while you sleep.
- Articles decay. They peak in days and are effectively archived within weeks.
- Files compound. A reserve model, a pricing calculator, a diligence checklist, or a one-page underwriting worksheet gets adopted into somebody’s workflow and copied every time a new analyst joins.
- Only one of them carries your name into the future, and only if you put it there. Files do not come with bylines by default.
2. How much do these files actually move? Show me numbers.
The best public evidence comes from an unexpected place. When Enron collapsed, federal regulators subpoenaed the company’s email, and researchers Felienne Hermans (Delft University of Technology) and Emerson Murphy-Hill (North Carolina State University) pulled every spreadsheet out of it. Their 2015 paper is the closest thing we have to a photograph of how business files really behave inside a large organization.

Chart 1. Source: Hermans & Murphy-Hill, “Enron’s Spreadsheets and Related Emails: A Dataset and Analysis,” ICSE 2015.
| Finding from the Enron archive | Figure | Why it matters to you |
|---|---|---|
| Email attachments extracted | 265,586 | Of these, 51,572 were Excel files, containing 16,189 unique spreadsheets |
| Spreadsheets emailed per day | ~100 | Sharing was not an event. It was the ambient weather of the company |
| Emails sending or mentioning a sheet | 9.6% | Roughly one in ten messages in a 717,102-email archive touched a spreadsheet |
| Sheet emails about revisions/updates | 20.4% | One in five forwards was a new version, so the file kept mutating in transit |
| Files with formulas containing an error | 24% | The thing being copied around was frequently wrong, and copied anyway |
| Average worksheets per file | 5.1 | The record holder had 175 tabs. Plenty of room for a footer |
Note the number that should worry you: not one of those 15,770 analyzable spreadsheets recorded who built it. The researchers had to reconstruct authorship from surrounding emails, and often could not.
3. What does a Commodore 64 have to do with my pricing calculator?
More than you would guess. The C64 launched in August 1982 at $595 and, per Guinness World Records, remains the best-selling desktop computer model ever made, with credible estimates from 12.5 million to 17 million units. Its software library ran to nearly 10,000 commercial titles. It also ran spreadsheets: Microsoft’s Multiplan, Calc Result, PractiCalc. Dan Bricklin and Bob Frankston even ported VisiCalc to the Commodore PET.
Here is why that machine is the perfect case study. The C64 was the single most heavily copied computing platform of its era. Software moved from bedroom to bedroom on cassettes and floppies, stripped of packaging, manuals, and boxes. It was, in other words, a full-scale experiment in what happens to a creator’s name when files travel without them. That experiment ran for a decade, and it produced two very different outcomes depending on one design decision.
4. What was the outcome when the name was left out?
Consider the most famous program ever written for the machine:
10 PRINT CHR$(205.5+RND(1)); : GOTO 10
One line. It fills the screen with an endless random maze, and it is genuinely beautiful. It appeared in a three-line variant in the Commodore 64 User’s Guide in 1982, spread through magazines and user groups, and became a folk artifact of home computing.
- Ten academics wrote a whole book about it. MIT Press published a 294-page study of this single line in 2012, authored by Nick Montfort, Ian Bogost, Casey Reas, Mark Sample and six others.
- They still could not tell you who wrote it. The line is treated as a text that surfaced in many printed sources. Its author is lost.
- The lesson is brutal and simple. You can create something so good that a university press assembles ten scholars to study it, and still be anonymous, because the file had no place to put your name.
5. And what was the outcome when the name was left in?
The C64’s sound chip produced music in a file format, the .sid, whose header reserves 32 bytes each for title, author, and copyright. That single field is the reason the 1980s bedroom composers are famous today and the maze programmer is not.

Chart 2. Source: High Voltage SID Collection release notes, updates #64 to #85, Dec 2015 to Jun 2026.
- The archive is enormous and still growing. As of update #85 in June 2026, the High Voltage SID Collection holds 61,157 tunes, each naming its composer.
- Attribution is treated as maintenance. That June 2026 release alone logged 312 credit fixes, 12 tunes moved out of the unattributed folder into their composers’ directories, and 3 previously unknown demo tunes identified.
- The payoff is a career, decades later. Rob Hubbard, whose name rode inside those files, was hired by Electronic Arts in 1987, received an honorary doctorate from Abertay University in 2016, and has had his C64 work performed by orchestras in the 8-Bit Symphony concert series.
6. Did anyone else figure this out at the time?
The software pirates did, and they were ruthless about it. When crackers removed copy protection from a game, they attached a short animated screen in front of it announcing who had done the work. It is called a cracktro.
Researchers Markku Reunanen, Patryk Wasiak and Daniel Botz documented the practice in the International Journal of Communication in 2015. Their key structural observation: moving the cracker’s signature into a separate screen, rather than defacing the game’s own title, meant the credit could be reused across every future release. The signature became a portable asset. The site intros.c64.org catalogs 9,225 of these intros from 1,980 named groups.
Read that again in business terms. The pirated copies traveled far wider than the legitimate ones, and because the crackers stapled a reusable credit to the front of every file, their names outlasted most of the publishers whose games they were copying.

Chart 4. Sources: HVSC update #85; intros.c64.org via Reunanen et al. 2015; Wikipedia C64 software library; Hermans & Murphy-Hill 2015; EUSES collection (Fisher & Rothermel, 2005).
7. Does this really happen to serious business frameworks?
Constantly. The most widely used strategy tool on earth is an orphan.
| Framework | What happened to the credit | Evidence |
|---|---|---|
| SWOT analysis | No universally accepted creator. Usually credited to Albert Humphrey at Stanford Research Institute, and disputed | StatPearls (NIH) states the attribution “remains debatable” and there is no accepted creator |
| The SRI origin report | Robert F. Stewart’s 1963 report, the most cited source in the field, is not publicly available today | Journal of Management History, 2025 historical reconstruction |
| Humphrey’s own claim | He asserted authorship in a Christmas letter published posthumously in an alumni newsletter in 2005 | “I’m told that Harvard and MIT have claimed credit for SWOT…not so!” |
| Scholarly cleanup | Researchers began digging in 2016 and published the reconstruction seven years later | Emerald / ScienceDirect, “The origins of SWOT analysis” |
| In-copyright works generally | Up to 40% are orphan works whose rights holder cannot be identified or found | British Library estimate, 2011; the UK later opened a scheme covering 91 million orphan works |
8. How does a footer physically fall off a file?
Mostly through ordinary, well-intentioned software behavior. This is the part people underestimate.
| Mechanism | What the evidence shows |
|---|---|
| Platforms strip metadata | The IPTC’s Social Media Sites Photo Metadata Test examined 15 major platforms. Only one, Behance, correctly displayed and preserved all four credit fields (caption, creator, copyright notice, credit line). Several displayed none |
| Resizing destroys it silently | IPTC found metadata often survived a download at original size and was stripped the moment the image was rescaled. Engineers frequently did not know their own pipeline was doing it |
| Newer credentials fare no better | AFIP’s platform testing found C2PA content credentials fully stripped in five of six platforms tested |
| Humans edit the visible parts | A recipient who adapts your template deletes the row that says who made it, because it is not their name and it is in the way |
| Email is the version control system | At Enron, 20.4% of spreadsheet emails discussed revisions. Every hop is a chance for provenance to vanish |
The countermeasure is not technical. It is editorial. Put the credit in the visible content of the document itself, in a place that survives copy-paste, rather than trusting file properties no one looks at.
9. Who has actually made money giving files away with their name on them?
Three people worth studying, in three different fields, all using the same move.

Chart 3. Source: Vertex42 company page and Jon Wittwer’s author page, vertex42.com.
| Who | What they gave away | What it returned |
|---|---|---|
| Jon Wittwer, Vertex42 | 450+ Excel templates, free, no form, founded April 2003 while finishing a mechanical engineering PhD | 170 million site visitors and 80 million downloads. The Gantt chart template alone passed 3 million. He left Sandia National Laboratories in 2008 to run it full time; Microsoft made Vertex42 a templates partner |
| Aswath Damodaran, NYU Stern | Valuation spreadsheets, datasets and lecture notes, published free and updated annually | Became “the Dean of Valuation,” quoted constantly across financial media, with a body of books and courses built on that reputation |
| Alexander Osterwalder & Yves Pigneur | The Business Model Canvas, released under Creative Commons and downloadable without a gate | Business Model Generation has sold in the millions across roughly 40 languages. Strategyzer, founded 2010, serves Colgate-Palmolive, Mastercard and Merck |
The pattern is identical in all three cases. The artifact is free, ungated, unlocked, and permanently stamped with a name and a destination. The business is downstream of the artifact, never in front of it.
10. But doesn’t gating the download capture leads?
It captures a smaller number of better-qualified leads, which is a real trade and sometimes the right one. It is the wrong trade for a template, because a template’s whole value is its ability to keep traveling after the first download.
- The reach gap is very large. Marketing author David Meerman Scott’s widely cited estimate is that ungated assets get downloaded 20 to 50 times more often than gated ones.
- Buyers gate-shop by asset type. Demand Gen Report survey work finds most B2B buyers will trade contact details for original research, and far fewer will do it for something they expect to find free elsewhere. A blank template is squarely in the second category.
- Search cannot index what it cannot open. Gated files are invisible to the crawlers, and increasingly to the AI assistants your future buyer will ask instead of Google.
- The gate stops the second hop, which was the valuable one. The colleague your download gets forwarded to never sees your form. They only ever see the file. Make sure the file introduces you.
11. Fine. What exactly do I put in the footer?
The practical half of this is three lines long, and it genuinely is that short.
| Rule | Specification |
|---|---|
| Sign every sheet | Name, firm, one URL, version and date, in row 1 or the bottom-left cell of every single tab. Not just the cover tab, because tabs get copied out individually |
| Lock nothing | No password, no protected ranges, no hidden formula sheets. People adapt templates. A locked file gets rebuilt from scratch by somebody else, with their name on it |
| Never gate the download | Direct link. No form, no email wall, no “request access.” You are buying distribution, and the price is the lead capture you were not going to convert anyway |
| Add a license line | One sentence: free to use and share, keep the attribution. Creative Commons BY is the standard version of this and costs you nothing |
| Use the file properties too | Set Author and Company in document properties as a backup, while assuming they will be stripped |
| Date it visibly | “v2.1, March 2026” tells a 2031 reader the model is stale and sends them back to your URL for the current one. That is the whole flywheel |
12. What is the worst case if I skip this?
Ask Dan Bricklin. In 1979 he and Bob Frankston shipped VisiCalc, the first spreadsheet for personal computers, the application Steve Jobs credited with propelling the Apple II. Software patents were rarely granted then, their attorney advised against pursuing one, and they let it go. Lotus 1-2-3 arrived in 1983, VisiCorp was bankrupt by 1985, and every spreadsheet you have ever opened is a descendant of their work.
Bricklin’s own summary: “I’m not rich because I invented VisiCalc, but I feel that I’ve made a change in the world.” Harvard Business School named a classroom after him in 1999. That is the good version of losing the credit, and it took a plaque at Harvard to get there.
13. If I were reporting this out, where would I start?
The durability angle is the story, and it is reportable. Pick three or four templates that circulate constantly in your industry and trace them backward.
- Find the templates everyone uses and no one sources. The comparison grid, the reserve model, the diligence checklist. Ask five people where theirs came from and watch the answers dissolve into “someone sent it to me.”
- Trace the file backward through the forwards. Check document properties, formula quirks, formatting fingerprints and stray defined names. The Enron researchers reconstructed provenance from exactly these signals.
- Find the original author and ask the money question. Did they ever get a client, a job, a speaking slot, or a dollar out of it? Most will say no, and most will tell you the footer came off somewhere around the third forward.
- Contrast with someone who kept the footer on. Wittwer, Damodaran and Osterwalder are all reachable and all have said publicly that giving the artifact away was the business decision, not a charity one.
- Close on the C64. Volunteers are still repairing SID credits in 2026, forty years after the fact, because in 1987 someone thought to add an author field. Your spreadsheet has 175 available tabs and you cannot spare one row.
Sources
Every figure above is traceable. Links below, in order of first appearance.
Hermans, F. & Murphy-Hill, E., “Enron’s Spreadsheets and Related Emails,” ICSE 2015 (TU Delft technical report). https://repository.tudelft.nl/file/File_6407791c-a000-4062-acaf-63e04502dfdc?preview=1
NC State University news release on the Enron and Fuse spreadsheet datasets. https://news.ncsu.edu/2015/04/murphy-hill-spreadsheet-2015/
Felienne Hermans, “A modern day Pompeii: Spreadsheets at Enron.” https://www.felienne.com/archives/3634
Guinness World Records, best-selling desktop computer (Commodore 64). https://www.guinnessworldrecords.com/world-records/72695-most-computer-sales
Wikipedia, Commodore 64 and Commodore 64 software (units sold, library size, Multiplan, Calc Result). https://en.wikipedia.org/wiki/Commodore_64_software
Montfort, N. et al., 10 PRINT CHR$(205.5+RND(1)); : GOTO 10, MIT Press, 2012. Free PDF under Creative Commons. https://10print.org/
MIT Press open-access edition of 10 PRINT. https://direct.mit.edu/books/oa-monograph/5343/10-PRINT-CHR-205-5-RND-1-GOTO-10
SID file format description, including the 32-byte author field (HVSC documentation). https://www.hvsc.c64.org/download/C64Music/DOCUMENTS/SID_file_format.txt
High Voltage SID Collection, Update #85 release notes, June 2026 (61,157 files, 312 credit fixes). https://www.lemon64.com/forum/viewtopic.php?p=1101572
High Voltage SID Collection, Update #84 release notes, December 2025. https://www.lemon64.com/forum/viewtopic.php?t=88425
Reunanen, M., Wasiak, P. & Botz, D., “Crack Intros: Piracy, Creativity, and Communication,” International Journal of Communication 9 (2015). https://ijoc.org/index.php/ijoc/article/download/3731/1345
Rob Hubbard career details, C64-Wiki and 8-Bit Symphony. https://www.c64-wiki.com/wiki/Rob_Hubbard
SWOT Analysis, StatPearls, NIH National Library of Medicine (“attribution remains debatable”). https://www.ncbi.nlm.nih.gov/books/NBK537302/
“The origins of SWOT analysis,” Long Range Planning / ScienceDirect, 2023. https://www.sciencedirect.com/science/article/pii/S0024630123000110
“From SOFT approach to SWOT analysis, a historical reconstruction,” Journal of Management History, 2025. https://www.emerald.com/jmh/article/31/2/333/1239245/From-SOFT-approach-to-SWOT-analysis-a-historical
British Library 40% orphan works estimate, cited in the National Library of Wales consultation response. https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/327042/National_Library_of_Wales.pdf
UK Government, “UK opens access to 91 million orphan works.” https://www.gov.uk/government/news/uk-opens-access-to-91-million-orphan-works
IPTC, “Many Social Media Sites Still Remove Image Rights Information From Photos” (15 sites tested). https://iptc.org/news/many-social-media-sites-still-remove-image-rights-information-from-photos/
IPTC Social Media Sites Photo Metadata Test results, 2019. https://iptc.org/standards/photo-metadata/social-media-sites-photo-metadata-test-results-2019/
AFIP, “Metadata Stripping: Why C2PA Fails on Social” (stripped in five of six platforms). https://afip.org/research/metadata-stripping/
Vertex42 company page (170 million visitors, 80 million downloads, 450+ templates). https://www.vertex42.com/about.html
Jon Wittwer author page (Gantt chart 3 million downloads, Debt Reduction Calculator 1.5 million). https://www.vertex42.com/author/jon-wittwer.html
Microsoft Office Templates blog announcing Vertex42 as a partner. https://learn.microsoft.com/en-us/archive/blogs/templates/free-financial-calculators-from-vertex42
Strategyzer, official Business Model Canvas download page (free, ungated). https://www.strategyzer.com/library/the-business-model-canvas
“The Story Behind the Business Model Canvas,” Entrepreneur & Innovation Exchange. https://eiexchange.com/content/the-story-behind-the-business-model-canvas
Aswath Damodaran’s free data, spreadsheets and lecture notes at NYU Stern. https://pages.stern.nyu.edu/~adamodar/
Dan Bricklin, “Patenting VisiCalc,” in his own words. https://www.bricklin.com/patenting.htm
“Dan Bricklin invented the spreadsheet,” Quartz. https://qz.com/578661/dan-bricklin-invented-the-spreadsheet-but-dont-hold-that-against-him

Discover more from 1000 Startups
Subscribe to get the latest posts sent to your email.
Claude Penland builds the marketing and business strategy for companies that are good at what they do and hard to find. Thirty years operating, one exit, eight of them as a practicing casualty actuary.
The free two-page read is genuinely free. Email claude@1000startups.com and I'll send back what I can see from the outside. Or see the work samples and how to work with me.