A business plan is read one section at a time, so the cost on one page and the revenue on another never sit side by side. The Business Model Canvas puts every part of a business on a single page, where two parts that disagree show up in the same glance.
A plan can argue every section well and still describe a business that cannot pay for itself. Webvan, the online grocer, went public in 1999 and planned to expand to 26 US markets. In 2000 it took in $178.5 million and spent $525.4 million, close to three dollars out for every dollar in, and it shut down in July 2001.
Each half of that business held up alone. People did want groceries brought to the door, and the warehouses and vans did get built. What failed was the relationship between the two halves, and a document read in order never makes anyone look at that.
The canvas exists so that every part of a business can be checked against every other part at once.
The canvas changes the format. Who you serve, what you promise, how you deliver and what it costs all sit on one sheet, so you read the business as a set of relationships between blocks instead of a sequence of chapters.
A friend hands you a thirty-page plan for a meal-delivery start-up. Every section is well argued. What is the most useful first check?
Put what one order costs to deliver next to what one order brings in
Check the market-size estimate against an outside figure
Re-read the operations section for anything that cannot be built
Ask for a one-paragraph summary of each section
A plan's sections can each be sound while the business fails between them. Setting what one order costs to deliver beside what one order brings in tests the relationship the plan's format keeps apart. Re-reading sections, checking the market size or asking for summaries all stay inside one section at a time.