Customers buy the product and adoption stays low

The contract is signed, accounts are created, and daily users keep working the old way: in spreadsheets, on paper or in the previous system.

In B2B software the buyer’s decision does not make users adopt the product. Track adoption before renewal using agreed task completion, usage and customer feedback measures.

Before attributing the problem to interface design, check missing functionality, integrations, permissions, data quality, training and implementation requirements. Use observed task failures and customer evidence to choose the intervention.

• The buyer and the daily users of B2B software are usually different people.
• Users keeping spreadsheets next to the product is a common signal that the interface limits the business.
• Exports and spreadsheets identify tasks to investigate; measure the proportion of task instances completed outside the product directly.
• Users stop during setup before they have seen the product work.
• If chat dominates usage, test whether structured workflows offer relevant value and whether users understand when to use them.

The product is slower than the workaround. A spreadsheet accepts any input immediately. If a screen takes longer than the spreadsheet for a daily task, users keep the spreadsheet.

The interface follows the system’s structure. Business users think in terms of their work, and menus organised around entities and configuration make each task a translation.

The first days were never designed. Setup without a working state, and learning features without history, give new users the least value when they are forming an opinion.

• Spreadsheets kept next to the product, often with formulas the product lacks
• Licences assigned and rarely used
• Data entered in bulk at the end of a shift or a week
• Users describe an AI product as a chat and ignore its structured modes
• Long-term users praise a feature that new users ignore
• Renewal conversations start with questions about value

• Make the daily task faster than the workaround. This usually means redesigning the screens where users spend their day, which takes longer than adding features.
• Bring the spreadsheet’s purpose into the product. The columns and formulas users added show what is missing. Adding all of them recreates the spreadsheet inside the product.
• Design the first days as a state. An explicit learning state and useful defaults give value sooner, and processing past history costs compute.
• Route users from chat into a better workflow. It adds a step to the fastest path, and it is justified only if the workflow output is clearly better.

• Treating the parallel spreadsheet as user resistance
• Measuring adoption by logins instead of by work done in the product
• Adding training where the screen needs redesigning
• Leaving the first days of a learning feature undesigned

Where we’ve worked on this

• StartProto: manufacturing ERP and MES