According to Salesforce’s State of Sales 2026 report, the average B2B sales rep spends 40% of their workweek selling and 60% on admin work, data entry, internal meetings, and CRM upkeep. That’s roughly 24 hours every week that don’t put a rep in front of a buyer.
The worst part: after all that effort, the data still isn’t clean. Only 35% of sales professionals completely trust the accuracy of their CRM data, and 47% say data accuracy is harder now than it was a year ago.
The Cost of the Data-Entry Tax
The math is straightforward. A 25-rep team losing 12 hours per rep per week to admin work represents 300 weekly hours of capacity. That’s roughly seven full-time equivalents of selling time eliminated by data entry.
At the organizational level, Gartner estimates that poor data quality costs the average organization $12.9 million per year. That figure includes forecast inaccuracy, missed follow-ups, and the downstream cost of every system built on top of incomplete CRM records.
And the problem compounds. Every AI tool, every forecasting model, every pipeline report your team relies on is only as good as the data underneath it. When reps skip fields or log notes three days after a call, the entire intelligence stack degrades.
Why Reps Don’t Fill the Fields
According to Salesforce’s State of Sales research, 68% of reps say note-taking and data input are their most time-consuming tasks. 43% report that administrative work occupies between 10 and 20 hours of their workweek.
The reasons are structural, not motivational:
- CRM fields were designed for management reporting, not for sellers. The information a manager needs to forecast is different from the information a rep needs to close.
- The translation step loses information. A 30-minute discovery call contains dozens of signals. Reducing it to a picklist value and a two-sentence note discards most of the context.
- Timing kills accuracy. Reps update Salesforce at the end of the week, by which point details are lost and the data is already stale.
Interestingly, LinkedIn’s State of Sales report found that top-performing reps update their CRM 18% more often than average reps. The issue isn’t laziness. It’s that the field architecture forces a translation step that loses information regardless of how diligent the rep is.
The Point Solution Trap
Several tools have tried to solve this. Momentum (now part of Salesforce) automates CRM field updates from call data. People.ai captures activity data automatically. Scratchpad gives reps a faster interface for pipeline updates.
Each of these reduces friction in one part of the workflow. But none of them solves the underlying problem: the data that matters most — what was said on a call, what the buyer cares about, whether the champion is still engaged — lives in unstructured conversations, not in CRM fields. Automating the entry of structured data is useful. Reasoning across structured and unstructured data together is the step that changes the outcome.
What Fixes It
The fix isn’t a better data-entry interface. It’s removing the need for reps to be the data-entry layer in the first place. An intelligence layer like Von connects to your CRM, call recorder, email, and data warehouse, then builds a context graph of your business autonomously. It captures the signals that matter — from calls, emails, and deal activity — without requiring reps to translate them into picklist values.
The result: reps spend their time selling, and the intelligence stack gets better data than manual entry ever produced. Forecasts improve because they’re built on what was said in conversations, not what a rep remembered to log three days later.
For a deeper look at how this works, see What Is Revenue Intelligence? and Why Your Sales Forecast Is Wrong.
Frequently Asked Questions
How much time do sales reps spend on CRM data entry? According to Salesforce’s State of Sales 2026 report, the average sales rep spends 60% of their workweek on tasks other than selling, including data entry, internal meetings, and CRM administration. That’s roughly 24 hours per week. 68% of reps identify note-taking and data input as their most time-consuming tasks.
Why is CRM data still incomplete even when reps update it? CRM fields capture structured data (stage, close date, amount), but the most valuable deal signals live in unstructured conversations: what the buyer said on a call, whether the champion is engaged, what objections came up. Reducing a 30-minute call to a picklist value and a two-sentence note discards most of the context.
What is the cost of poor CRM data quality? Gartner estimates that poor data quality costs the average organization $12.9 million per year. For sales teams specifically, the costs include forecast inaccuracy, missed follow-ups, degraded AI model performance, and lost selling time when reps have to clean data their tooling should have captured.
Can AI automate CRM data entry? Partially. Tools like Momentum and People.ai automate the capture of structured activity data. But the higher-value problem is reasoning across unstructured data (calls, emails) and structured data (CRM fields, pipeline) together. An intelligence layer that does both eliminates the need for reps to be the translation layer between conversations and the CRM.
Previous in this series: Why Your Sales Forecast Is Wrong: The Real Causes
Next in this series: [The AI VP of RevOps: What This New Role Means]
