Why Your SDRs Spend 67% of Their Time Not Selling
Here is a number that should make every revenue leader uncomfortable: 67% of an SDR’s day is spent not selling.
Not prospecting. Not qualifying. Not closing. Just… working around broken systems.
I see it in almost every GTM audit I run. The reps are fine. The process is the problem.
Where the Time Actually Goes
Let me break it down based on what I typically find when I instrument a sales team’s workflow for a week.
| Activity | Share | What it looks like |
|---|---|---|
| Manual research | ~25% | Toggling between LinkedIn, the CRM, a company website, maybe a news aggregator to build a picture that should already exist in the system. Copy. Paste. Tab. Tab. Copy. Paste. Repeat for 40-60 accounts per week. |
| Tool-hopping | ~18% | Roughly 8 tools, each with its own login, its own UI, its own data model. None of them talk to each other natively, so the rep ferries data between them by hand. |
| Unstructured qualification | ~15% | Gut feel and tribal knowledge instead of scored, routed leads. “This one looks good.” “That company seems too small.” No shared framework, no signal-based prioritization. Just vibes. |
| CRM hygiene | ~9% | Updating fields, logging activities, filling in required fields that nobody reads. Not because the data matters in real time, but because a manager built a dashboard that depends on it. |
That tool count is not my estimate alone: Salesforce’s State of Sales research finds sellers use an average of 8 tools just to close deals, and the prospecting and qualification workflows I audit run in the same range. So the rep becomes the integration layer. A human API, moving data between systems that should be connected but aren’t.
Add it up: 67% of the day gone before a single meaningful conversation happens.
The Dollar Cost Nobody Tracks
The fully-loaded cost of an SDR in B2B SaaS runs $85-110K per year, depending on market and seniority. Public benchmarks reconstruct that number from two directions: RepVue’s verified comp data puts median SDR base at $60K and OTE at $85K, and the widely cited MIT rule of thumb adds 1.25 to 1.4 times base for payroll taxes and benefits, before tooling and management overhead even enter. If anything, $85-110K is conservative.
67% of that is $57-74K. Per rep. And the ratio is not unique to my audits: a Forbes-published time study found 64.8% of rep time going to non-revenue activities.
But that overstates it a little because some of that non-selling time is genuinely necessary. After automating what can be automated, the recoverable waste lands between $17-23K per rep per year. That is money being spent on work that a well-built system handles in milliseconds.
For a team of 8 SDRs, that is $136-184K annually. Not in software costs. In human time doing work that machines should do.
Why “Just Buy Another Tool” Does Not Fix This
The instinct is always to add another tool. A better enrichment provider. A new sequencing platform. An AI writing assistant.
But tools without infrastructure are just more tabs to switch between. I have walked into orgs with $200K+ in annual SaaS spend across their GTM stack, and their reps are still copying data between spreadsheets.
The problem is not a missing tool. The problem is missing infrastructure. No unified data model. No automated enrichment pipeline. No signal-based routing. No closed-loop feedback between what marketing generates and what sales actually converts.
What Infrastructure Looks Like in Practice
When I build a revenue system for a client, the SDR workflow changes fundamentally.
Automated enrichment pipelines pull firmographic, technographic, and intent data the moment a lead enters the system. By the time a rep sees the record, it already has company size, tech stack, recent funding, hiring signals, and competitive intelligence attached. No manual research needed.
Unified data architecture means every tool writes to and reads from the same source of truth. The CRM becomes the system of record, not a reporting afterthought. When a rep updates a deal stage, that change propagates everywhere it needs to go.
Signal-based qualification replaces gut feel with scored, weighted, and time-decayed signals. A lead that downloaded a pricing page, matches the ICP on 4 of 5 criteria, and works at a company showing hiring intent in the relevant department gets surfaced automatically. No scrolling through lists.
Activity capture eliminates manual logging. Emails, calls, meetings, and engagement data flow into the CRM automatically. Reps stop being data entry clerks.
What Infrastructure Actually Removes
Manual research. Tool-hopping. Unstructured qualification. CRM hygiene. Those four rows in the table above are the ones a well-built system takes off the rep’s plate, and together they are the bulk of the 67%.
What it cannot remove is the genuinely necessary overhead: internal meetings, training, strategic account planning. That is the floor, and it is a real one. The point of the build is to push the non-selling share down to that floor, not to zero.
The time that comes back goes straight to pipeline-generating activity. More conversations. Better-prepared conversations. Faster follow-up. Not because the reps got better, but because they finally have time to do the job they were hired for.
The Uncomfortable Truth
Most companies treat SDR productivity as a people problem. Hire better reps. Train them harder. Add more KPIs.
But when 67% of the day is consumed by system failures, no amount of training fixes it. You would not blame a carpenter for being slow if you handed them a butter knife.
The fix is not motivational. It is structural. Build the infrastructure that eliminates the non-selling work, and the selling takes care of itself.
If you are running a revenue team and haven’t audited where your reps’ time actually goes, start there. The number will be worse than you think. But at least then you will know what to build.