Someone else is calling the lead you paid for

A buyer who fills out your form on Tuesday morning is not waiting for you. They filled out two other forms the same hour. That's how buyers evaluate now. Whichever vendor gets a competent human into the conversation first frames the problem, sets the criteria, and becomes the standard the others get compared to. Inbound intent is perishable, and it's measured in days, not weeks.

Now the CRM question: what is your median time from lead creation to first logged activity? Not the SLA you announced, but the median your portal would actually report if you filtered for it. Most leaders have never run the filter. Most who do don't enjoy the result.

Where the gap comes from

The first-touch gap rarely has one cause. It's five small mechanisms stacking:

• Routing rules built for an org chart that no longer exists send leads to the wrong desk, where they sit in the wrong queue until someone notices, which is never.
• Handoffs between marketing and sales each assume the other side is watching. Marketing counts the MQL and moves on; sales assumes anything important comes with a Slack ping. The lead that arrives quietly belongs to nobody.
• Reps triage visible urgency: the demo request gets called, the whitepaper download waits forever. Reasonable per-lead, expensive in aggregate, because buying signals are noisier than form types.
• A lead assigned isn't a lead worked, but dashboards count assignment as motion. The metric that looks healthy (time to assignment) hides the one that matters (time to touch).
• After-hours and Friday-afternoon leads age three days before business resumes, and by Monday they've been buried by the weekend's new arrivals.

Each mechanism drops a few; together they drop a pile. And a lead that misses its first week rarely gets touched in its second. The decay is steep, and then it is total.

The two numbers to put on a wall

First: median hours from creation to first logged activity, by source. The median, not the average; one heroic five-minute response hides a lot of nine-day silences in an average. Cut it by source because sources decay differently: a demo request is a hand-raise with a clock on it, while a content download can tolerate a slower touch. If you only track the blended number, the demo requests subsidize the neglect of everything else.

Second: the count of inbound records older than 30 days with zero logged activity, which is the graveyard the median leaves behind. The first number tells you how the system is performing now; the second tells you what its history has already cost. (The exact filter recipe for that graveyard is filter one in the four places revenue leaks out of a HubSpot portal.)

Building the filters in HubSpot

Both take minutes. For the median: a report on contacts, "create date" against "date of first activity," grouped by original source, shown as a median. For the graveyard: a saved contact filter, create date more than 30 days ago, last activity date is unknown, lifecycle stage is lead or beyond. Save both to the dashboard your revenue team actually opens, not the one built for board meetings.

Making it hold

Set the standard where intent actually lives: same business day for hand-raisers, 48 hours for everything else. Then make the miss visible: a saved filter of unworked inbound, reviewed weekly by name. Not to shame anyone; to make silence show up somewhere before it becomes permanent.

Three implementation notes from portals that made it stick:

• Route by capacity, not just territory. A rep at 130% of pipeline capacity is a black hole for new leads no matter what the routing rule says. If someone's untouched-lead count keeps growing, the fix is redistribution, not reminders.
• Put the SLA clock on the record. A task with a due date fires a visible overdue state; an SLA that lives in a slide deck fires nothing.
• Watch the miss rate, not just the median. A team can hold a two-hour median while 15% of leads fall through entirely. The median is the brochure; the miss rate is the leak.

The teams that hold a first-touch standard don't have better reps. They have a system where an untouched lead is loud instead of quiet. The structural reasons reps go quiet on leads, and why it isn't laziness, are covered in why good reps let leads go quiet.

The pile the standard leaves behind

The leads already past the standard, the ones sitting at zero activity from last quarter, aren't automatically dead. Buyers' problems persist longer than their patience. A late, honest touch ("your note reached us at the worst possible week; if the problem's still open, I'd like to fix that") recovers more of them than pride predicts, because almost nobody else bothers. But the recovery only happens if someone counts the pile first, and counting it is exactly what a record-level audit does: every zero-touch lead named, valued, and ranked by recoverability.