I throw away every lead older than 48 hours

On 27 August 2026 I added one line to the filter that decides which hiring posts become leads for VALORAE Media.

Older than 48 hours: rejected. Before anything else is checked.

It runs first because it is the cheapest test and the one that saves the most wasted work. There is no point scoring a lead's budget, scope and fit if the job was filled on Tuesday.

And on its own, the rule would have done nothing at all, for a reason I only found while wiring it in.

Step 1: we were stamping the wrong date

Our leads come from people publicly asking for an editor or someone to run their channels. A script finds the post, a filter scores it, and a tracker holds the survivors.

The tracker had a column for the post date.

It was being filled with the date we found the post. Hardcoded to today.

So every lead looked fresh on the day it arrived, whether the post was two hours old or three weeks old. A freshness rule reading that column would pass everything.

That is the complication with any time-based rule in a pipeline. The rule is only as good as the timestamp, and the timestamp is the field people treat as a formality.

We fixed it the same day, so the tracker records when the post went up. The filter now reads relative wording like "3h", "2 hours ago" or "yesterday" as well as proper timestamps, because that is what most pages show.

Step 2: a blank date switched the rule off

The second decision was harder.

Some sources hide when something was posted. If the filter rejected every undated lead, we would lose real ones for a reason that has nothing to do with the person hiring.

So an unknown age gets flagged, not rejected.

The cost of that choice is ugly. Anyone who calls the filter without passing a post date silently turns the freshness rule off, and nothing complains. That warning now sits in writing next to the filter, because it is exactly the kind of thing that breaks quietly six weeks from now.

Speed beats cleverness here.

Step 3: the cutoff comes from how hiring posts age

Why 48 hours?

I will be straight about this: it is a judgement, not a number I measured. I did not run an experiment across cutoffs and pick the best one.

What I did was look at how hiring threads age. By the second day, most are either filled, deep in replies from a crowd of other editors, or the person has moved to talking privately with the three they liked. Your message lands at the bottom of a pile the reader has stopped opening.

A day felt too tight, because posts found on a weekend would die before anyone read them on Monday. Three days felt like a polite way to keep sending messages into filled jobs.

So two days.

Inside the window, fresher still wins. A post under six hours old scores higher than one at a day and a half. The cutoff is a floor.

Step 4: borrow the research, but read what it measured

The study everyone quotes on this is from Harvard Business Review, March 2011, "The Short Life of Online Sales Leads" by Oldroyd, McElheran and Elkington.

They audited 2,241 companies by sending each a web-generated test lead, and the average response time among those who replied within 30 days was 42 hours, per the article.

In a separate study of 1.25 million leads, the same HBR article reports that firms contacting a lead within an hour were nearly seven times as likely to qualify it as firms that waited even an hour longer, and more than 60 times as likely as those that waited 24 hours or more.

That is a huge effect. But read what it measured.

Those were inbound leads. People who contacted the company. A hiring post is someone else's public request that I answer without being asked.

So I take the direction and not the multiplier. Speed matters enormously when someone is actively looking. The exact size of that effect on cold replies to public posts has not been measured by anyone I can cite, and I am not going to invent it.

Step 5: work that has not started yet is also stale

A sibling rule came out of the same week.

Some posts are fresh by date and stale by substance. "Editing will not begin until next year." "Looking to hire once the channel is monetised."

We reject those too. A lead that cannot pay for months is a lead we will have forgotten by the time it can, and it clogs the tracker with names that look active.

The freshness question is really "can this turn into work soon?", and the date is only the easiest part of that to check.

Step 6: the rule is for strangers

This whole rule applies to cold hiring posts from people who do not know us.

It does not apply to anyone we have spoken to. A past client who posts that they need help again is worth a message whatever the date. A relationship does not expire in two days.

It also does not replace answering fast when someone contacts us directly. That part is the one the HBR numbers were measuring, and it is where the rule matters most.

VALORAE Media treats an inbound message as a sales conversation, not a support ticket. The reasoning is laid out in your DMs are sales calls, and it is the same principle from the other side: whoever replies while the person is still looking has the conversation.

Step 7: what I would tell another agency

If you run outbound off public posts, three things.

Store the date the post went up, and check the column today. It is probably wrong.

Decide what a missing date means, write it down, and accept the cost you chose.

And pick a cutoff you can defend in one sentence, even if the sentence is "it is a judgement." A rule you can explain is a rule the next person will keep.

The short version

• Drop cold hiring posts older than 48 hours before scoring anything else.
• Record when the post went up, never when you found it.
• Flag undated leads rather than rejecting them, and know that this switches the rule off.
• Use the HBR 2011 findings for direction, not for a forecast on cold replies.
• Treat work that starts months out as stale too.
• Never apply the cutoff to past clients or to people who contacted you.

More notes like this, on running agencies and what broke, are at tgsidd.com.