Intelligence Reports · Guide
Most buying signals are noise
Every list of B2B buying signals tells you what to watch. Almost none tell you what to throw away. That is the harder half, and it is the half that decides whether a monitoring program is still being read in three months.
Written from running weekly opportunity monitoring for organizations from lean teams to a Fortune Global 500 enterprise. The worked examples are invented; the failure mode is not.
These programs die of volume, not of blindness
The common story is not that a team missed the signals. It is that the team pointed something at eight categories, it worked exactly as promised, and it returned two hundred items a week. Nobody reads item forty. Within a month the digest has a folder rule and an unread count.
A signal you do not act on is worse than one you never received. You paid for it, and it taught the team that the report is safe to skip — which means the one item that mattered in March gets skipped too.
So the question worth asking is not “what else could we be watching?” It is “what are we prepared to drop, so that what is left gets read?”
Four questions that kill most of them
Run anything you are currently tracking through these. Most categories fail at least one, and a category that fails one is a category you are reading out of habit.
Can you name the role you would approach and a reason that person would take the call this month? If the honest answer is “we would put them on a list,” it is not a signal. It is news about a company you find interesting.
What this removes: most market-level developments. A sector growing 9 percent is a fact. It is not a conversation.
This one cuts both ways and is the test teams most often skip. A signal that buys eighteen months of warning is wasted on a team that works a three-week cycle — nobody will still be holding it. A signal that gives you nine days is wasted on a team whose deals take nine months, because you cannot mobilize in time.
What this removes: whole categories that are genuinely valuable to somebody else. Fit is specific to how you sell, not to how important the event is.
Write the opening line you would send having seen the signal. Now write the one you would have sent anyway. If they are the same sentence, the signal did no work — it only supplied a pretext, and buyers have read that pretext before.
What this removes: the congratulatory opener. “Saw your funding round — thought I’d reach out” is the same email as no signal at all, with a timestamp on it.
Organizations do interesting things continuously. Only a fraction of those things move a budget, create a deadline, or empty a seat a supplier was sitting in. The rest is corporate activity that reads like opportunity because it arrived in a feed labeled opportunities.
What this removes: more than you expect. This is usually the test that takes a list of eight categories down to three.
Four that look strong and usually are not
These appear on nearly every list of signals worth tracking. They are not worthless. They are just far weaker than their popularity implies, and they are expensive in attention.
Money raised is not money allocated. The gap between the announcement and a purchase order is usually two or three quarters, and the announcement reaches every competitor you have on the same morning. What is worth watching is not the round; it is what the company does in the two months after it — the hires, the site, the first contract.
Popular because they are easy to collect, which is a reason for a vendor to offer them and not a reason for you to act on them. A posting says a team is short-handed. Hiring frequently replaces a purchase rather than preceding one — the company decided to do it in-house. Read alongside a capital commitment it means something. Alone it rarely does.
Published precisely because somebody wanted them published. There is no deadline attached, no seat has come free, and no budget has moved. These fail the money test outright, and they crowd a digest because there are so many of them.
Not false — just already priced in. If the item reaches you through a portal or newsletter your three closest competitors also read, you are not early, you are on time. Being on time is the same as being late in a market with one buyer and four bidders.
What survives the cut
Three shapes tend to pass all four tests, for most businesses that sell to other businesses.
Something with a date attached. A rule with a compliance deadline, a contract with an expiry, a framework that has to be rebid. These are the only signals that tell you when the money arrives rather than leaving you to guess.
Something that empties a seat. A supplier failing, a merger collapsing two approved lists into one, a long-standing incumbent losing a certification. Somebody is about to be replaced, and it is worth knowing whether that is an opening or a warning about your own account.
Something that creates a site. A plant, depot, office or market entry. Physical commitments generate long, predictable chains of purchasing, and a company arriving somewhere new has no incumbent to displace.
A week, filtered
Invented but ordinary. A mid-size industrial supplier turns on broad monitoring across its sector and two neighboring ones. The first week returns roughly a hundred and eighty items.
Around ninety are company activity with no money attached — appointments at firms they do not sell to, awards, product announcements, conference appearances. Gone on test four.
Around fifty are real developments at real accounts, but nothing they could open a conversation about this quarter. Gone on tests one and two. Some are worth a note in the account file; none are worth a line in a weekly report.
That leaves roughly forty. Of those, most are known already — the account manager saw it, or it came through the same trade title everyone reads. Gone on test three.
Six to nine items remain. That is a report somebody reads on a Monday morning, and it is the same report whether the feed behind it returned a hundred and eighty items or twenty thousand. The value was never in the collection. It was in being willing to throw away a hundred and seventy.
How to make the cut yourself
- Take last quarter’s won deals and work backwards. What was observably true about each buyer before they contacted you? Three or four patterns will repeat. Those are your categories, and they are probably not the ones on a vendor’s list.
- Put a lead time against each one and compare it to how long your deals actually take. Drop anything that does not fit at either end.
- Set a ceiling on the report before you set the sources. Ten items, or five. A fixed ceiling forces a ranking; an open-ended digest never has to choose, which is why it grows until nobody opens it.
- Name who acts, per category, before the first item lands. A signal routed to nobody costs more than a signal missed, because you paid for it twice — once to find it and once in the credibility of the next report.
The discipline is the deliverable. Collecting more is cheap now and getting cheaper; deciding what not to send is the part that still takes judgment.
Related guides
Report designWhat a competitor monitoring report should actually containThe structure, a worked example, and why the hard part is deciding what to leave out.Read the guide →SourcesWhere market signals actually appearWhich signals show up where — and what monitoring in English only costs you.Read the guide →CadenceWhat to monitor, and how often each one movesThe eight categories worth tracking, and why only two of them reward watching daily.Read the guide →The hard part is what we leave out.
We read broadly and report narrowly — a short weekly brief that names the opportunity, why it matters to you, and the first move. Six to nine items, not two hundred.
Request a free sample