Your company does not need more alerts. It needs relevance.

Why does a company with more alerts still miss what matters?

A company misses what matters because its alerts are sorted by keyword and recency, while relevance depends on the company's own context: what it sells, to whom, against whom, and what it is betting on. Adding alerts adds volume without adding that context, so the important item arrives looking exactly like the unimportant ones.

Most teams already have more coverage than they can read: Google Alerts on competitor names, a Slack channel fed by a monitoring vendor, newsletters forwarded by executives. The problem is rarely that a development went unpublished. It is that it was published, delivered, and lost among forty other items that matched the same words.

The fix is not another feed. It is a written statement of what the company needs to know, a test that applies it to every item, and fewer signals, each explained and sent to the person who can act.

Why does the same news matter differently to different companies?

The same news matters differently because relevance is a relationship, not a property of the item. A headline has no importance on its own; it acquires importance when it touches a company's products, customers, competitors or plans. The same sentence can be urgent for one company, an opportunity for a second and noise for a third.

Most monitoring is built as if this were false. A vendor feed scores items by how many outlets covered them, how often a keyword appears, or how recently they were published. Those scores are the same for every subscriber. Two companies in different positions receive the same ranking, which means at least one of them is receiving the wrong one.

What is wrong with generic competitive intelligence feeds?

Generic competitive intelligence feeds fail in three predictable ways: they match words instead of consequences, they broadcast instead of route, and they deliver items without explaining them. Together they produce a channel that everyone mutes.

• Matching words. A rule on a competitor's name catches their sponsorship of a conference and misses the pricing page change that never mentions the name in a headline.
• Broadcasting. Items land in a shared channel where sales, product and leadership all see them. When everyone sees an item, each person assumes someone else owns it, and the item is acknowledged with an emoji and left there.
• No explanation. A link and a headline force each reader to work out the implication alone. Busy people skip that step, so the item is read as news, not as a reason to act.

Health care has studied the end state of this pattern closely. The Agency for Healthcare Research and Quality describes alert fatigue as workers becoming desensitized to warnings and ignoring them, and names two causes: the volume of alerts and the low relevance of most of them. Business channels follow the same mechanism. Once most items can be ignored safely, people stop distinguishing the ones that cannot.

What should a company intelligence mandate contain?

A company intelligence mandate is a short written statement of the context a filter needs: what the company sells, who buys it, who it competes with, which strategic bets it has made, and who owns each topic. It is the company profile turned into a set of instructions for reading.

A useful mandate has six parts:

1. Products and the problems they solve, in the customer's words, not the internal names.
2. Customers and channels: the segments that make up most of revenue, and the partners or retailers the company depends on.
3. Competitors, split into the three to five you lose deals to and the wider set you only need to notice.
4. Strategic bets: the markets, technologies or pricing moves the company has committed to, each phrased as an assumption that news could confirm or break.
5. Owners: one person per topic who receives signals and decides what happens next.
6. Out of scope: the adjacent subjects the company will deliberately not track.

The idea has a long history in competitive intelligence. Jan Herring, a former CIA officer who later ran intelligence at Motorola, adapted it for companies as key intelligence topics, grouped into strategic decisions, early-warning topics and key players. The principle is the same: start from what decision makers need, then choose what to read.

Unlike a personal reading list, a company mandate has to work across people: its value lies in agreeing, in writing, who hears about what. Drafting it with sales, product and leadership keeps it from becoming one department's reading list.

What is a relevance test you can apply to any item?

A relevance test is a short set of questions that decides, for any item, whether it should be routed to an owner, filed for context, or dropped. Five questions are enough, asked in order, stopping at the first one that fails.

1. Which line of the mandate does it touch? Name the product, customer, competitor or bet. If no line applies, drop it.
2. What is new? Does it change something the company believed, or repeat what was already known? Repetition gets filed.
3. Which decision does it affect, and by when? If no decision is affected within the next planning cycle, file it.
4. Who owns that decision? The owner is the destination. If nobody owns it, the mandate has a gap worth fixing.
5. What would that owner do differently? If the honest answer is nothing, the item is context, not a signal.

An item that passes all five is routed, with the answers to questions three and five written as one sentence on top. That sentence is what turns a forwarded link into intelligence.

How does one news item matter to three different companies?

One news item can be urgent, an opening, or background depending on the company reading it. Consider a hypothetical announcement: a large national retailer says that, within eighteen months, every supplier must report the carbon footprint of each product it sells through the retailer's stores.

Company A: a packaged snacks maker that sells a large share of its volume through that retailer.

• Mandate line: a key customer and channel.
• What is new: a hard requirement with a deadline, not a voluntary program.
• Decision: whether to build product-level emissions data in house or buy it, and how to budget for it this year.
• Owner: head of operations, with the key account manager informed.
• Verdict: route now. Missing the deadline puts shelf space at risk.

Company B: a software vendor selling carbon accounting tools to mid-sized manufacturers.

• Mandate line: a strategic bet that supply chain reporting drives demand.
• What is new: a named buyer creating a deadline for thousands of suppliers at once.
• Decision: whether to build a campaign aimed at that retailer's supplier base, and whether the product covers product-level footprints well enough.
• Owner: head of marketing, with product informed.
• Verdict: route now. The window is the eighteen months before the deadline.

Company C: a regional grocery chain that competes with the retailer.

• Mandate line: a competitor, but on store experience and price, not supplier policy.
• What is new: little that affects the chain directly in the near term.
• Decision: none within the planning cycle, though shared suppliers may pass costs on later.
• Owner: none needed today.
• Verdict: file it, and note a single follow-up question for the quarterly review: are shared suppliers raising wholesale prices?

The headline was identical in all three inboxes. The value came entirely from the context each company brought to it.

How should signals reach the right owner?

Signals should reach owners directly, one at a time when urgent and in a weekly batch otherwise, each with a reason attached. The goal is fewer items per person, every one of which earns its place.

A few operating rules make routing work:

• Route by owner, not by topic. A channel per person or function beats a channel per keyword, because ownership is what produces action.
• Attach the why. Every routed item carries one sentence: what changed, and why it matters to this company.
• Set a volume budget. Decide how many urgent items each owner should receive in a normal month. If the budget is exceeded every month, the filter is too loose.
• Close the loop. Once a month, ask each owner which items they acted on. Rules that produce nothing actionable get rewritten.

Filed items still matter: they become the record a team consults when a question comes up later.

What are the risks of filtering this hard?

The main risk of a strict relevance filter is missing something important that falls outside the mandate. A company that writes down only what it already knows to watch can be surprised by what it did not think to write down.

Three habits reduce that risk without reopening the floodgates. Keep one small, explicitly peripheral topic for adjacent developments, reviewed monthly rather than daily. Treat every surprise as a mandate defect and fix the line that would have caught it. And review the mandate every quarter, because strategy changes and a filter tuned to last year's company will confidently ignore this year's competitors.

Where does software fit in company-level industry monitoring?

Software fits after the mandate is written, as the layer that applies it to more sources than a team could read. Industry monitoring at company scale means checking competitors' sites, regulators, trade press, filings and specialist voices every day, and a person cannot run the relevance test on all of that by hand.

The mandate translates into configuration: competitors and customers become sources, bets become questions, owners become destinations. Kindal is one system built around that order: you choose the sources, it reads them and writes a short brief when something changes, saying why the change matters and linking each line to its source, and staying silent when nothing moved. Teams focused on competitor tracking or traffic data can compare market intelligence options for small teams.

Whatever tool runs the reading, the mandate stays a company document, owned and edited by people.

What changes when relevance comes first?

When relevance comes first, the number of items each person receives drops, and the share they act on rises. The work moves from reading everything to deciding, once and in writing, what the company needs to know.

That is the whole argument. More alerts give a company more of the same headlines everyone else receives. A mandate, a relevance test and named owners give it the few items that are about its own products, customers, competitors and plans, explained, and delivered to the person who can do something about them.