How to Find Out What Your Competitors Are Doing — 7 Signal Channels for Manufacturers

Most manufacturers do not lack information about their competitors. They lack the right signal channels, opened in the right order.

"Just research harder" is not a method

When a competitor launches a product, enters a region, or changes a price, the news reaches most manufacturers through the same door: a lost deal, retold in a sales postmortem. By that point the move is typically three to six months old. The decision behind it is older still.

Subscribing to more news does not fix this, because the lag is structural and it compounds at every step. A capacity move, for example, travels through three distinct stages:

• Action. The competitor approves a capex line in an internal budget cycle. Nothing is visible from the outside.
• Visible trace. Equipment tenders go out, process engineers get hired, a tooling supplier receives an order. Only someone watching these channels sees anything.
• You can read it. A press release, an analyst note, a booth at a trade show. Now everyone sees it at once, including your other competitors.

Each arrow in that chain adds months. A search-based workflow sits at the last arrow and calls the result intelligence. It is not intelligence; it is the public record, read late.

The fix starts before any channel is opened. Write down the three questions you need answered earlier than the market answers them. In CI practice these are key intelligence topics, or KITs: "Is Competitor A adding capacity for product line X?", "Is Competitor B moving upmarket into our accounts?", "Which technology route is Competitor C committing to?" Channels earn their budget only if they answer a KIT. That is what gives the seven channels below their order.

The 7 signal channels, and what each one actually tells you

Each channel below comes with its typical signal, its lag relative to the competitor's decision, the cost to stand it up, and the time to first usable signal. None of them requires exotic tooling. All of them require rhythm.

• 1. Public disclosures. Annual reports, prospectuses, capex guidance, segment revenue, customer concentration, and expansion announcements. For listed competitors this is the cheapest lead time available: capex language and segment shifts telegraph moves quarters ahead. Cost is low. Expect a first usable reading within 30 to 60 days, once you have a baseline to compare against.
• 2. Hiring signals. The type of role matters more than the count. A cluster of process and automation engineers means a line is being built or converted; regional sales hires mean a market entry; quality-systems hires mean a certification push into a new customer segment. Hiring traces appear months before announcements. Low cost, first signal in 30 to 60 days.
• 3. Patents and standards. Filing is not publication: applications typically surface around 18 months after filing, so the visible record is a delayed view of R&D intent. Read patent families, citations, and claim scope rather than raw counts, and watch who sits on the standards committees in your category. Medium cost, 60 to 90 days to first signal.
• 4. Supply chain and procurement. Suppliers reveal their customers through their own behavior. Tooling and mold orders, production-line equipment purchases, new certifications, capacity additions at shared suppliers, and shifts in logistics lanes all precede the announcement they serve. This channel leads by three to nine months but costs the most to build, because it runs on relationships. First signal in 60 to 90 days.
• 5. Trade shows and customer sites. Booth size and placement, what gets demoed versus last year, the customer mix at the stand, and the engineering questions visitors ask are all readable signals. This channel is event-driven and bursty: moderate cost per event, and the calendar sets your rhythm.
• 6. Policy and subsidies. Industry catalogs, technical-upgrade project lists, export-control changes, and rules-of-origin shifts often name who is expanding before any press release does. Policy lists are a preview of capex, not a reaction to it. Low cost; first signal in 30 to 60 days in markets that publish such lists.
• 7. Tenders and customs. Award records, the technical specifications inside tenders, unit prices, and changes in export destinations give you hard numbers where everything else gives you direction. Low to medium cost through data subscriptions; first usable signal in 60 to 90 days, once you know the baseline pattern.

Notice what the table does not contain: a channel that is simultaneously cheap, early, and continuous. That combination does not exist. Lead time is bought with money (supply chain), patience (patents), or rhythm (disclosures, hiring, policy). That is why the order you open them in matters more than the list itself.

The order to turn them on

Days 1 to 90: open two channels, not seven. For most manufacturers the right pair is public disclosures and hiring signals. They are cheap, every competitor leaves these traces whether it wants to or not, and they build the baseline that every later channel gets compared against. If your competitors are private companies with thin disclosure, swap disclosures for policy and subsidy lists, which often carry the same information in industrial markets.

Months 4 to 6: add tenders and customs data, then supply chain. These two need plumbing: data subscriptions, a supplier-conversation routine, someone who owns the read. Opened in week one, they produce a spreadsheet nobody maintains. Opened in month four, against an existing baseline, they produce lead time.

Year two: complete the set with patents and standards and a systematic trade-show protocol. These are long-cycle channels whose value compounds over years, and they are the hardest for a competitor to notice you running.

The discipline that keeps this sequence honest is a single rule: a channel must answer at least one of your three KITs before it gets opened. If you cannot name the question a channel answers, you are not opening a channel. You are buying noise with a workflow attached.

What "good" looks like

A channel is working when it passes three tests. The signal is traceable to a source you can show a skeptic. The signal arrives ahead of the public news cycle, not inside it. And the channel produces every month, whether or not anything dramatic happened; a channel that only speaks during crises is a news feed with extra steps.

The failure modes are just as concrete:

• Digest without channels. A weekly email summarizing public news feels like a CI function and fails like one at the worst moment: everyone reads the same summary at the same time.
• Summaries treated as intelligence. A news abstract tells you what happened. A channel tells you what is about to happen. Confusing the two is how teams end up informed and surprised at once.
• A channel owned by one person, maintained by hand. If that person leaves or gets busy, the channel dies and nobody notices for a quarter. Document the rhythm, automate collection where possible, and make the read a role, not a favor.

The manufacturers who consistently know first are not smarter and do not spend the most. They opened two channels before they needed them, added two more on schedule, and refused to confuse a news subscription with an early-warning system.