Most businesses find out the same way. A customer, mid-conversation, says the other quote was lower — or mentions a service they assumed you also offered. The information arrives in the worst possible setting: in front of the person you were trying to sell to, with no time to check whether it is even true.
The instinctive fix is to check competitors' websites more often. That mostly does not work, and it is worth understanding why before spending money on it.
The change you need rarely appears on the homepage
Homepages are marketing. They change when a business rebrands, which is roughly never, and they are the last place a company reflects an operational decision.
The signals that actually matter sit in less glamorous places:
- A line on a specification or price sheet, often a PDF, often not linked from the main navigation.
- A distributor or marketplace listing, which frequently moves before the manufacturer's own site does.
- A job posting. The most underrated signal there is. A competitor advertising for three installers in a city they do not currently serve has told you their expansion plan months before any customer-facing page will. Same for a first-ever quality or compliance hire, which usually means they are chasing a class of contract they could not previously bid for.
- Trade press and association notices, where capacity expansions, certifications, and supply changes surface first.
- Review platforms, where a sudden cluster of complaints about lead times is a competitor in trouble — and an opening.
None of this requires anything clever. It requires knowing where to look, which is a different problem from looking more often.
Why most monitoring gets abandoned in three weeks
This is the failure that repeats, and it is not a technology failure.
Compare any page against itself daily and you will find differences daily. A testimonial rotates. A copyright year ticks over. A promotional banner appears for a weekend. A phrase gets reworded. Every one of those is a change, and none of them mean anything.
Report them all and you produce a document nobody reads by week three. Report none and you miss the price move. The entire difficulty sits in the middle, and it has to be settled before any collection starts:
What counts as a change worth knowing?
That question is answered per source, not in general. On a pricing page it might be a published figure moving by more than 3%, and nothing else. On a services page it might be a new named service or a new city, ignoring all copy edits. On a competitor's careers page it might be any posting at all, because in that context every one carries information.
Get that definition right and a weekly briefing runs to five lines and gets read. Get it wrong and you have built something that generates work instead of removing it.
The question that decides what to watch
Before adding anything to a watch list, answer this:
If this changes, what will I do differently?
If there is no answer, do not watch it. Competitor headcount is interesting; unless you are recruiting against them, it changes nothing you do. Their social media following is a number that goes up.
Whereas: if a directly comparable product drops below our price by more than 5%, we review our published price and brief the sales conversation. That is a decision with a trigger attached, and it justifies the watching.
If you cannot say what you would do differently, you are not monitoring — you are subscribing to anxiety.
Applied honestly, this test usually cuts a wish-list of thirty things down to about eight. That reduction is the main deliverable, and it happens before anything is built.
What is legitimate to watch
Worth being clear about, because the question comes up and the answer is more straightforward than people fear.
Reading public information about companies you compete with is ordinary commercial practice. It is precisely what your customers do when they compare you, and there is nothing furtive about doing it deliberately.
The boundaries are simple. Respect the terms of the sites involved. Stay out of anything behind a login or a paywall. Do not collect personal data about individuals — this is about companies and their public offers, and a monitoring brief that starts drifting toward named people has stopped being competitive intelligence.
Public business information, collected respectfully, is also entirely sufficient. In practice, the constraint is almost never what you are permitted to see; it is that nobody has decided what is worth seeing.
Cadence, and the one thing worth interrupting you for
Weekly suits nearly every business. Long enough that the briefing has something in it, short enough that you are not learning about a month-old price change.
Reserve immediate alerts for a very short list — realistically, price movements on directly competing offers, and a competitor entering your service area. Everything else can wait until Monday, and treating everything as urgent is how the urgent things stop being noticed.
Two details separate a briefing that gets used from one that gets skimmed:
Every line links to its source, with the date it changed. Without that, you are asking someone to act on an assertion, and the first time a line is wrong the whole thing loses credibility. With it, verification takes eight seconds.
Someone decided what mattered before it was sent. Automated collection can gather and summarise far more than a person would ever read, which is exactly what it should do. But a briefing that faithfully reports everything a machine noticed has handed the judgement back to you, which was the work you were trying to hand over.
Where to start if you are starting from nothing
Pick your three most direct competitors — the ones you actually lose work to, not the biggest names in the industry. For each, write down the two or three pages where a change would genuinely alter what you do. Add their careers page.
Check them yourself, once a week, for a month, and write down what you find. Most people discover one of two things: either almost nothing changes, in which case you have saved yourself a project and can stop; or several things changed that you did not know about, in which case you now know exactly what to watch and why.
Either outcome is worth a month of Mondays.
If you do find it worth continuing, the honest next question is whether it should stay manual. The test is the same one that applies to any repeating task: frequency and consistency, not how tedious it feels.
Our Monitoring Blueprint is $599: what to watch and what to deliberately ignore, a source-by-source registry with a note on whether each is practically and legitimately collectable, a written definition of what counts as a change worth reporting, one sample briefing built from your real market, and a fixed quote to run it. Fully credited against the build.
Every finding links to its source with a date, and the data is yours to export at any point. We do not promise to catch everything — nobody can, and a service that claims otherwise is telling you something about its honesty rather than its coverage.