Every platform in this category leads with a number. Sources monitored, signals detected, data points tracked. The numbers are large and they are usually true.
They are also close to meaningless as a purchase criterion, and a prospect made the point better than we would have. Reviewing a demo, he observed that most of what he had been shown was irrelevant to his business, and that pulling lots of data is table stakes while relevant data is the actual bar.
He was right, and it is worth taking seriously, because the failure mode he was describing is the most common way these tools end up unused.
Why volume is the wrong metric
A signal platform has two jobs: find things, and decide which of them matter. The first is largely solved. Data is abundant, APIs are plentiful, and any competent team can ingest a great deal quickly.
The second is where the difficulty lives, and it barely appears in marketing material because it does not reduce to a number.
The asymmetry matters because the costs are asymmetric. A missed signal costs you one opportunity. A feed full of irrelevant signals costs you the rep, because they stop opening it. Once a rep concludes a tool is noise, it does not matter what else it detects, and that judgement is usually formed inside the first two weeks.
More is not better past the point where a human can read it. A rep will look at a feed for a few minutes a day. Everything past what fits in that window is not intelligence, it is a backlog.
See Salesmotion in action
Take a self-guided interactive tour — no signup required.
Three ways relevance fails
Volume problems present in different ways, and the fix differs for each.
Wrong subject. The signals are about your accounts but concern things you cannot sell into. Usually a keyword configuration problem, and the most fixable of the three.
Wrong weight. The signals are relevant but ranked badly, so genuinely important events sit below trivial ones. A commercial leader at a large scientific instruments company described this precisely: he worried that ten articles about a single acquisition would push a low-value account above accounts that mattered more, purely on volume of coverage. That is a scoring design question, and it is worth asking a vendor directly.
Wrong timing. The signals are relevant and well ranked but arrive after the moment has passed. A director at a software development firm told us job-change alerts in his existing stack were firing around three weeks after the move, by which point the value had largely evaporated. Freshness is a relevance problem, not a separate category.
Why volume beats relevance in a demo
There is a structural reason this keeps happening, and being aware of it helps you resist it.
Volume demos well. A feed filling with activity looks like a product working. Relevance demos poorly, because the correct behavior is often an empty feed. If nothing meaningful happened at your accounts this week, the honest output is "nothing meaningful happened," and that is a difficult thing to show in a sales meeting.
So the incentive across the category runs toward showing more. Buyers reward the busy-looking demo, and then discover in month two that busy is the problem.
This also explains a pattern in tool abandonment. Teams do not usually abandon these tools because the data was wrong. They abandon them because the ratio of useful to useless fell below the threshold where checking was worth the time.
Evaluating relevance in a demo
Five questions that separate a scoring system from a firehose. Ask them live.
"Show me a week where nothing important happened." The most revealing question available. A platform that can say "these three accounts moved, the other hundred and ninety did not" is making judgements. One that always shows a full feed is not.
"Why is this account ranked above that one?" You want an explanation in terms of what happened and why it matters, not a score with no derivation. If nobody can explain the ranking, your reps will not trust it.
"What happens when one event generates twenty articles?" This tests the weight problem directly. Good systems recognize one event covered many times as one event.
"How do I tell it this was irrelevant?" If there is no path for feedback, relevance can never improve for your specific business, and generic relevance is the thing that failed.
"How quickly did this specific signal appear after the news?" Pick a development you already know about and check the lag. Use your own knowledge as the benchmark.
The configuration half
An uncomfortable truth: a meaningful share of relevance problems are configuration, not product.
Keywords written to describe what you sell rather than what triggers a purchase will produce irrelevant results in any tool. Priorities set uniformly high leave the scoring nothing to work with. Keyword sets shared across teams with different motions serve neither.
This is not a deflection, because it has a real implication for evaluation. Ask what the tuning process looks like, who does it, and how long it takes to see improvement. A platform with no tuning path is one where relevance is whatever it shipped with. A platform that requires an engineer to tune is one that will never be tuned.
Our own view is that keyword setup is the highest-leverage hour in onboarding, which is why we walk through it in a worked example rather than leaving it to a settings page.
What good looks like
A signal system is working when a rep opens it and closes it in ninety seconds with a clear answer about who to contact and why. Not when it has the most sources.
Three properties that produce that:
It can be quiet. Silence when nothing happened is a feature, and a system incapable of silence is not ranking.
Rankings are explainable. A rep can see why an account surfaced and judge whether they agree. Trust follows from being able to check.
It gets better. Feedback changes future output. Static relevance decays as your market moves.
Buy on those, and treat source counts as a floor rather than a differentiator. Every serious platform has enough sources. Few have enough judgement.
For how refresh cadence affects the timing dimension, see how fresh is your sales signal data. For what happens when the ratio tips the wrong way, see bad leads kill tool adoption.
