XP OneWhat is signal-based prospecting
Traditional B2B prospecting works from a static list. You filter the database (ICP fit), upload to your sequencer, send. Signal-based prospecting inverts this: you watch the world for events, and only when an event fires do you trigger outreach to the specific person at the specific company at the specific moment. The static list is replaced by a live event stream.
Why it works: at any given moment, ~3% of your TAM has an active need. The rest don't. Signal-based prospecting talks to the 3% — list-based prospecting talks to the 100% and hopes.
The 14 signals that work in 2026
| # | Signal | Avg reply uplift | Detection difficulty |
|---|---|---|---|
| 1 | Job change to buyer role at ICP account | 7.2x | Easy (LinkedIn) |
| 2 | Funding round (Seed-Series C) | 5.8x | Easy (Crunchbase, news) |
| 3 | Pricing page / demo page visit | 6.5x | Medium (intent tools) |
| 4 | Hiring posts for relevant role | ~5x | Easy (LinkedIn jobs) |
| 5 | Engaged with competitor's content | 5.1x | Medium |
| 6 | Tech-stack change detected | 4.6x | Medium (BuiltWith, Wappalyzer) |
| 7 | Leadership change (CEO/CMO/CRO) | 4.3x | Easy |
| 8 | Product launch / new feature | 3.9x | Easy |
| 9 | Engagement with your content | 4.2x | Easy |
| 10 | Negative review of current vendor | 5.5x | Hard (G2, Reddit monitoring) |
| 11 | Office expansion / new location | 3.5x | Easy |
| 12 | Partnership announcement | 3.2x | Easy |
| 13 | Trial/free-tier signup at your tool | 8.4x | Easy (your own data) |
| 14 | RFP / RFI posted publicly | 6.8x | Hard (gov + bid sites) |
Uplift figures are directional, measured on XP One accounts, May 2026, against the same message sent to a list with no signal.
Where XP One sits in this — and where it does not
Eight of these signals live on LinkedIn, and that is exactly where XP One collects. Its Chrome extension pulls prospects from reactions, comments, reposts, connections, followers, event attendees, direct profiles and search results — the people who just did something, with the date attached — then finds their email and phone at 80%+ accuracy and drops them into a real-time list in the native CRM.
What it does not do, today: watch Crunchbase, monitor G2, de-anonymise your website traffic, or detect signals for you while you sleep. Automated signal detection and ICP scoring are on the roadmap, clearly labelled as such. Collection is deliberate, on-demand, and capped at 100 profiles per day. Anyone selling you autonomous signal detection in 2026 is selling you a demo.
The observed conversion scale across those LinkedIn sources is the most useful table in this article:
| Signal the prospect emitted | Observed conversion band |
|---|---|
| Like on a post | 1–3% |
| Comment or repost | 3–8% |
| Event attendee or new follower | 5–15% |
| Targeted search result or direct profile | 10–25% |
| Post asking for a tool in your category | 15–30% |
Measured on XP One accounts, May 2026.
The 6 signals that are noise
- Profile views alone — Too noisy; viewing your profile rarely means intent
- Likes on competitor posts — Polite engagement, not intent
- Generic "intent data" from third-party providers — Most signals are stale or wrong account
- Twitter/X mentions of category terms — Too broad, too late
- Birthday / work anniversary — 2010s tactic, now read as cringe
- Webinar registration without attendance — Email harvesting, not intent
Priority scoring framework
When multiple signals fire on the same prospect, score and prioritize:
| Component | Weight | Example |
|---|---|---|
| Signal strength | 40% | Job change = 10, partnership = 3 |
| ICP fit | 30% | Exact ICP = 10, adjacent = 5, off = 0 |
| Signal recency | 20% | <24h = 10, <7d = 6, >30d = 0 |
| Account size / value | 10% | $10k+ ACV = 10, $1k = 3 |
Score >7.5: contact within 4 hours. Score 5–7.5: within 24 hours. Score 3–5: weekly batch. Below 3: ignore.
How to detect signals
| Signal type | Best source |
|---|---|
| Job changes | LinkedIn Sales Navigator alerts, Champify |
| Funding | Crunchbase, Pitchbook, Pulley |
| Hiring | LinkedIn Jobs, OnLoop, RolePoint |
| Website visits | RB2B, Common Room, Default, Vector |
| Tech stack | BuiltWith, Wappalyzer, HG Insights |
| Content engagement | Common Room, LinkedIn Sales Navigator |
| Reviews | G2 alerts, manual Reddit monitoring |
| News / announcements | Google Alerts, Meltwater, AlphaSense |
For the LinkedIn half of that table — engagement, event attendance, followers, job-change chatter — XP One collects the people directly and enriches them in the same pass, so the gap between "spotted the signal" and "have a verified email and phone" is minutes rather than an afternoon of copy-paste. For funding, tech-stack and review signals, you still need the sources above. See also LinkedIn prospecting in 2026.
Speed: the 24-hour rule
The half-life of a B2B buying signal is roughly 36 hours. Reply rates by time-to-touch:
- 0–4 hours: 100% baseline reply rate
- 4–24 hours: 85% of baseline
- 24–48 hours: 50% of baseline
- 2–7 days: 20% of baseline
- 7+ days: cold list-blast equivalent (~0.3%)
This is the real argument for tooling: not that software has better judgement than you, but that the window closes while you are in a meeting. Cutting the distance between "they commented" and "they have a message from me, with their verified number already on file" is worth more than another hour spent on the wording. What happens next is in the multichannel sequence playbook.
The fastest message to a signal wins. The most personalized message to a stale list loses.
FAQ
Do I need an intent-data tool?
Optional. The top 6 signals (job change, funding, hiring, leadership, content engagement, product launch) can all be sourced free from LinkedIn + Crunchbase + Google Alerts. Intent data is useful for website visits and tech-stack changes.
How many signals should I monitor?
Start with 3 — the ones most aligned with your buyer journey. Expand to 7–10 once your team can act on them within 24h. Beyond 10 without automation = signal overload.
What if a signal is wrong?
Acceptable false-positive rate is ~15%. Above that, refine the signal definition. Below that, you may be filtering too tightly.
Key takeaways
- Signal-first beats list-first by a wide margin: 1–3% on a passive like, 15–30% on someone asking for a tool in your category
- 14 signals worth tracking; 6 to ignore
- Half-life of a B2B signal is short — act within 24 hours, ideally within 4
- Score on signal strength × ICP fit × recency × account value — by hand, deliberately
- Automated detection and ICP scoring do not exist in XP One today; LinkedIn signal collection and enrichment do
Measured on XP One accounts, May 2026.