Comparison
Ballot Signal vs. IP-based targeting
IP-based approaches infer without asking. They are probabilistic, degrading, and sit squarely inside the pen-register litigation theory.
What IP targeting actually does
It takes the network address a request came from and infers something — a household, a building, an employer, a neighborhood — usually by matching against a commercial mapping of addresses to places. Then it targets ads at that inference, or claims to identify the person behind it.
Why it degrades
- Mobile traffic routes through carrier-scale address pools that map to nothing useful.
- Privacy relays and VPNs are now default-on for a large slice of consumer devices.
- Shared and dynamic addressing means the address you saw yesterday is somebody else in a different town today.
- CGNAT puts thousands of unrelated subscribers behind one address.
The accuracy claims in this category were never great and the underlying substrate is actively getting worse.
The legal posture is materially different
The pen-register theory under California Penal Code § 638.51 is specifically about scripts that collect IP addresses and similar routing information. A method whose entire mechanism is IP collection is not well positioned against that theory, and adding a privacy policy does not change the mechanism. See CIPA & website tracking.
Ballot Signal’s answer is structural rather than rhetorical: nothing runs before consent, so there is no pre-consent collection to characterize.
The honest comparison
| IP targeting | Ballot Signal | |
|---|---|---|
| Asks the visitor first | No | Yes — always |
| Output | An inferred household or company | A named supporter record |
| Accuracy trend | Degrading | Stable |
| Works on mobile | Poorly | Yes |
| Audit trail | None | Timestamp, version, source URL, categories |
| Coverage of your traffic | Broad but low-confidence | Narrower and high-confidence |
IP targeting reaches more of your traffic. It reaches it with a guess. Which trade you want is a real decision — we are only asking that you make it knowing which one you are buying.
Related
See what your own website already knows.
Book a 20-minute walkthrough. We will look at your site's current consent posture first — that part is useful whether or not you buy anything.