IP targeting for B2B.
What IP targeting is, where it genuinely works, why remote work and shared IPs break it, and how person-level identity matching reaches actual people instead of buildings.
IP targeting serves ads to every device on a given internet connection by matching the IP address to a physical location, usually an office. It reaches a building, not a person. You can't tell the CEO from the receptionist, you can't target by job title, and if your buyer is working from home that day, you miss them entirely.
I've watched teams spend real budget on IP targeting and walk away confused about why their pipeline didn't move. The mechanic isn't broken. It's just aimed at the wrong unit.
You don't sell to buildings. You sell to people.
This article covers what IP targeting actually is, where it genuinely works (it does have real uses), where it quietly breaks, and the person-level alternative that fixes the thing IP was always missing.
It's part of the contact-level marketing system — the larger argument that B2B advertising should be aimed at named contacts, not proxies for them.
What IP targeting is.
Every device on the internet has an IP address. Companies that buy their own internet connection get a block of those addresses, registered to their organization.
IP targeting works backward from that.
A vendor builds a map of company names to the IP ranges their offices use. You hand them a target account list, they resolve it to IP ranges, and they serve display or social ads to any traffic coming from those networks.
Anyone in that office, on that connection, can see your ad.
That's it. You're targeting a network address that happens to sit inside a building.
Reverse IP is the same trick, flipped.
The mirror image of IP targeting is reverse-IP identification. Instead of serving ads to an IP, you watch your website traffic and try to name the company behind each anonymous visit.
Same database. Same assumption: this IP belongs to this company.
Both depend on a clean line from an IP block to an org. And that line is messier than most vendors admit. More on that below.
Where IP targeting genuinely works.
I'm not here to tell you IP targeting is useless. It's not. There are a few jobs it does well, and you should use it for them.
Event and venue geofencing.
This is the best use. Draw a boundary around a trade-show hall, a conference center, a competitor's user conference. Serve ads to devices inside that boundary during the event.
You don't need to know names. You know intent by location — these people walked into this room on purpose. Geofencing earns its budget here.
A single large enterprise headquarters.
If your whole campaign is one giant account — a Fortune 500 with its own registered IP block and thousands of people in one tower — IP targeting can blanket that building with awareness.
Large enterprises with dedicated IP ranges are exactly the case where reverse-IP accuracy holds up best.
Coarse, top-of-funnel account awareness.
If all you want is for the logo to appear in front of "someone at the account," IP targeting gets you a cheap, broad impression. No personalization, no job-title precision, but a presence.
That's a real use. It's also the ceiling.
Where IP targeting breaks.
Now the part the IP-targeting vendors don't lead with.
Remote work shattered the office IP.
IP targeting assumes your buyer is sitting on the office network. A lot of them aren't.
Roughly a fifth to a quarter of US workers telework on a given month, per the Bureau of Labor Statistics. And among remote-capable knowledge workers — the exact people on a B2B buying committee — Gallup puts the share working fully remote or hybrid at around 80%.
So the VP of Marketing you're trying to reach is at her kitchen table three days a week. Her IP that day is Comcast residential, indistinguishable from her neighbors. Your office-IP campaign never touches her.
The signal IP targeting depends on has been leaking out of the building for years.
Most companies don't own their IPs.
Here's the structural one. To resolve a company to an IP range, that company has to actually be registered as the owner of the block.
Most aren't. As Demandbase puts it plainly, most companies aren't the proprietors of their own IP addresses — they lease them from an ISP. So when you do a reverse lookup, you get the name of the ISP, not the customer.
Large enterprises with dedicated blocks are the exception. Everyone smaller shares dynamic IPs allocated by their provider. For most of the mid-market, the IP-to-company link is a guess.
Shared IPs put strangers in your audience.
A coworking space. An office building with twenty tenants behind one connection. A VPN that routes thousands of users through one exit node.
IP targeting can't see inside any of that. It serves your ad to the whole pipe. You pay to reach the startup down the hall and the VPN provider's entire customer base, none of whom are on your list.
No person-level precision. At all.
This is the one that matters most.
IP targeting reaches a network. It has no idea who is on the other end. The CEO, an intern, a contractor, the office printer phoning home — same IP, same ad, no way to tell them apart.
You can't run different content for the CFO and the champion. You can't sequence a buyer through awareness stages. And you can't frequency-cap a person, because IP targeting doesn't know people exist — only addresses.
When a deal stalls, IP targeting can't tell you which stakeholder went quiet. It never knew their name in the first place.
The person-level alternative.
So if the unit is wrong, fix the unit. Target the person, not the pipe.
That's what contact-level targeting does. Instead of resolving a company to an IP range, you start with a list of named contacts and resolve each one to their personal identifiers — the personal email, personal phone, and account identifiers that LinkedIn, Meta, Google, Reddit, and X actually match on.
This runs through an identity graph. Your CRM has a business email and 1-3 fields per contact. Identity enrichment maps that business identity to the personal identifiers the platforms use and adds 50-70 data points per contact.
The ad now follows Sarah Chen. Office, home, phone, weekend — wherever she logs into her accounts. Her IP can change every day and it doesn't matter, because you're not chasing her network. You're reaching her.
And because every contact is matched to a real identity, you get back the thing IP targeting can never give you: you know who saw the ad, who clicked, and who went quiet. By name. That's how you find the stalled stakeholder in a deal.
This is also the only way to reach a B2B audience on Meta, Google, Reddit, and X, which have no professional targeting filters and no concept of "company IP." An enriched contact list is the key that opens those platforms — at CPMs as low as ~$3 on Meta versus ~$30 on LinkedIn.
IP targeting vs contact-level targeting.
An honest side-by-side. I run a contact-level platform, so read the table knowing that — but every row here is a structural fact, not a preference.
| IP targeting | Contact-level targeting | |
|---|---|---|
| Targeting unit | A network address (a building) | A named person |
| Precision | Whole office, anyone on the IP | Specific contact, by name |
| Job-title / persona control | None | Different content per persona |
| Remote-work resilience | Fails — buyer off the office IP is invisible | Holds — follows the person across networks |
| Shared IPs / coworking / VPN | Strangers land in your audience | Unaffected — matched to identity, not IP |
| Accuracy by company size | Best for large enterprises; weak for SMB | Consistent — doesn't depend on IP ownership |
| Frequency cap a person | No — doesn't know people exist | Yes — capped per contact |
| Platforms | Mostly display + some social | LinkedIn, Meta, Google, Reddit, X |
| Measurement | Aggregate, "someone at the account" | Contact-level — who saw, clicked, stalled |
| Match rate to your target list | Low for most companies (leased/shared IPs) | 70-99% after enrichment vs ~30% native upload |
The pattern is simple. IP targeting is a location signal. Contact-level targeting is an identity signal. Location degrades the moment your buyer leaves the building. Identity doesn't.
When to use which.
You don't have to pick a religion. You have to match the tool to the job.
Use IP targeting when location is the signal.
→ Geofencing a trade show, conference, or competitor's office.
→ Blanketing a single large enterprise HQ with its own IP block.
→ Coarse, cheap, top-of-funnel logo awareness where precision doesn't matter.
Use contact-level targeting when the person is the signal.
→ Reaching specific people on a buying committee, by name.
→ Running different content for the CFO, the champion, and the CTO.
→ Reaching buyers off LinkedIn — on Meta, Google, Reddit, or X.
→ Reaching people who work from home, which is most of them now.
→ Measuring who actually engaged, so sales knows who's warm.
Most B2B teams reach for IP targeting because it's the version of "account targeting" they heard of first. But what they actually want — reach the right people, run the right content, see who responded — is the person-level job. The IP was always a stand-in for the person. Now you can skip the stand-in.
If naming the anonymous companies on your site is the problem you're trying to solve, that's website visitor identification, and the same person-level-vs-company-level split applies there too.
Go deeper.
IP targeting is a proxy for the thing you actually want: the right person, on the right platform, with the right content. Contact-level marketing is the system that targets the person directly.
→ Contact-level targeting — how identity enrichment works, match-rate mechanics by platform, and cost per contact reached.
→ Contact-level advertising — the paid distribution layer: how to deliver specific content to specific named contacts across five platforms.
→ Website visitor identification — the reverse problem: putting names on the anonymous companies and people already visiting your site, person-level vs company-level.