Personalization Tokens: Definition and Usage Guide
One data point worth knowing before you touch your first template: adding a recipient's first name to a LinkedIn connection request lifts acceptance rates, but adding three or four scraped details to the same opening line frequently tanks them. The mechanism matters more than the volume.
What Personalization Tokens Are
Personalization tokens are dynamic placeholders, written as {{first_name}}, {{company}}, {{job_title}}, or similar, that your outreach tool replaces with real contact data the moment a message goes out. The template stays generic; the delivered message reads specific.
Every serious outreach platform supports them. Where they differ is in which fields they expose, how they source the data, and what happens when a field is empty. That last part is where most campaigns quietly break.
In Ampliflow, tokens pull from whatever data you imported, either from a LinkedIn search, a Sales Navigator export, or a CSV uploaded directly. The lead list building process determines what's available: if a field wasn't in your import, the token has nothing to draw on.
Which Tokens Actually Move the Needle
Short answer: first name, company name, and job title, in roughly that order of impact.
First name is the baseline. Skipping it makes a message feel like a broadcast. Using it correctly feels conversational. That's not a nuance, it's a real perceptual shift for the reader.
Company name works well in the opener when it frames a relevant observation, not just when it's dropped in for its own sake. "I noticed [Company] expanded into APAC recently" lands differently from "I wanted to reach out to someone at [Company]." The first is a reason. The second is padding.
Job title earns its place in mid-funnel sequences where you're addressing a specific pain point. A VP of Sales and a Head of Engineering have different concerns; acknowledging that in message two or three of a drip campaign is worth the extra variable.
Beyond those three, the marginal value drops fast. Tokens built from LinkedIn headline text are notoriously inconsistent because people write their headlines in wildly different formats, and pulling from a scraped "about" section risks truncated or nonsensical output.
| Token | Source field | Practical use | Risk if missing |
|---|---|---|---|
{{first_name}} |
Contact name | Opening line | "Hi ," broken output |
{{company}} |
Current employer | Context hook | Generic opener |
{{job_title}} |
Role field | Pain-point framing | Audience mismatch |
{{location}} |
Profile location | Localisation | Irrelevant filler |
{{mutual_connections}} |
LinkedIn graph | Social proof | Overstated claim |
{{custom_note}} |
Manual field | Hyper-specific hook | Visible empty gap |
Fallback Handling: The Part Everyone Skips
A fallback value is what your tool inserts when a token has no data. Set {{first_name}} to fall back to "there" and your opener becomes "Hi there" instead of "Hi ," with a hanging comma. Simple. Critical.
The mistake we keep seeing in campaign reviews is people who configure tokens but leave the fallback field blank. Some tools default to an empty string; others leave the literal placeholder visible in the sent message. Neither outcome is good. A recipient who sees "Hi {{first_name}}" knows instantly they're in a bulk sequence.
In Ampliflow, fallbacks are configured per token directly in the message editor. We'd recommend going further than first name: set a fallback for every token you use, including the ones you're confident are populated. Data quality in imported lists is almost never as clean as it looks on first glance.
A second pattern worth building is conditional logic. If {{company}} is empty, show a generic line; if it's populated, show the personalised version. Ampliflow's If/Else branches in the workflow builder handle this natively, so you can route contacts down different message paths based on whether a field exists, without maintaining two separate campaigns.
Over-Personalization: A Real Pattern Worth Naming
There is a threshold past which adding more tokens makes messages worse, not better.
We've seen openers like: "Hi [First], I noticed you're the [Title] at [Company] based in [City] and recently posted about [Topic]." Five merge fields in two sentences. It reads like a mail-merge test, not a human note. The reader can feel the scaffolding.
The problem is psychological. When a message contains too many individually "targeted" details, it signals effort that no human would actually spend on a cold reach-out. It triggers the same suspicion as a phishing email that knows too much. Specificity is valuable up to the point where it becomes obviously assembled.
Our working rule: one strong personalised detail per message is usually enough. Two is fine if they're genuinely related. Three or more requires a very good reason and a very short sentence.
This is separate from the account safety question, though they're connected. Sending variable messages at volume is actually a mild positive signal from LinkedIn's detection perspective because it looks less automated. But the content still has to read naturally, and a message with six tokens rarely does. The cloud-based LinkedIn automation guide covers the architecture side in more depth, including what Ampliflow's anomaly detection is watching for.
How Ampliflow Handles Tokens in Practice
In the Ampliflow message editor, you insert tokens via a dropdown or by typing the field name in double curly braces. Every token shows a live preview against a sample contact so you can see the rendered output before the sequence runs.
Because Ampliflow runs in the cloud via the Unipile API, there's no browser extension involved and no local session to manage. Messages go out on the schedule you set, with human-like timing variation built in, whether your laptop is open or not. That timing randomisation also means token-driven messages don't all arrive within seconds of each other from the same account, which matters for how the activity pattern reads.
The A/B testing feature lets you run two message variants with different token configurations against the same audience. That's the only reliable way to measure whether a specific personalisation approach is lifting your reply rate or just adding noise. We use this ourselves when we're unsure whether a company-name hook is earning its place or just making the message longer.
If a prospect replies at any point, Ampliflow auto-pauses that contact's sequence immediately. Worth mentioning here because it means a bad fallback rendering, the kind that might prompt someone to reply just to point it out, stops the sequence rather than continuing to send follow-ups with the same broken field.
Founding-member pricing is $19/month locked for life for the first 100 accounts, compared to $39/month Starter and $79/month Pro at public launch. For context, Dripify and HeyReach both start at $79/month; Zopto is $197/month. Cheaper tools like Linked Helper at $15/month are genuinely less expensive and worth considering if conditional token logic isn't a priority for your workflow. The full breakdown is on the pricing page.