Start free trial →
Feature · Source Outreach

Knowing you're not cited is the easy part. This is the part that fixes it.

Every AI-visibility tool can tell you a competitor is cited and you aren't. Truffle turns that gap into a contact, a pitch seeded with your own data, and a pipeline you can work. Then it keeps watching until the new citation shows up in the answer.

7-day trial · no credit card · Source Outreach included in the trial

1 · GAP FOUND saas-vergleich.de Cites 3 competitors, never you DA 54 2 · CONTACT + PITCH Editor found · draft ready Seeded with your data + the prompt YOU SEND 3 · PLACED Citation live in the answer ChatGPT · Perplexity · tracked daily +38 CITATIONS

Illustrative data · demo brand "Flowmetrics"

Why the source list decides the answer

An AI answer isn't written from nothing. The model builds it out of a small set of pages it already trusts. Look at which pages those actually are and the pattern gets uncomfortable: they're mostly platforms and reference sites, and your own domain is barely in the running. That's why the reflex to publish more content rarely moves an AI answer. You can't publish your way onto a shortlist assembled from somebody else's pages. Somebody has to earn a place on those pages instead. Every tool on the market will happily show you that you're missing from them, then leave the hard part to you.

23.3%
Across nearly every industry, three domains dominate AI Overview citations: YouTube (~23.3%), Wikipedia (~18.4%) and Google.com (~16.4%). The concentration gets sharper per vertical. YouTube takes ~32.4% of e-commerce citations and Reddit ~11.3%. In gaming, Reddit alone takes ~78%. The engines lean on surfaces you don't own, so getting cited means getting onto them.Surfer, "AI Citation Report 2025" (36M AI Overviews, 46M citations, March–August 2025), October 2025. Vendor research; sample and method disclosed.

What source outreach actually is

Source outreach is the practice of getting your brand onto the third-party pages that AI engines already cite when they answer your category's questions. The target list matters here. It's built from the specific pages Truffle watched an LLM reach for, so it starts with observed citations rather than a domain list you bought.

The vocabulary

Cited source: A URL an AI engine actually referenced while answering one of your tracked prompts. Truffle logs it on every daily run.
Citation gap: A source that cites your competitors and skips you. This is the raw material: the page is already trusted, you're just absent from it.
Opportunity: A citation gap with a contact route attached: author, editor or generic press address for the domain.
Placement: The outcome. Your brand is now on that source, and Truffle watches the tracked prompts to see the citation appear.
Classic link building

Targets come from a bulk domain list sorted by authority metrics. The goal is a link, a ranking signal for a search engine. Success gets counted in domains acquired. Nobody checks whether the page was ever going to influence an AI answer about your category.

Citation-source outreach

Targets are the pages an LLM was observed citing for your prompts. The goal is a mention in the answer. Success gets counted in citations that show up afterwards in ChatGPT and Perplexity. The list stays short and specific, because every page on it has already steered a model.

How it works in three steps

From an observed citation gap to a placement you can point at, without leaving the dashboard or building a prospect list by hand.

1

Find the sources AI already cites

Truffle runs your tracked prompts daily across every AI engine on your plan and records which URLs the answers lean on. It then splits those domains into the ones that cite you and the ones that cite your competitors instead. That second group is your target list. Every entry is a page a model has already chosen to trust, so you can stop guessing at which sites carry influence.

2

Get the contact and a pitch seeded with your data

For each cited domain, Truffle discovers a contact route: the author, the editor, or the generic press address. That saves you the hunt for an email. It then drafts a pitch template seeded with your strongest data points and the prompt context where the gap shows, so the editor sees the actual question their readers are asking and what you'd add to the page. You review and send it yourself from your own inbox.

3

Track it until the citation lands

Move the opportunity across the board as it progresses: identified, contacted, in conversation, won. Follow-ups stay visible and nothing rots in a spreadsheet. Once a placement goes live, the same daily tracking that found the gap watches for the payoff. The new citation shows up per engine, so you can see it land in ChatGPT and Perplexity.

A CRM for the sources that steer AI answers

Digital PR usually dies in a spreadsheet. The outreach board keeps every opportunity in one place, ranked by whether the source actually moves the answer.

Truffle Source Outreach board for demo brand Flowmetrics: four kanban columns — Identified (12), Contacted (5), In conversation (3), Won (7) — with domain cards showing DA scores, HOT tags and status notes, above summary tiles reading 27 source opportunities, 7 placements won and +86 new citations from won sources over 60 days.

Four columns, one loop. Every card is a real cited source. It carries the domain's authority score and the reason it's on the board: saas-vergleich.de, DA 54, cites 3 competitors, never you. Or salestech.review, DA 47, top source for Gemini. The HOT tag flags the gaps worth working first. Opportunities move left to right as you work them, from Identified through Contacted and In conversation to Won, each carrying the status you logged (emailed Jul 8 · follow-up due). You send the mail; Truffle keeps the record. The tiles up top close the loop: 27 opportunities extracted from AI answers, 7 placements won, and +86 new citations traced back to those won sources over 60 days. That last number is the one to judge the work by, because it measures the outcome the outreach was supposed to buy. Illustrative data for demo brand "Flowmetrics."

This is the part that closes the loop

Every AI-visibility tool ends at the diagnosis: here's your score, here's the competitor cited instead of you, good luck. The hardest question stays yours to answer. So who do I actually email? Source Outreach answers it with a name, a contact route and a draft, then tracks the placement back into the citation data that started it. Observation becomes outreach becomes a citation you can watch land.

What you get with Source Outreach

Every capability below ships today. Nothing here is a roadmap promise.

Author · editor · press
Contact discovery for every cited domain, so an opportunity arrives with a route in rather than a URL you have to go research.
Seeded pitches
Pitch templates pre-filled with your strongest data points and the prompt context where the gap shows. You edit and send.
4 stages
A pipeline view covering identified, contacted, in conversation and won, so you always know who's pitched, who replied and who placed.
Per-LLM
Citation tracking per engine. Watch the new citation land in ChatGPT while Perplexity still hasn't picked it up.
Daily refresh
Your tracked prompts re-run every day, so new citation gaps surface as opportunities while the answers shift.
In the trial
Source Outreach is included in the 7-day trial, then ships on Pro and Agency. Starter doesn't include it. See the pricing page for current tiers.

Frequently asked questions

Where do the contacts come from?
Truffle discovers them per cited domain: the author of the piece where the gap shows, the site's editor, or the generic press address when no individual is identifiable. The starting point is never a purchased prospect list. It's the set of URLs Truffle observed an AI engine citing for your tracked prompts, so a domain is on the board because a model already trusts it. Contact discovery is best-effort and depends on what the site publishes. Some domains resolve to a named editor, others only to a general inbox.
Is this just link building with a new name?
The overlap is real. It's still a human pitching an editor, and a good digital PR person will recognize most of the motion. What changes is the target list. Link building starts from an authority-sorted domain list and counts links acquired for a search engine. Source outreach starts from pages an LLM was observed citing when answering your prompts, and counts citations that appear in the answer afterwards. A DA 80 domain that never gets cited in your category is a fine link-building target and a waste of a source-outreach pitch.
Which plan is Source Outreach on?
The 7-day trial includes it, so you can work the board before you pay anything. On the paid ladder it starts at Pro. Starter doesn't include it, and that's worth knowing before you pick a tier: if you use Source Outreach during the trial and then buy Starter, you won't have it. Citation tracking itself runs on every paid plan, and Source Outreach is the layer on top that adds contact discovery, pitch templates and the pipeline board. Check the pricing page for the current tiers before you buy on this page's word.
Do you send the emails for me?
No. Truffle handles discovery, drafting and tracking. It finds the source, finds the contact, seeds the pitch and tracks the outcome. You send the mail from your own inbox and log what happened on the card. That line is deliberate. Automated outreach blasts are what trains editors to ignore pitches, and the sources that steer AI answers usually have a human reading that inbox. The board keeps the record of your sending. It doesn't do the sending.
How do you know when a citation lands?
The same mechanism that found the gap. Truffle re-runs your tracked prompts daily across the AI engines on your plan and logs which URLs each answer cites. When a domain you placed on starts appearing in those answers, it shows up in your citation data per engine, so you can watch it register in ChatGPT while Perplexity hasn't picked it up yet. Attribution here is observational. Truffle shows you the citation appearing after the placement, which is good evidence and still short of a controlled experiment.
How fast do I get the first result?
Two clocks run here and they're worth separating. Opportunities are fast. Once your prompts have run, cited domains and their gaps sit on the board within the first day or so. The placement is slow, because it depends on a human editor. Pitching, replies and publication run on the same weeks-long timeline as any digital PR, and plenty of pitches get no answer at all. The citation follows the placement over the days and weeks after the page goes live, as engines re-crawl and re-weight the source. Anyone promising you a citation on a fixed date is selling something.

Stop reading about the gap. Start closing it.

The 7-day trial includes Source Outreach. Find the sources the AI engines already cite in your category, get the contacts, and send your first pitches before you pay anything.

No credit card · cancel anytime · Source Outreach included in the trial, then on Pro and Agency

Newcomer AI-Visibility Tracker · known from