The guide
Why outreach in your own voice gets more replies
First Person Outbound, explained by SimplyB2B
Cold outreach fails at scale for a predictable reason: the message arrives wearing a stranger's face. Recipients do not reply to a pitch, they reply to a person they can place. When the words, the tone, and the LinkedIn profile telling the same story all point to the same real human, something shifts. The reply rate shifts with it.
Key takeaways
- Trust is built by signal coherence: your words, your profile history, and your mutual connections all have to match.
- A voiceprint seeded from your own sent messages is not a style guide, it is a trained model that only your edits can move.
- Autonomy that graduates only as your corrections shrink means the output stays yours, not a shared template slowly drifting from you.
- Profile age, post history, and mutual connections are trust artifacts a rented persona cannot manufacture.
- One B2B advisory practice running outreach in its own voice recorded a 31% cold reply rate, against a 3 to 10 percent industry norm.
Why does personal outreach get more replies?
Recipients scan for social proof before they read a word. Profile age, a consistent post history, and shared connections signal that the sender is a real peer, not a rented persona. When the message tone matches that public record, the cognitive friction that normally kills a reply disappears. That coherence is what personal outreach produces and templated outreach cannot.
A cold message is a small trust test. The reader's first move is almost never to read carefully. They glance at the profile photo, skim the headline, check whether they share any connections, and only then decide whether the words are worth a second. If anything in that scan feels borrowed or generic, the message dies before the ask lands.
Personal outreach passes that test because every signal points the same direction. The writing sounds like the person behind the profile, the profile shows a real professional history, and the mutual connections confirm the sender exists in the reader's world. No template can assemble those three things because two of them predate the message by years.
Seed
Your sent messages and posts train your per-account voiceprint
Draft
Outreach is written in your voice from that model
Review
You approve, edit, or reject before anything sends
Train
Your edits update only your voiceprint, never a shared model
Graduate
Autonomy grows only as your corrections shrink
Does outreach in your own voice actually work better?
One B2B advisory practice runs its entire outreach on SimplyB2B in its own voice. Over a single month the practice recorded roughly a 31% reply rate on cold outreach, 172 genuine replies from 552 people contacted, against a 3 to 10 percent cold-outreach norm. Around 10 meetings booked came from those replies, with over 1,100 outbound conversations handled in the same period.
That figure is not a product-wide claim and it should not be read as a guarantee. Reply rates depend on the quality of the target list, the relevance of the offer, and how well the outreach reflects a voice that readers find credible. What the number shows is that the mechanism, outreach that sounds and reads like a real person with a real profile, does something that volume-blasting cannot.
The difference between 6% and 31% is not a subject-line trick. It is the compounding effect of a message that reads right, from a profile that looks right, to a recipient who can verify the sender is real before deciding to reply.
How does a voiceprint actually capture someone's writing style?
SimplyB2B seeds each voiceprint from the user's own sent messages and posts on that specific LinkedIn account. Not a questionnaire, not a shared model, not a style prompt. Every edit the user makes to a drafted message trains only their own voiceprint. The model shifts toward the user as corrections shrink, not toward a platform average.
The practical effect is that the output does not sound like AI-assisted outreach, it sounds like the person who has been writing in that inbox for years. A reader receiving the message has no shared-model artifact to detect because there is no shared model. The only data shaping the output is the user's own writing history on that account.
Autonomy expands on a graduation model. Early on, the user approves everything. As edits shrink, the system earns more room to act. That is not a convenience feature. It is the mechanism that prevents drift, the slow slide toward generic that makes most AI outreach indistinguishable from every other AI outreach.
What makes a rented AI persona less credible than your own profile?
A rented persona can copy a tone but it cannot manufacture profile age, post history, or mutual connections. Those three artifacts are what a recipient checks before trusting a cold message. They take years to build and they belong to a real person. No tool that runs outreach from a fabricated or borrowed identity can replicate them because they are chronologically impossible to fake.
This is why the borrowed-face robot is a foil, not a solution. It solves the wrong problem. The problem is not that you need more messages sent. The problem is that the recipient needs a fast reason to trust the sender. Volume from a fake profile makes that harder, not easier, because every additional message from a non-credible source trains the reader to ignore the whole category.
Your real profile compounds over time. Every post, every mutual connection added, every replied-to comment makes the next cold outreach slightly easier to trust. Running outreach from your own account is not just safer for your LinkedIn standing, it is the only approach where the trust infrastructure grows in your favour.
How does SimplyB2B keep a real account safe while running outreach?
SimplyB2B connects to a LinkedIn account via a permissioned integration that stores no password and can be revoked at any time. A warm-up ramp eases a new account into activity gradually. A fixed daily activity ceiling cannot be raised. Per-account behavioural timing variance means no two accounts act in lockstep, removing the pattern that LinkedIn's automated-activity detection looks for.
These are not optional settings. They are the defaults the system runs on because the account being protected is the user's real professional identity, not a throwaway. The ceiling being fixed and non-negotiable is a deliberate design choice. Raising it to hit a volume target is exactly how accounts get flagged, and no short-term reply gain is worth the loss of an account that took years to build.
Account safety is also where the first-person model earns its credibility back. Every protective constraint, the ramp, the ceiling, the timing variance, exists because the outreach is running on something that matters. That is a different posture from a tool built to exhaust disposable accounts at scale.
Common questions
Will prospects know the outreach was assisted by AI?
Not from the writing alone. The voiceprint is built from your own sent messages, so the output reflects your actual patterns, not a generic AI register. The bigger signal recipients check is your profile, and that belongs entirely to you. Authenticity is a function of coherence across message and profile, not whether a tool was involved.
Does SimplyB2B work for someone who is not a natural writer?
Yes, and that is a meaningful part of the value. The system drafts in the voice your sent history already shows, which means it works from what you actually sound like, not an idealised version. You correct what feels wrong, those corrections train the model, and the drafts improve toward your real voice rather than toward a house style.
Can I try it without connecting my LinkedIn account?
The free Voiceprint tier lets you see how SimplyB2B reads and models your voice before you connect anything. You only pay when you connect an account, and the Voice plan at $59 a month writes posts and outreach for you to post yourself, so your account is never touched. Account connection comes with the Presence plan from $199 a month.
Is LinkedIn the only channel SimplyB2B covers?
Yes, today the productized channel is LinkedIn: connection requests, direct messages, follow-ups, comments, and posts. Email is not part of the current offering. The focus on one channel is deliberate, it is where the per-account safety model and the voiceprint approach are built to run.
Related: Why outbound at scale fails · First-person outbound explained · AI SDR vs founder-led outreach · The graduation model · Own it or outsource it
See how SimplyB2B reads your voice
Start with the free Voiceprint, no account connection required, and see the model before you commit to anything.
Free to set up · pay only when you connect · 30 day money back guarantee