What the Product Knowledge Base is
The Product Knowledge Base is a single profile of what you sell: your product, who it is for, how you win, and how you want to sound. You fill it in once, in plain language, and Linkeddit reads it every time it generates something on your behalf.
Think of it as the brief you would hand a sharp new hire on day one. Without it, the AI has to guess. With it, every draft, score, and recommendation is grounded in your actual positioning instead of generic best practices.
Every field is optional and you can edit it any time. The more you add, the better the output, so treat it as a living document rather than a one-time form.
Where your knowledge base is used
The profile is not a settings page you fill in and forget. It feeds directly into the features you use every day:
- Competitor briefs (Compete): grounds the "why it matters" analysis so each competitor move is read against your product, not in a vacuum.
- Lead scoring: ranks prospects against your ideal customer profile so the strongest fits rise to the top.
- AI reply drafts: writes on-brand replies to matched posts and conversations across Reddit, review sites, and the open web.
- Lead DM drafts: personalizes outreach using your value props, proof points, and signature.
- Content ideas: suggests topics tied to your product and the problems your buyers actually describe.
How to fill it out
The page groups fields into four categories. Start with Product, Audience, and Positioning. Those three move the needle most. Voice and outreach then shapes how everything reads.
Product
What you sell and why it wins.
- Product description: what you sell, who it is for, and the core problem it solves, in a short paragraph. Example: "A scheduling tool for solo accountants that automates client reminders and intake forms."
- Value propositions: the two to four reasons buyers choose you over the alternatives. Example: "Set up in a day. No per-seat pricing. Built for a single practitioner, not an enterprise team."
- What sets you apart: the capability or approach only you have. Example: "The only intake flow that syncs directly with QuickBooks Self-Employed."
- Top use cases: the main jobs customers hire you for. Example: "Automating client onboarding; chasing overdue documents; booking quarterly review calls."
- Problems you solve: the pains buyers describe, in their own words. This powers intent matching, so phrase it the way a real buyer would complain. Example: "I waste hours every week emailing clients for the same paperwork."
Audience
Who you are built for.
- Ideal customer / persona: the roles, seniority, and traits of your best-fit buyers. Example: "Solo or two-person accounting practices, owner is also the person doing the admin."
- Target market: the industries, company sizes, and regions you focus on. Example: "Independent bookkeepers and accountants in the US and Canada, under five staff."
Positioning
How you stack up against the field.
- Competitor positioning: how you position against key competitors and what you beat them on. Example: "Versus the big practice-suites, we are cheaper, faster to set up, and do not charge per client."
- Pricing & value: your pricing model and how you frame value against alternatives. Example: "Flat 29 dollars a month, unlimited clients, no annual contract."
- Proof & social proof: notable customers, metrics, case studies, or awards you can point to. Keep this to things that are actually true. Example: "Used by 400 independent practices; average user saves four hours a week."
Voice & outreach
How the AI should sound when it writes for you.
- Voice & tone: the personality behind the writing. Example: "Friendly founder, plainspoken, no hype or jargon."
- Writing style: how replies and DMs should read, plus phrasing you like. Example: "Short sentences. Lead with the answer. Never open with a compliment."
- Do NOT claim / avoid: claims, topics, or language the AI must never use. Example: "Never say 'best in class'. Do not compare on price. Do not promise tax advice."
- Signature: how you sign outreach. Example: "Jordan, founder at Ledgerly, ledgerly.com".
Tips for great inputs
- Be specific. "Saves time" tells the AI nothing. "Cuts client onboarding from two days to twenty minutes" gives it something to work with.
- Use your buyers' words. For "Problems you solve," paste the phrases customers actually use. Matching on real language is what surfaces buying intent.
- Keep it current. Revisit the profile after a launch, a pricing change, or a new competitor. Stale positioning produces stale drafts.
- Do not fabricate proof. Only list metrics and customers that are real. The AI will repeat what you put here in outreach.
Privacy
Your knowledge base is private to your account and is never shared with other users. It is used only to ground the AI features you run inside Linkeddit.
Common questions
Do I have to fill in every field?
No. Every field is optional. A strong Product description alone already improves your output. Add the rest over time, starting with Product, Audience, and Positioning.
How often should I update it?
Whenever your product or market changes. A ten-minute review after a launch, a pricing change, or a new competitor keeps every draft accurate.
Does editing it change drafts I already generated?
No. Existing drafts stay as they are. The updated profile applies to everything you generate after you save.
Where do I edit it?
On the Product Knowledge Base page. Fill in the fields, watch the completeness meter, and hit Save. Your changes take effect on the next thing you generate.