There's a difference between "chat with our AI assistant" and "chat with Doc."
The first sounds like every other chatbot on the internet. The second sounds like a person you'll remember tomorrow. The second is also exactly what DocuTrain's chat character feature lets you build — an assistant with a name, a face, and a personality that's unmistakably yours.
DocuTrain itself uses Doc as the friendly example character in our own demos and blog artwork. Here's what that can look like:
This post walks through how to create one for your organization, what your users see, and why this small piece of configuration carries more weight than almost anything else in the product.
What a chat character is
A chat character is an organization-wide persona configured in Owner Settings → Chat Assistant. It has four main pieces:
| Piece | What it does |
|---|---|
| Name | The character's display name — e.g. Doc, Henley, Dr. Helena, Atlas |
| Avatar | A circular image — uploaded or generated from a short description |
| Embed intro | A rich welcome message that appears in embed experiences |
| Accent color | The brand color used for highlights, buttons, and message accents |
Once enabled, assistant replies are labeled with your character instead of a generic "Assistant" label:
[Avatar]
Doc Says…
─────────────────────────────────
Cardiac telemetry should be initiated for patients
presenting with chest pain and ECG changes consistent
with acute coronary syndrome…The accent-color stripe runs down the left side of every message. The avatar appears next to every response. The name labels every reply. The character isn't just decoration — it's how the AI presents itself, every single turn.
Setting it up
Inside Owner Settings, open the Chat Assistant tab. There's a master toggle: Enable Custom Chat Assistant Character. Turn it on (Business plan or higher), and the fields appear underneath.
Step 1: Pick the name
The name appears before every assistant message. Pick something short, pronounceable, and on-brand:
- A friendly default like Doc if you want a clear, professional mascot tied to "documentation" and "doctoring" answers from sources.
- A persona that fits your audience: Henley for a warm product assistant, Dr. Helena for a clinical guideline, Atlas for a developer-tools company.
- An animal or mascot tied to your industry: Owen the Owl for an education platform, Pixel for a design tool.
- Your founder's first name if your brand is personal: Ask Maya for a solo consultant's knowledge base.
The form suggests examples like Doc, Henley, Atlas — simple patterns that work well in the UI.
Step 2: Give it a face
The avatar is where the character stops being a label and starts feeling like a person. Two paths:
Upload your own. PNG, JPG, GIF, or WebP up to 2 MB. Square works best — the avatar shows in a circle in chat and in embeds. Use a brand mascot, illustration, or an approved photo if someone on your team is the face of the product.
Generate one. Type a short description — "friendly robot", "owl mascot", "helpful assistant in glasses" — and DocuTrain proposes an image tuned for small circular crops: tight framing, simple background, no tiny text.
You can try a few times until it looks right; there are reasonable limits so casual experimentation doesn't get out of hand.
Images are stored securely and delivered quickly to visitors.
Step 3: Write the embed intro
The embed intro is the welcome block under the avatar on your embedded character experience. It's where you explain what the character does in your voice:
Ask Dr. Helena about our clinical protocols, drug-interaction guidelines, and patient-safety standards. She'll cite the exact source documents so you can verify her answers.
Hi, I'm Doc. I've read every line of our product documentation — ask me anything about setup, troubleshooting, or advanced features.
The intro supports formatting — bold, italics, links, lists. Keep it short. Two sentences is usually plenty.
Step 4: Match the accent color
One highlight color ties the character UI together: the message accent stripe, focus states in search, active tabs, buttons.
Pick from your brand palette with the color picker, or paste a hex code. A default blue ships with the product if you don't have brand colors yet — swapping to your real palette pays off the moment someone lands on your embed.
Live preview, before you save
The configuration tab includes a live preview of how your character will appear in chat — avatar, name strip, accent border, sample text — updating as you edit so you can experiment without save-and-refresh.
Per-document opt-in
Not every document wants a chat character. A casual product FAQ might use Doc or another mascot. A regulatory disclosure might need a sober, neutral assistant voice.
Use the per-document toggle under UI Options. The organization-level character must be enabled first; then you can turn the character on or off per document — friendly mascot here, neutral mode there, all under one account.
Why a character changes everything
Most chatbots feel like a search box that talks back. The bar for trust is low. The bar for being remembered is even lower.
A named, branded character flips both:
- Users remember the experience. "I'll ask Doc" (or Henley, or Helena) is something a user actually says out loud. "I'll ask the AI" is something they think and then forget.
- Trust transfers from your brand. If users trust your company, they transfer that trust to the named assistant. A generic assistant is judged as "another bot." Doc (or your chosen name) is judged as a representative of you.
- Mistakes are forgivable. When something isn't in the documents, "Doc doesn't have that information" feels human. "The AI cannot answer" feels broken.
- Embeds feel native. Your character, colors, and intro don't look like a generic widget bolted on — they look like part of your site.
There's a reason many AI products use a name: it turns a capability into something people relate to. DocuTrain lets you ship your name and face — and Doc is our go-to example of how that feels.
What the user sees
In live chat:
- Header on every assistant message with avatar and "{Name} Says…"
- Accent stripe along the message
- The same character presence during loading — so the wait feels on-brand, not empty
In the character embed:
- Large avatar up top
- Page title like "Welcome to Doc's Database" (using whatever name you chose)
- Your intro under the avatar
- Accent color throughout
On your organization landing page, the character can appear beside your logo so visitors see the assistant's face immediately.
Plan gating
Full chat-character customization is a Business feature. On Free and Pro you'll see the controls with an upgrade note; on Business and Unlimited everything is available.
That's intentional — the character is one of the highest-leverage parts of a Business deployment. Lower tiers still get grounded answers; Business pays for the branded presentation.
The character is the product
A lot of DocuTrain settings are invisible — advanced limits, permission details. Important, but visitors never see them.
The chat character is the opposite. It's the most visible part of the product. It's what people remember and repeat to colleagues. ("Have you tried asking Doc?") If you're going to spend focused time on any single setting, spend it here.
The next post covers the character-database embed — your avatar, searchable library, and chat on your own site in one snippet.
→ Next: Your Character's Document Database: The Embeddable Hub