What is Seth's knowledge?
Knowledge is where you add information about your business. When a customer asks a question, Seth AI always checks the knowledge base for the solution.
Where can I find Seth's knowledge base?
In the menu on the left of your screen, click Seth AI support > 'Train' tab > Using knowledge and rules.
What are the limits of Seth's knowledge base?
Here is how much you can add as knowledge to Seth:
500 webpages (urls), which are auto-synced every 24 hours. Product pages are not counted toward this limit.
20,000 products synced from Shopify (Shopify stores only). Inventory is tracked too.
1,000,000 characters of plain text in total (~200,000 words) or 50 text files, whichever you reach first.
25 documents (each one being max 50MB)
How to upload knowledge to Seth
First, click on what type of content you want to upload:
Option 1) Add a website
Click on 'Websites' and then add your website url. Ringly will then start scraping all the pages in the website url you pasted.
You'll see all the pages we found. Review then and only add relevant pages for Seth to solve tickets. Then, hit confirm and upload.
Notes on scraping a website:
If you add a url like ringly.io we will also scrape help.ringly.io or docs.ringly.io
if you add https://www.ecomstore/products we will scrape all urls of your website that have the /products path.
Sometimes the scraper didn't find all your urls. In this case you can add additional urls manually later.
Your website will start syncing a bit. It usually takes 10-30 seconds. But bigger websites take longer.
When the website is synced, you can click on it to open all the content inside. Here you can see what pages are synced.
Possible statuses for the pages:
Live: the page is live knowledge of Seth
Excluded: the page is not live knowledge
Failed: This page couldn't be accessed. It may be slow to load or blocked by anti-crawler settings or a firewall. Check your site configuration and try syncing again. If the issue keeps happening, contact us.
Resyncing: You can resync a failed or excluded page. Then, it will show as "Resyncing..." some time until it becomes either live or failed.
Adding an additional URL
Sometimes the scraper didn't find all your urls. In this case you can add additional urls manually. In the top right of the website contents, click "+ Add additional url".
You can only add new urls from the same domain as the website. If you want to add urls for another website, you can just create a new website.
Option 2) Upload plain text
This is useful for things not on your website, but you still want Seth to know. Just paste or type plain text.
Steps:
Click 'Plain text'.
Give your text file a name
Paste or type your text content into the input field
Click 'Attach text'.
Option 3) Upload a document
Useful if you have SOPs, product brochures, manuals etc.
Supported formats: You can upload documents in PDF, CSV, HTML, RTF or XML (max. 50MB per document)
Steps:
Click on 'Documents'.
Give your document a name
Click on the upload area or drag and drop a file
Click 'Attach file'
Option 4) Sync products from Shopify
If your Shopify store is connected to Ringly.io, Seth automatically syncs your full product catalog. No setup needed.
What gets synced:
Up to 20,000 products from your Shopify store
Product titles, descriptions, variants, and prices
Live inventory levels, so Seth knows what's in stock
Products are stored separately from your website knowledge, so they don't count toward the 500 web page limit. Retrieval is also tuned for product questions, so Seth picks the right product faster.
Not connected to Shopify yet? Follow this guide.
Knowledge best practices
You can check what knowledge Seth used in the Call History.
Images will use up a lot of storage space, so it's best to avoid using documents with images or styling.
Give your files descriptive names. This way the AI phone agent knows where to look.
Call your agent to see if it gives the right answers
Use markdown format for the source, with clear paragraphs: it will be easier to chunk and is easier to be retrieved.
For tabular data or image data, it’s sometimes hard for the system to recognize the correct data and put relevant information together in one chunk, so it might get suboptimal results.
Group relevant information together in one chunk: it will improve the retrieval accuracy.
Avoid ambiguity and use specificity in references: when the chunk is retrieved, it’s best to contain specific references like name, dates, and less ambiguous terms like
itorthis. Because without the prior chunk, it’s hard to understand ambiguous terms.