Sync website- teaching NexoFlow from your existing site
Sync Website reads the business's existing website and stores what it finds as brand knowledge. NexoFlow then writes from real facts on the site rather than guessing.
What it is
Sync Website reads the business's existing website and stores what it finds as brand knowledge. NexoFlow then writes from real facts on the site rather than guessing.
This is the difference between "we offer roofing services" and "we offer the same-week tile roof replacement you already advertise on your services page, with the 15-year warranty you already mention".
Where to find it
Sidebar > Brand knowledge > Sync Website
The page is titled "Read your website" in the app. It only appears for projects that have a website address set.
What it reads
For a WordPress site:
- Published posts and pages
- Internal links and navigation
For any other website:
- Your pages and their text
- Your site menu links
For both:
- Your logo, brand colours, and contact details
Nothing is published or changed on your website by syncing. It is read-only.
Why it matters
Syncing improves three things at once:
1. Accuracy
Content references your real services, real service area, and real claims.
2. Internal linking
NexoFlow learns which pages exist on the site, so new blog posts can link to your real service pages and city pages instead of dead links.
3. Brand consistency
Your logo, colours and contact details get picked up and reused in visuals and calls to action.
Running a sync
Open the page, choose the project, and start the sync. Each run is recorded in a history list with its status, so you can see what was read and when.
Automatic re-sync
There is an auto-sync toggle. When it is on, NexoFlow re-reads the site periodically so the knowledge does not go stale as the site changes. The page shows when the next automatic sync is due.
Turn it off if the site is static and you would rather control when it runs.
When to re-sync manually
- After adding or rewriting service pages
- After changing the phone number, address or business name
- After a site redesign or a move to a new domain
- If content is referencing pages that no longer exist
Other knowledge sources
The website is not the only thing NexoFlow can learn from. You can also add:
- PDF files- Brochures, spec sheets, price lists, service guides
- Word documents- Internal briefs, style guides, FAQs
- Web links- Any URL worth learning from, including pages on other sites
- YouTube videos- Spoken content from a video
- Google Drive- Connect a Drive folder and sync documents from it
Each source is processed and becomes searchable brand knowledge, so when the AI writes about a topic it can pull the relevant passage from your own material.
A source moves through: pending, then processing, then ready. If a source shows as failed, the file or link could not be read- see troubleshooting below.
Common questions
Does syncing change anything on my website?
No. It only reads. Nothing is edited, published or deleted on your site.
How often should I sync?
Turn on auto-sync and forget about it. Run a manual sync after any significant change to the site, such as new service pages or a new phone number.
I updated my website. Why is NexoFlow still using old information?
The knowledge was captured at the last sync. Run a sync again to refresh it.
Can NexoFlow learn from things that are not on my website?
Yes. Add PDFs, Word documents, links, YouTube videos, or connect a Google Drive folder as knowledge sources.
My site is behind a login / staging password. Can it still read it?
No. It can only read pages that are publicly reachable.
I don't have a website. Can I still use NexoFlow?
Yes. Fill in the Brand Profile by hand or with the "describe your business" box, and add PDFs or documents as knowledge sources.
Will it copy my existing pages into new posts?
No. It learns facts and tone from the site; it writes new content.
Troubleshooting
SYMPTOM: Sync finds very little, or fails
- Confirm the website address on the project is correct and includes the right domain (with or without www, as the site actually resolves).
- Confirm the site is publicly reachable and not password-protected.
- Sites that render entirely in JavaScript with no server-rendered text give the crawler very little to read.
- Aggressive bot protection or a firewall can block the read.
SYMPTOM: A PDF or document shows as failed
- Scanned PDFs that are just images of text contain no readable text.
- Very large files may not process. Try splitting it.
- Password-protected files cannot be opened.
SYMPTOM: Google Drive documents are not syncing
- The Drive connection may need reconnecting, particularly after permissions change. Reconnect it from the knowledge sources area.
- Only the folder you selected is synced.