Key takeaways
- Profound doesn't offer a direct data export to Promptwatch, so migration requires a structured manual process -- but it's manageable in a few days.
- Your most valuable historical assets are your prompt library, competitor benchmarks, and citation snapshots. Export these first.
- Promptwatch covers the same 10+ AI models as Profound and adds capabilities Profound doesn't have: Reddit citation tracking, ChatGPT Shopping monitoring, AI crawler logs, and built-in content generation.
- Rebuilding prompt tracking in Promptwatch takes about 30-60 minutes if you've documented your Profound setup properly.
- The migration is also an opportunity to audit your prompt strategy -- most teams discover gaps they weren't tracking.
If you're reading this, you've probably already made the decision to switch. Maybe Profound's pricing jumped at renewal. Maybe you hit a wall with its monitoring-only approach and wanted a platform that actually helps you fix visibility gaps, not just report on them. Maybe your agency needs Reddit tracking or ChatGPT Shopping data that Profound doesn't surface.
Whatever the reason, the migration question is the same: how do you move without losing months of historical data that you've been using to benchmark progress?
This guide walks through the full process -- what to export, what to rebuild, and how to set up Promptwatch so you hit the ground running instead of starting from zero.
Understanding what "historical data" actually means in this context
Before you export anything, it helps to be clear about what you're actually trying to preserve. AI visibility data from Profound falls into a few categories:
- Prompt library: the specific questions and queries you've been tracking
- Visibility scores over time: your brand's mention rate, citation rate, and sentiment trends
- Competitor benchmarks: how your visibility compared to specific competitors across models
- Citation sources: which pages, domains, and third-party sources AI models were citing for your tracked prompts
- Response snapshots: the actual AI-generated answers captured at specific points in time
The bad news: there's no automated migration path between Profound and Promptwatch. They're separate platforms with different data architectures, and neither offers a "import from competitor" feature.
The good news: the data that matters most -- your prompt library and your benchmark snapshots -- is exportable from Profound and can be used to configure Promptwatch quickly. The historical trend lines won't transfer, but you can establish a clean baseline on day one in Promptwatch and use your Profound exports as reference context.
Step 1: Export everything from Profound before you cancel
This sounds obvious, but people cancel first and export later. Don't do that. Once your Profound subscription lapses, access to historical data typically ends with it.
Here's what to pull before you close the account:
Prompt library export
In Profound, navigate to your prompt management section and export your full prompt list. Most enterprise plans allow CSV export. If yours doesn't, copy the prompts manually into a spreadsheet -- you'll need them to rebuild tracking in Promptwatch.
For each prompt, record:
- The exact prompt text
- Which AI models you were tracking it on
- The category or topic group you assigned it to
- Any persona or region filters you applied
Visibility score history
Export your brand's visibility trend data for as far back as the platform allows. Even if Promptwatch won't ingest this directly, having a dated CSV of your mention rates and citation rates gives you a baseline to compare against once you're six months into Promptwatch.
Competitor benchmark snapshots
If you've been tracking competitors in Profound, export those comparison reports. Specifically, capture:
- Which competitors you were tracking
- Their visibility scores at key dates (quarterly snapshots are usually enough)
- Any prompts where competitors were significantly outperforming you
Citation source data
Export any reports showing which domains, pages, or sources AI models were citing for your tracked prompts. This is useful context when you start doing Answer Gap Analysis in Promptwatch -- you'll already know which sources were influential.
Response snapshots
If Profound captured the actual AI responses (not just scores), export those too. They're useful for qualitative comparison later.
Step 2: Organize your exports into a migration document
Before you touch Promptwatch, spend an hour organizing what you've exported. Create a simple migration document with four sections:
- Master prompt list (with categories and model coverage)
- Baseline metrics (your visibility scores at the point of migration)
- Competitor list (who you were benchmarking against)
- Key citation sources (domains that were appearing frequently)
This document becomes your source of truth during setup. It also helps if you're onboarding a team -- everyone works from the same reference point.
Step 3: Set up your Promptwatch account
Promptwatch has a free trial, so you can start configuring before you commit. The onboarding flow is guided, but here's how to approach it strategically given that you're migrating from Profound.
Connect your website
Promptwatch integrates with Cloudflare, Fastly, Vercel, server logs, Google Search Console, or a tracking snippet. Connect whichever integration fits your stack first -- this enables AI crawler log tracking, which shows you when ChatGPT, Claude, Perplexity, and other AI agents are actually crawling your pages.
This is something Profound doesn't offer, and it's genuinely useful. You'll see which pages AI engines are reading, how often they return, and whether they're encountering errors. It's the kind of data that explains why your visibility looks the way it does, not just what it is.
Configure your brand entity
Set up your brand name, domain, and any brand variations or product names you want to track. Promptwatch tracks entity mentions separately from direct citations, so be thorough here -- include product names, executive names if relevant, and any common abbreviations.
Step 4: Rebuild your prompt library
This is the core of the migration. Take your master prompt list from Step 2 and add those prompts to Promptwatch.
A few things to keep in mind:
Promptwatch assigns volume estimates and difficulty scores to each prompt, which Profound doesn't do. As you add prompts, you'll immediately see which ones have high query volume and which are easier to win. This is useful for prioritization -- you might discover that some prompts you were obsessing over in Profound have low volume, while high-volume prompts you weren't tracking are wide open.
Promptwatch also surfaces query fan-outs: when you track one prompt, it shows you the sub-queries that branch off from it. This often reveals 10-20 related prompts you weren't tracking in Profound.
The plan tiers matter here. The Essential plan ($99/mo) covers 50 prompts. Professional ($249/mo) covers 150. Business ($579/mo) covers 350. If your Profound prompt library was large, you may need to prioritize which prompts to migrate first based on business impact.
Categorize as you go
Promptwatch lets you organize prompts by topic, funnel stage, or custom categories. Use the same category structure you had in Profound -- it makes comparison easier when you're reviewing data six months from now.
Set up competitor tracking
Add the same competitors you were tracking in Profound. Promptwatch's competitor heatmaps show you who's winning for each prompt across each AI model, which gives you a more granular view than Profound's aggregate scores.
Step 5: Establish your day-one baseline
On the day you start tracking in Promptwatch, capture a manual baseline snapshot. This is the bridge between your Profound history and your Promptwatch future.
In your migration document, record:
- Date of migration
- Your visibility scores from your last Profound export
- Your Promptwatch visibility scores on day one (these will be available within the first 24-48 hours of tracking)
The numbers won't be identical -- different platforms measure visibility differently, and Promptwatch tracks some models Profound may not have covered. But having both data points lets you say "when we migrated, our Profound score was X and our Promptwatch baseline was Y" -- which is useful context for any stakeholder who asks about the data gap.
Step 6: Run an Answer Gap Analysis immediately
This is where Promptwatch starts earning its keep beyond what Profound offered. Once your prompts are set up, run an Answer Gap Analysis. This shows you which prompts your competitors are visible for that you're not -- the specific content gaps that AI models are exposing.
In Profound, you'd see that a competitor was outperforming you on a prompt. In Promptwatch, you see why -- which sources they're getting cited from, what content is filling the gap, and what you'd need to create to compete.
This is also where your exported Profound data becomes useful context. If you noted in your migration document that a competitor was consistently outperforming you on certain prompts, you can now see the specific content and citation sources driving that advantage.
Step 7: Set up AI crawler log monitoring
Once your website integration is live, check the crawler logs section. This shows you real-time data on which AI agents are crawling your site, which pages they're reading, and whether they're encountering errors.
Most teams migrating from Profound are surprised by what they find here. Common discoveries:
- AI crawlers are hitting pages that aren't being cited (often a content quality issue)
- High-value pages have crawl errors that are blocking AI indexing
- Some pages are being crawled frequently but never cited (a signal that the content isn't answering the prompt well)
None of this is visible in Profound. It's the kind of operational data that explains the gap between "we have content" and "we're being cited."
Step 8: Configure offsite citation tracking
Profound focuses primarily on your owned content. Promptwatch also tracks offsite citations -- Reddit threads, YouTube videos, third-party listicles, and external domains that AI models are citing when users ask questions relevant to your brand.
Set up offsite tracking for your key topics. You'll often find that AI models are citing Reddit discussions or competitor blog posts for prompts where you have relevant content on your own site. That's an actionable gap: either you need to get your content cited instead, or you need to participate in the external conversations AI models are drawing from.
What you'll lose (and what you won't)
Let's be honest about the tradeoffs.
What you lose:
- Continuous trend lines from your Profound tracking period. There will be a visible break in your historical data.
- Any Profound-specific features you were using (if you were on an enterprise plan with custom integrations, those don't transfer).
- The time investment in rebuilding your setup.
What you keep:
- Your prompt library (exported and rebuilt)
- Your baseline metrics (documented in your migration file)
- Your competitor list and benchmark context
What you gain:
- AI crawler logs showing how AI engines interact with your site
- Reddit and YouTube citation tracking
- ChatGPT Shopping monitoring
- Built-in content generation tied directly to your gap analysis
- Page-level visibility tracking showing which specific pages are being cited
- Prompt volume and difficulty scoring for prioritization
- Query fan-outs that expand your prompt coverage
Platform comparison: Profound vs Promptwatch
| Feature | Profound | Promptwatch |
|---|---|---|
| AI models covered | 10+ | 10 (ChatGPT, Claude, Gemini, Perplexity, Grok, DeepSeek, Copilot, Meta AI, Google AI Overviews, Mistral) |
| Prompt tracking | Yes | Yes |
| Competitor benchmarking | Yes | Yes |
| AI crawler logs | Enterprise tier only | Professional tier and above |
| Reddit citation tracking | No | Yes |
| YouTube citation tracking | No | Yes |
| ChatGPT Shopping tracking | No | Yes |
| Content generation | No | Yes (Content Agents) |
| Answer Gap Analysis | Limited | Full analysis with content briefs |
| Page-level visibility tracking | No | Yes |
| Traffic attribution | No | Yes |
| Prompt volume/difficulty scores | No | Yes |
| Query fan-outs | No | Yes |
| Pricing (entry) | Higher (enterprise-focused) | $99/mo (Essential) |
| Free trial | Yes | Yes |
How long does the migration take?
Realistically, plan for this:
- Exporting from Profound: 1-2 hours
- Organizing your migration document: 1 hour
- Setting up Promptwatch and connecting your site: 30-60 minutes
- Rebuilding your prompt library: 30-60 minutes (depending on size)
- Establishing your baseline and running initial reports: 1-2 hours
Total: most teams complete the migration in a single working day. The first meaningful data starts appearing within 24-48 hours of setup.
Common mistakes to avoid
Canceling Profound before exporting. Once access ends, you lose everything. Export first, cancel after.
Trying to recreate every prompt from Profound. Use the migration as an opportunity to audit. Some prompts you were tracking in Profound may not be worth tracking. Promptwatch's volume and difficulty scores will help you decide which ones to prioritize.
Ignoring the crawler logs. Teams that skip this step miss the most actionable data Promptwatch offers. Set up your website integration on day one.
Not documenting your baseline. The data gap between platforms is only a problem if you can't explain it. A simple migration document with your last Profound scores and your first Promptwatch scores solves this for any stakeholder conversation.
Treating the migration as a copy-paste exercise. Promptwatch surfaces data Profound doesn't, which means your strategy should evolve. Use the Answer Gap Analysis to find prompts you weren't tracking at all.
After migration: the first 30 days
Week one is about getting data flowing. Make sure your crawler logs are active, your prompts are tracking, and your competitor benchmarks are set up.
Week two is about running your first Answer Gap Analysis and identifying the highest-priority content gaps. Use Promptwatch's Content Agents to generate briefs for the top three gaps -- these are grounded in real prompt data and citation analysis, not generic SEO templates.
Weeks three and four are about watching the data. Check which pages AI crawlers are reading, which prompts your visibility is improving on, and whether the content you've created is starting to get crawled. The Agent Analytics timeline shows you the path from publish to crawl to citation.
By the end of 30 days, you'll have a cleaner picture of your AI visibility than you had in Profound -- not because Promptwatch's numbers are higher, but because the data is more specific and more actionable.
The historical trend line will have a gap. That's fine. What matters is what you do with the data from here.