Claude — best ways to get ad platform data without getting banned?

[Malan19]

Check out Segmentstream — they have MCP that works Claude and ChatGPT

[haldiii4o]

The workaround we found here is not to use unofficial MCPs directly, but use middleware tools that have official API integration underneath

We’re currently use segmentstream free plan, and pull data from 50+ Meta and TikTOk accounts which works fine (we focus on paid social, don’t do paid search so can’t comment on how it works for google ads and bing)

[Hypgamer12]

Would using the Meta Graph API not be a viable option in this case?

[Novara_Paradise]

Use middleware like n8n or Make to get around this alot of these have native integrations that will allow you to get the info

[X-3SY]

honestly I’d recommend avoid MCPs for now.. just import CSVs and you’ll be just fine. Yes it takes 10 min work daily, but you don’t need to pay for any tools or risk losing your client accounts..

[Oblap]

Use adzviser to connect all your accounts to sheets or better yet get everything in bigquery and then you can see what’s what. 

[shashank_223]

Use ChatGPT lol

[Former_Engineer6582]

So I’m an associate performance director at large network media agency (we’ve also been in the news due to sheer amount of recent people cuts, and AI is one of the reasons). I stay as luckily i’m on the AI side and actively try to implement it throughout our agency.

We’ve tested a few tools — pipe dream (honestly not good quality) and windsor (good, but only raw data). So far we’ve found the most useful SegmentStream. It brings all data in our own BigQuery instance, connects to GA4, Hubspot, etc. applies attribution logic, good skills library, etc.

Honestly so far it’s too good to be true, because it’s free (for now?). It’s kind of earlyresearch preview, so I hope it won’t turn into some big $$$ after next few months of extensive usage.

The alternative approach would be getting raw data via Make/n8n and other tooling that has offiical data connectors, but then you’d need to use AI to somehow clean and unify this data to prepare for normal AI usage.

[ppcbetter_says]

Couldn’t you just export the data to a google sheet and then have Claude read it from there?

[BMXviper]

If you want to avoid conversion deduplication, then the best way is to move away from exporting raw ad platform conversion data, and apply unified attribution for all your channels

We export Google Analytics data in BigQuery using GA4 link in settings

We also get ad platform data in there, and unify it to have table with all costs, clicks, impressions

Unfortunately since its GA4 we have last-click attribution only, but at least we don’t have inflated revenue numbers 😀

Our data analysts does it, so I cannot help with telling you how to do this exactly

Once data in GBQ , our Claude can read from it and write html reports, and do its AI magic

[Puzzled-Smoke-6349]

I recently started using windsor.ai to pull data. It works for me so far but there some limitations

[cameo11]

Have you tried Manus? Meta acquired them and I believe that’s the only AI platform they’re allowing in at all!

But yea like other said my thoughts would be to use a data warehouse like Funnel or similar?

[Icy-Satisfaction8793]

I’ve recently made a research of marketing analytics tools that support MCPs to Claude/Cursor/ChatGPT and similar

Here are four i’ve found (in no specific order):

segmentstream.ai

windsor.ai

pipeboard.co

supermetrics.com

The first one probably matches your needs best since its support more than just raw data piping but also bigquery storage and custom cross channel attribution. Other more focused on data connectors but not on its analysis. All seems legit and popular

have a good day

[CroRad1987]

Coupler.io has Claude connection

[basse1985]

We use Catchr to send data to bigquery and then analyse data there. We have Google, meta, Shopify, TikTok and more in our bigquery. So the data is analysed across platforms.