Databox is an AI-powered business intelligence and analytics platform for teams that need clear, trusted answers fast. It offers the best of BI, without the complicated setup, steep price, or long learning curve. It provides a blend of powerful, but easy-to-use, features, from preparing datasets and creating the custom metrics your company needs to track, to building beautiful dashboards, customizing your reporting, and receiving AI-powered insights.
This is the 7th launch from Databox. View more
Skills Marketplace by Databox
Launched this week
A library of ready-made skills for paid ads, website performance, ecommerce, revenue, and more. Each skill pulls from your live performance data and returns a finished, shareable report in under a minute. Run them in Claude when you want answers in chat, or schedule them in n8n to land in Slack or email automatically. Free to download.
Free Options
Launch Team / Built With
Databox
Hey Product Hunt 👋
Today, I'm excited to share the launch of our new Skills Marketplace for AI-powered data analysis. It's the culmination of a decade-plus of work by the Databox team.
Nine years ago, I left HubSpot to become CEO of Databox, a tiny startup no one had heard of. I did it because they were tackling a problem I knew needed solving: making data-informed decisions easy for anyone, not just analysts.
Back then, just pulling the data together was the hard part. It was scattered across a dozen tools, and getting it into one place took real work before anyone could make sense of it.
We've largely solved that part.
The bottleneck moved downstream, to the analysis, and AI looked like the answer.
Last year, prospects started telling us they were using it for their data analysis. The output looked confident, but they kept finding the math wasn't always right. We weren't surprised once we saw how they were doing it: feeding the AI partial data by uploading CSVs, screen-grabbing charts from different tools, or wiring up one MCP server at a time.
The fix isn't to stop using AI. It's to connect the AI to something that aggregates your data, applies your real metric definitions, and runs the math, so what comes back is accurate, not just confident. That's what Databox does, whether you use Genie, our built-in AI analyst, or connect your own LLM through our MCP server.
We launched both a few months back, and already a quarter of our customers use it every week.
The more they used it, the clearer the next problem got. Running the analysis was no longer the hard part. Knowing which analysis to run was.
We realized we could speed up adoption by giving people proven skills instead of a blank chat box. So we built the Skills Marketplace.
It's a free library of pre-built AI analytics skills that connect to live data and return a finished report in under a minute. Each skill is a specific analytical workflow that already knows which metrics to pull, what to compare against, what to flag, and how to structure the output.
A few examples of what's in there:
A weekly GA4 traffic report: sessions, channel breakdown, top pages, conversion rate, anomaly flags, all from live data
A cross-channel paid ads summary across Google, Meta, and LinkedIn from one trigger
Recurring workflows that deliver to Slack or email on a schedule, so no one has to remember to run them
We launched with 20+ skills and workflows. All free to download.
Take a look at what we built and let us know if it's helpful for you. 🙏
PS — What's the one report your team rebuilds by hand every week that you wish just ran itself? Drop it in the comments. We're prioritizing what to build next based on these replies.
PPS — A growing group of AI experts are building skills for our marketplace, getting exposure to our audience of 100Ks of people, including customers and partners. Want in? Reply below and we'll reach out with instructions.
@pc4media Congrats on the launch Peter. is it possible to define or extract a “what changed and why?” growth report that pulls GA4, paid ads, CRM, and revenue data together?
Databox
@zolani_matebese Yes. That's very straightforward. Our system automatically determines % change mathematically. So, you can literally ask the question in our chat interface.
A skill could also be built that allows you to tell it exactly what metrics you want to monitor, in order to avoid a lot of noise in metrics that might not matter much.
You can also ask it "Why did this change?"
Try it here: https://databox.com/ai-analyst
Further, you could setup smart alerts for key metrics that automatically alert you when something changed and doesn't require you to run a report. This'll help you detect changes as soon as they are significant, not at the end of the week or month.
https://help.databox.com/set-up-a-metric-notification
Wannabe Stark
@pc4media Great launch Peter! Love all of the free tools you and your team offer! Keep it up!
@pc4media Many congratulations Davorin, Ziga and team on shipping yet again! :)
I’m excited to be hunting Databox once again today after their previous launches Custom Integrations by Databox, Genie by Databox and Databox MCP.
The Databox Skills Marketplace is a free library of AI analytics skills that turn your live business
data into finished reports - in under a minute.
Why I hunted and endorse it? Every other AI tool makes you bring the data. Skills connect directly to
your live sources through Databox and run against your real metrics, definitions, and historical context - so the output is something you can share with your team the moment it lands, not something you have to verify before you trust it.
Download a skill, connect your sources once, trigger it in Claude. First report in under 10 minutes. Nothing else does this with 130+ data sources, zero code, and output you can trust from the first run.
Give it a spin today!
Databox
@rohanrecommends thank you - both for hunting us again and for framing it so well. 'Output you can share the moment it lands' is exactly the bar we were building to, good to see it land that way. Appreciate the continued support across launches, it means a lot to the team.
The metric-map-fails-loudly default is the right call, but the case that bit us with shareable skill files was the quiet one: a source renames a field and quietly repoints it at slightly different semantics, so the map still resolves structurally and the run succeeds with subtly wrong numbers. Do skills pin a schema version or hash the source's field definitions, or is the metric map matched purely by name?
Databox
@dipankar_sarkar Fair challenge, and that's the harder failure mode to catch. Right now the matching is by name against the metric map, not a pinned schema version or field hash, so you're right that a silent rename-with-different-semantics could slip through where an actual schema break wouldn't. That's a real gap, not something we've solved yet. Appreciate you flagging the specific mechanism, it's useful for thinking about what schema validation should actually check for.
Databox
The insight behind the Skills Marketplace didn't come from a product roadmap. It came from watching how people actually used Databox MCP after we launched it.
Teams connected their data sources, started asking questions in Claude, and got good answers. But then they hit a wall. They had to figure out which questions to ask, how to structure the analysis, what to compare against. The blank canvas problem. The AI was capable - but the analytical workflow still lived in someone's head.
That's what a skill solves. It's not a prompt. It's a complete analytical workflow - the questions, the structure, the context, the output format - packaged into a single file. The expertise is baked in. Someone who has never written an analytics report in their life can trigger a GA4 skill and get back something a senior analyst would be proud to send.
We built the first skills ourselves to prove the format. Then something interesting happened - people started asking if they could build their own and share them. That's when we knew the marketplace was the right move.
What excites me most about where this goes:
Skills as a new content format. A blog post explains how to analyze paid ads. A skill just does it. The knowledge is executable, not just readable.
The quality bar is structural. Every skill ships with a metric map, setup guide, and troubleshooting flow. The output is grounded in live data. It's not a prompt someone shared on Twitter - it's something that holds up against real numbers.
It compounds. Every new skill added to the marketplace is another reason to connect another data source. More sources, richer context, better output. The whole system gets more useful over time.
We're launching with 20+ skills and workflows today. That number is going to grow fast.
The bigger vision: a world where every repeating analytical task a marketing or ops team runs has a skill that does it automatically - grounded in real data, structured for sharing, trusted enough to act on. We think that world is closer than most people expect.
I tested the cross-channel paid ads skill across a few accounts with Google Ads, Facebook Ads, and LinkedIn Ads all connected. The output covered ROAS by platform, CTR trends, budget pacing, and anomaly flags - all in one pass, in under a minute. No manual data pull. No prompt engineering. Just the analysis a paid ads analyst would have produced, ready to send. For anyone doing this kind of reporting weekly across multiple clients, the time difference is not incremental. It's the entire task removed.
Databox
@tadej_kelc This is exactly the use case we built it for, agencies and in-house teams running the same report across multiple clients or accounts every week. Glad it held up under real conditions with live accounts connected. Curious how it compared to your usual workflow time-wise, even a rough before/after would be great to hear.
Databox
@tadej_kelc Coming from a financial analyst, this is a pretty awesome endorsement. This skill might be removing barriers for non-ad experts to do analysis. Not sure ad people are ready for this level of scrutiny! :-)
Databox
I want to give a proper shout out to the partners who have built skills. Their skills are really well written and extremely practical.
These partners have spent years (if not decades) doing this work by hand for themselves and clients. They've now encoded that experience and judgement into these skills for the rest of us to run.
It's extremely valuable and generous of them!
A few worth checking out...
Rick Kranz (AI Marketing Labs, https://ai-marketinglabs.com) has built a whole library of analytics skills. His Sales Pulse skill connects to your CRM through Databox, pulls 14+ pipeline, activity, and funnel metrics, compares two rolling four-week periods, and surfaces the 3-4 findings that actually matter, like deal concentration risk and coverage ratio, before it shows you a single chart. It's pipeline interpretation, not another dashboard. https://databox.com/skills-marketplace/skill/sales-pulse-dashboard
Manav Mehra (Wagman Digital, https://wagmandigital.com) spent 18 years in finance before building his Quick Financial Health Check. It pulls P&L, balance sheet, cash, and AR/AP through Databox and outputs a monthly financial dashboard for services firms: cash flow projection, collections aging, missing-invoice detection. If a number isn't there, it shows a visible data gap instead of inventing one. That discipline is the whole point. https://databox.com/skills-marketplace/skill/quick-financial-health-check
Vasundhara Gupta (Atidiv, https://atidiv.com) built Morning Marketing Summary for people who don't want to open dashboards at all. It reviews your marketing data every morning and emails or Slacks a plain-English briefing under 250 words: what happened, what matters, and the 2-4 moves to make. It learns your channels and targets on the first run and tailors every summary after that. https://databox.com/skills-marketplace/workflow/morning-marketing-summary
Keith Gutierrez (Modgility, https://modgility.com) built Content Performance MAX, which turns any URL into a ready-to-review SEO metadata rewrite in about two minutes. It uses GA4 and Search Console data in Databox to decide which pages are even worth optimizing first, then drafts new titles, H1s, and meta descriptions with a human approval gate before anything touches your CMS. https://databox.com/skills-marketplace/skill/content-performance-max-starter
Marc Woodland (Winbox, https://www.winbox.co) is a LinkedIn Ads specialist who co-hosts the LinkedIn Ads Insider podcast, and he built Monthly LinkedIn Ads Analytics for the thing every team struggles with: a great workflow one person builds shouldn't die on their laptop. His skill plugs the Databox MCP straight into Claude, pulls your LinkedIn Ads campaign data into the chat with no exports or spreadsheets, and produces a clean one-page monthly report benchmarked on reach, relevance, engagement, and pipeline. https://databox.com/skills-marketplace/skill/monthly-linkedin-ads-analytics
Laszlo Fazakas (Arcanian, https://arcanian.com) built Set Baseline to kill a trap I see constantly: treating last month as a baseline. For a seasonal business, December is not a fair reference for January. His skill pulls enough real history to see the seasonal pattern, then states the baseline as a seasonality-adjusted range with the math and a confidence level, so "did it work?" finally has something honest to compare against. https://databox.com/skills-marketplace/skill/set-baseline
Today's launch is a really important milestone. But, the marketplace is bound to get better every time someone with real expertise adds a skill.
So, if you've built something on the Databox MCP, or want to, reply here and we'll get you in
Databox
@pc4media Adding my thanks here too. Rick, Manav, Vasundhara, Keith, Laszlo - genuinely appreciate you putting your expertise into something the rest of the community gets to benefit from. That's the kind of contribution that makes a marketplace actually useful instead of just a list of integrations. If anyone else wants to submit a skill, or has one they wish existed, you can do that here: https://databox.com/skills-marketplace/build
Databox
@pc4media The partner skills are exactly what we hoped the marketplace would become: real expertise, encoded and shared. Rick, Manav, Vasundhara, Keith, and Laszlo are setting the bar high and it will be super exciting to see more builders join in 🙌
@pc4media A little humbled to be on this list, honestly. What strikes me reading through them is how different they look on the surface — finance health, LinkedIn Ads, SEO metadata, a morning briefing — and how much they share underneath: people handing over judgement they spent years (decades, in some cases) earning by hand, so the rest of us don't have to learn it the slow way.
That's the generous part. And the quiet thing about a marketplace like this is that it compounds — every skill someone adds makes the next person's work a little sharper. Grateful to be building alongside Rick, Manav, Vasundhara, Keith and Marc, and to whoever ships the next one. Together this gets good fast.
Really useful idea. Do the skills handle schema changes in connected sources without breaking reports?
Databox
@dhiraj_patel5 Good question. Each skill expects a specific metric map (listed on its product page), so if a source's schema shifts in a way that breaks that mapping, the run would fail or flag it rather than silently return bad data - that's what the troubleshooting flow included in every download is for. Since the skills are editable files, you can also adjust the metric map yourself if your setup changes. We're not auto-detecting schema drift yet, but it's a fair ask for the roadmap.
I have tried a few analytics AI tools before, but many still expect me to figure out the right prompts. I like that your approach starts with proven workflows instead. That feels like a more natural way for teams to adopt AI without a lot of trial and error.
Databox
@gwendolyn_kira That's the gap we were trying to close. Most AI analytics tools put the burden on you to know what to ask for, which assumes you're already the expert. Skills flip that - the expertise is baked into the file, so you start from a proven structure instead of a blank prompt and trial and error. Glad that distinction came through.