The monthly marketing report has a structural problem: by the time it arrives, the data is already between two weeks and a month old, the person receiving it cannot ask any further questions about it, and if they want to dig deeper into anything they have to wait for the next meeting or write to the analyst.

Los conversational AI dashboards solve that problem at its root: anyone on the team can ask questions in natural language about marketing data at any time and get immediate answers with visualisation included.

What sets a conversational dashboard apart from a traditional dashboard

A traditional dashboard (Looker Studio, Power BI, Tableau) is a set of predefined visualisations. It shows the metrics the analyst considered important when they set it up. If you want to see something different, you have to edit the dashboard or ask the analyst to do it.

A conversational dashboard answers questions that were not anticipated. "Which three campaigns generated the most qualified leads in the last quarter with a CPA below 40€?" is not a question that any predefined dashboard can answer directly. A conversational system connected to the right sources can.

The technology that makes this possible is a combination of NL2SQL (Natural Language to SQL, which converts the question into a database query) and language models for the interpretation and narrative of the result.

How it connects with marketing sources

The system needs access to data. The most common integrations for marketing:

  • GA4: via Google Analytics Data API. Traffic metrics, behaviour, conversions by source and channel.
  • Google Ads: via Google Ads API. Campaign, ad group, keyword, conversion and cost metrics.
  • Meta Ads: via Meta Marketing API. Reach, impressions, conversions, ROAS by campaign and ad set.
  • CRM: if the CRM has an API (HubSpot, Pipedrive, Salesforce), the system can query the pipeline, leads by status, and conversion rate by source.
  • Email marketing: if Klaviyo, ActiveCampaign or Brevo have an API, the system can query open rates, CTRs, and conversions by flow.

With those sources connected, the system can answer 80% of the questions a marketing team typically asks about its data.

Use cases that generate the most value

For the marketing director: "What was the ROI of each marketing channel in Q1?" or "Which month did we have the lowest CPA and what were we doing differently that month?"

For the advertising team: “Which ad groups are above the Target CPA this week?” or “Which ad format has the best CTR on Meta in the last 30 days?”

For the content team: "Which blog articles generated the most leads in the last 6months?" or "Which organic keywords are bringing high-quality traffic but we don’t have specific content?"

For general management: "How much does it cost us to acquire a new client per channel?" or "Are we on track to meet the quarterly leads target?"

Our Conversational Dashboard product implements this architecture connected to all marketing data sources.

Related reading