Porter Metrics

1. 1

   You sign in with the Google account you use for Looker Studio and connect a data source by OAuth. Access is read-only and revocable from your account at any time. The unit that matters for both pricing and configuration is the account: one Meta ad account, one GA4 property, one Shopify store, one HubSpot portal. Ten Google Business Profile locations count as a single account.

2. 2

   You then build a connection, which is where Porter differs from a plain connector. A connection can hold several accounts of the same source (a dozen client Meta ad accounts consolidated into one table) or several different sources at once. When you select more than one platform, Porter maps the shared fields automatically: dates at day, week, month, quarter, and year, campaign names matched against GA4 UTM campaigns, ad set and creative names, country and region, and spend, impressions, clicks, conversions, and revenue. No join is written and no schema is defined.

3. 3

   For sources with hard API rate limits (HubSpot, Shopify, ActiveCampaign, Amazon Seller, large Google Business Profile accounts) Porter stores and backfills the data into its own managed BigQuery layer rather than querying live, which is what stops those sources from timing out or failing partway through a long date range. Everything else is queried live against the source API, with storage available as an opt-in for speed and stability.

4. 4

   You send the connection to a destination. Looker Studio is the primary one and Porter connects through the Data Studio API, so it can create the report with the data already loaded rather than making you configure a community connector by hand; roughly 60 templates install in one click this way. The same connection also writes to Google Sheets, Power BI, and a BigQuery dataset in your own Google Cloud project, all included in the same subscription with no per-destination fee.

5. 5

   On top of that sit the analysis features. A natural-language custom field builder lets you describe a metric ("cost per lead excluding campaigns with brand in the name") and turns it into a formula or regex that is written as a column into the connection, and custom dimensions can parse a campaign naming convention into separate Country, Platform, and Funnel Stage columns. A visual, node-based alert builder runs a query on a schedule or on a threshold, optionally passes it through an AI summary prompt, and posts to Slack or out through Zapier. An MCP server at mcp.portermetrics.com exposes the same connected accounts read-only to Claude, ChatGPT, and Gemini at no extra charge.

Charged per connected data source account, where an account means one ad account, one analytics property, one business profile, one store, or one CRM portal, with ten Google Business Profile locations counting as a single account. There are no seat charges, no per-destination charges, and no premium connectors: every destination, blending, templates, custom fields, alerts, and the MCP server are included on every plan including the free one. The per-account rate falls as the account count rises. Annual billing gives two months free, roughly 16.6 percent.

Value assessment: The honest way to price Porter is per connector against the two products it actually competes with, and the answer changes direction depending on how many accounts you have. For a single business connecting Meta Ads, Google Ads, and GA4 once each, Porter costs $39.99 a month and includes every destination, unlimited users, blending, and templates. Supermetrics' Starter plan at 49 euros a month covers three data sources but gives you one destination, one user, and three accounts per source, so the moment a second person or a second destination is involved Porter is meaningfully cheaper for the same coverage. Windsor.ai's Basic plan at $23 a month monthly ($19 annually) covers three data source types and up to seventy-five accounts, which at three accounts is cheaper than Porter and comes with a far larger connector catalogue. Push the count up and the picture inverts twice. At twenty accounts Porter is a flat $100 a month, against Windsor Standard at $118 ($99 annually) for seven source types and Supermetrics Growth at 199 euros for seven sources, two users, and seven accounts per source: Porter wins clearly. At sixty accounts across five platforms Porter charges $4 per account, about $240 a month, while Windsor Standard covers seventy-five accounts across seven source types for $99 annually, and Porter loses badly. The rule that falls out of this is specific: Porter is the best value in this category between roughly three and twenty-five connected accounts across a handful of platforms, and it is the wrong pricing model for an agency running dozens of accounts of the same source. Add the free forever plan, which no direct competitor of this shape matches in permanence, and the entry risk is close to zero either way.

Setup time

Under an hour for a first working report. Sign in with the Google account you already use for Looker Studio, authorise a source by OAuth, select the accounts, and install a template through the Data Studio API with data already loaded. A blended multi-source connection with custom fields and a naming-convention parser is more like half a day. A BigQuery destination writing into your own Google Cloud project takes longer only because of Google's side of the setup.

Learning curve

Low for a marketer, and deliberately so, because the modelling interfaces are natural language and visual builders rather than SQL. The concepts that take time are what counts as an account for billing, when a source should be switched to stored rather than live, and how the automatic blend has decided to map fields when two platforms disagree about what a campaign is. Analysts used to writing their own joins will find the absence of a SQL editor the hardest adjustment.

Onboarding

Self-serve with a 14-day unlimited trial and no card, but the notable part is that the assisted path is free and unmetered. Onboarding calls are available on any plan including the free one, Porter will build or fix reports for you at no charge, and annual customers are offered custom reporting and training. A direct Slack channel with the developers and founders is part of the standard offer rather than an enterprise perk.

Migration notes

Moving in from Supermetrics, Windsor, or a hand-built connector means rebuilding data sources inside existing Looker Studio reports, since field names will not match one to one and calculated fields referencing old field names need rewriting. Budget an afternoon per non-trivial report rather than a swap. Moving out is easier than most: if you use the BigQuery destination, the data already sits in your own Google Cloud project and survives cancellation. The genuinely hard exit is not the data, it is the automatic blending, because a connection Porter blended for you becomes a set of joins someone has to write by hand elsewhere.

Platforms

Web application at app.portermetrics.comLooker Studio (Data Studio) connector via the Data Studio APIGoogle Sheets destinationPower BI connectorBigQuery destination into a customer-owned Google Cloud projectHosted MCP server at mcp.portermetrics.com for Claude, ChatGPT, Gemini, and other MCP clients

API

No public REST API for extracting data from Porter itself is published. The programmatic routes are the BigQuery destination, where data lands in tables in your own Google Cloud project and can be queried by anything, the Google Sheets destination, Zapier for routing alert output to around 5,000 apps, and the read-only MCP server for AI clients. Custom integration development exists but only for customers on plans from about $1,000 a month.

Compliance

No SOC 2 Type 2 certification as of August 2026; the company states it is expected during 2026The company's own documentation states the product is not GDPR compliant, and flags this specifically for European teams with data residency requirementsNDAs offered in place of certificationRead-only OAuth access to source platforms, revocable by the customer at any timeCustomer-owned BigQuery project offered as the data-control answer, so warehoused data sits in infrastructure the customer holds

Data residency

No published data residency or region-selection controls. The one mechanism that gives a customer control over where warehoused data lives is the BigQuery destination, since the dataset sits in a Google Cloud project the customer owns and configures.

SSO

Not published. Authentication is through a Google account, and no SAML single sign-on, role-based permission model, or audit log is advertised.

Security notes

Source connections are read-only and the MCP server is explicitly read-only, so the product cannot alter campaigns or spend. Porter states it does not transform, sample, or reinterpret source data, meaning reported figures should reconcile against each platform's own manager. The gap for a security reviewer is not the access model, it is the absence of a certification, an audit log, an SSO story, and a permissions model, all of which are reasonable to expect from a ten-person company and unreasonable to ignore if your procurement process asks for them.

Channels

Live chat on every plan including the free forever planEmail supportFree onboarding and setup calls with no minimum planUnlimited calls, including calls booked directly with the foundersA direct Slack channel with developers and foundersFree report building, fixing, and Looker Studio training

Documentation

A help centre with setup tutorials, a field-level documentation reference listing every metric and dimension exposed by each connector, a large tutorial library covering Looker Studio and BigQuery technique as well as the product itself, and a YouTube channel. The site is published in English, Spanish, and Portuguese.

Community

No large independent user forum, which is the usual consequence of a small vendor. The practical substitute is unusually direct access to the team, and reviewers do note that support response times can be slow despite the generous channels on offer, which is what a ten-person company looks like from the outside.

Founded

2020

Headquarters

Lewes, Delaware, United States, with the team based in Manizales, Caldas, Colombia

Ownership

Independent, angel and accelerator backed

Founders

Juan Jose Bello, Mateo Gomez, Carlos Alexis Gomez

Employees

~10 to 16 (2026)

Funding

About US$400,000 from angel investors and the Techstars Boulder 2021 cohort, per the company's own March 2026 statement. Earlier Colombian press reported the Techstars investment at US$120,000 alongside a US$250,000 angel round. No institutional venture round has been announced.

Porter Metrics is the connector layer, not the dashboard, and every sensible judgement about it follows from that. It does not compete with AgencyAnalytics, DashThis, or Whatagraph; it competes with Supermetrics and Windsor.ai, and inside that fight its case is packaging rather than breadth. One pricing dimension, every destination and every feature on every plan, automatic cross-source blending that removes the join nobody wants to write, managed warehousing for the APIs that break live connectors, a BigQuery dataset you own, and a permanent free tier make it the best value in this category for a business connecting roughly three to twenty-five accounts across mainstream marketing platforms. The disqualifiers are equally clear and the company states most of them itself: 26 connectors with no Salesforce, no Tableau or Snowflake destination, no SOC 2 Type 2 and an admission that the product is not GDPR compliant, no SQL editor, and a per-account model that turns expensive for an agency running dozens of accounts of the same source. Check your source list and your compliance checklist before anything else, because either one can end the evaluation in a minute. If both pass, start on the free plan, connect three accounts, and see whether a blended table that needed a data engineer last month arrives on its own this month.

Written by the SaaSTracker editorial team. Awards, when shown, are judged against the published criteria in our methodology.