Marketing automation tools, sorted by the job they do

There is no single best marketing automation tool. Pick an email or CRM platform to store contacts and send messages, add a connector such as n8n, Make or Zapier when data has to move between tools, and use AI steps for reading and writing tasks. The tables below compare the main options on facts the vendors published as of November 2026.

Updated · Published by BlackCard

Connectors: n8n, Make and Zapier

A connector watches for an event in one tool and acts in another. The three best-known products differ mainly in hosting, billing unit and how technical they feel. Integration counts are the vendors' own figures.

Pointn8nMakeZapier
What it isPlatform for workflows and AI agents with a node-based editorVisual builder for workflows and AI agentsAutomation platform that also includes Tables, Forms and Agents
Integrations500+ native, plus any API over HTTPMore than 3,000 appsMore than 9,000 apps
HostingSelf-hosted, or n8n's cloud with data stored in Frankfurt, GermanyVendor cloud, with EU or US zone chosen at sign-upVendor cloud
Billing unitWorkflow executions, regardless of the number of stepsCreditsTasks
Human approvalDocumented human review for AI tool calls via Slack, Teams, Gmail and other channelsMake's own Human in the Loop app is tied to the Enterprise plan; otherwise split the work into two scenariosBuilt-in Human in the Loop step on paid plans
Source codeSource available under the Sustainable Use LicenseClosedClosed
Current plansn8n pricingMake pricingZapier pricing

Email and CRM platforms

These tools hold your contact list and send the messages. Each has its own automation builder, which is enough for welcome emails, tagging and simple follow-up sequences.

PlatformWhat it isGood fitWorth knowing
HubSpot Marketing HubMarketing software built on HubSpot's CRMTeams that want marketing, sales and service data in one CRMHas a free plan with forms, email and live chat; advanced automation sits in the higher tiers
BrevoEmail and SMS marketing, CRM and automation platformSmall businesses that want email, SMS and a simple CRM in one accountFrench company, known as Sendinblue until 2023
MailchimpEmail and SMS marketing platformNewsletters and simple automated flowsAn Intuit brand; SMS is an add-on in selected countries
ActiveCampaignEmail, SMS and WhatsApp marketing with automation and AI agentsBusinesses that want detailed, branching automationsThe automation builder is the centre of the product
KlaviyoB2C CRM for email, SMS, WhatsApp and pushOnline shops, especially on ShopifyBuilt around order and browsing data
MailerLiteEmail campaigns, automations, landing pages and sign-up formsSolo operators and small teams that want a simple toolFree sign-up available
PipedriveSales CRM with workflow automation and built-in AISales-led teams that need the pipeline first and marketing secondConnects to email tools through its marketplace

AI steps and enrichment

An AI step sends text to a language model and uses the answer in the next step. There are three ways to add one: use the AI features built into your platform (HubSpot calls its assistant Breeze, and Mailchimp, ActiveCampaign and Klaviyo ship their own), call a model from a connector, or use the agent features that n8n, Make and Zapier now include.

Enrichment services add company data to a bare email address. Clay combines many data providers with an AI research agent. Dropcontact finds and cleans B2B contact data and says it does not keep or resell a database of personal data. Only enrich where you have a lawful basis for holding the extra data.

  • Drafting: first replies, follow-ups, variants of a social post. Always reviewed by a person.
  • Classification: enquiry type, language, urgency, product interest.
  • Extraction: company name, budget or deadline from free text.
  • Scoring: a fit score with a written reason, checked against real outcomes.
  • Summarising: call notes, long email threads, survey answers.

How to choose in practice

Work from the workflow to the tool, never the other way round.

  1. Write down the one workflow you want and every tool it touches.
  2. Check whether your current email or CRM platform can do all of it. If it can, stop there.
  3. If not, list the triggers and actions you need and confirm that the connector offers them for your tools.
  4. Estimate runs per month and steps per run, then read each pricing page with those numbers in hand.
  5. Check hosting region, the data processing agreement and the list of sub-processors.
  6. Build the workflow on a free plan or trial before you commit to an annual contract.

Data location and GDPR points to check

For European businesses the hosting question often decides the shortlist. n8n states on its pricing page that hosted plans store data in the EU, on servers in Frankfurt, and it can also be self-hosted. Make lets you choose an EU or US zone, and its help centre notes that AI features may process data outside the chosen zone. For any US-based vendor, check whether it is certified under the EU-US Data Privacy Framework or offers standard contractual clauses, and sign its data processing agreement. This is general information, not legal advice.

Common questions

Is n8n free?

The self-hosted Community Edition has no licence fee when used for your own internal business purposes under n8n's Sustainable Use License. You still pay for a server and for the time to maintain it. The hosted cloud plans are paid and billed by monthly workflow executions. See the n8n pricing page for current plans.

Which connector is easiest for a non-technical marketer?

Zapier and Make are aimed at people who do not write code and run entirely in the browser, so most marketers get a first automation working quickly on either. n8n gives more control over data, hosting and AI agents and suits teams with some technical confidence. The honest test is to build the same small workflow on two of them using free plans or trials.

Do I need a separate AI tool for marketing automation?

Usually not at the start. Most email and CRM platforms now include AI drafting, and all three major connectors can call a language model as one step in a workflow. A separate tool makes sense for a specific job the others do poorly, such as enrichment across many data providers.