The ultimate glossary of terms for integration platforms (iPaaS)

Introduction

If you have ever sat through a software demo and nodded along while secretly wondering what on earth a "webhook" is, this glossary is for you.

Workflow automation — connecting the apps your business already uses so they talk to each other without anyone copying and pasting — is one of the highest-impact investments an SME can make. The problem is that the industry wraps perfectly simple ideas in jargon that feels designed to keep non-technical people out. We think that is nonsense. Every concept in automation has a real-world equivalent you already understand: conveyor belts, post rooms, filing cabinets, telephone switchboards. Once you see the parallel, the jargon stops being intimidating and starts being useful.

This glossary covers every major term you will encounter across the four leading automation platforms — Zapier, Make.com, n8n, and Power Automate — translated into language that a Managing Director, Operations Director, or Finance Director can read over a cup of tea and actually enjoy. We have included real-world metaphors drawn from engineering, construction, consulting, and manufacturing, plus concrete examples using the apps you probably already have open in another tab: HubSpot, Xero, Google Sheets, Slack, and dozens more.

Whether you are evaluating platforms, briefing your team, or just trying to decode a proposal from an automation consultancy, bookmark this page. It is the reference guide we wish someone had handed us when we started.

Tip: New to automation? Start with these five foundational terms: Trigger, Action, Workflow, Webhook, and iPaaS. Once you understand these, every other concept in this glossary clicks into place.

How to Use This Glossary

Every entry follows the same format so you can scan quickly:

1. Plain-language definition — one or two sentences, no jargon.
2. Real-world metaphor — a comparison drawn from factories, construction sites, workshops, or everyday office life.
3. App example — a concrete scenario using tools you actually use, like HubSpot, Xero, Slack, or Google Sheets.
4. Platform comparison table — where relevant, a simple table showing what each platform calls the same concept.
5. Learn more links — direct links to vendor documentation if you want to dig deeper.

Terms are organised A–Z. Use your browser's search (Ctrl+F or Cmd+F) to jump straight to a term, or simply scroll through — you might discover something useful along the way.

How the Core Concepts Connect

flowchart LR
    A["Trigger"] -->|starts| B["Action"]
    B --> C{"Condition"}
    C -->|"Yes"| D["Action 2"]
    C -->|"No"| E["Action 3"]
    D --> F["Workflow Complete"]
    E --> F

A–Z Glossary

A

Action

What it means: An action is a single task that your automation performs after it has been triggered. It is the "then do this" part of an "if this happens, then do this" instruction.

Think of it like this: If a trigger is someone pressing the doorbell, the action is you getting up to answer the door. One trigger can set off a chain of several actions — hear the bell, stand up, walk to the door, open it, greet the visitor.

App example: A new contact is added to HubSpot (that is the trigger). The action could be: create a matching contact in Xero so your finance team has their billing details ready before anyone lifts a finger.

PlatformWhat they call itZapierActionMake.comModule (action type)n8nNode (action type)Power AutomateAction

Aggregator

Aggregator

What it means: An aggregator takes multiple separate items and combines them into a single bundle. It is the opposite of an iterator — instead of splitting things apart, it gathers them together.

Think of it like this: Imagine a packing station at the end of a production line. Individual items come down the belt one by one. The aggregator is the worker who groups them into a single box, seals it, and sends it on as one package.

App example: You receive ten separate line items from a WooCommerce order. An aggregator bundles them into a single formatted list, which you then paste into one Slack message to your warehouse team — instead of sending ten separate messages.

Learn more: Aggregators in Make.com

AI Agent

What it means: An AI agent is a step in your workflow that uses artificial intelligence to make decisions or generate content, rather than following a fixed rule. Unlike a simple "if/then" condition, an agent can interpret context, draft responses, or classify information using large language models.

Think of it like this: Most automation steps are like vending machines — put the right coin in, get a predictable result. An AI agent is more like a capable office temp: you give them a brief, some context, and they use judgement to produce a sensible output. They will not always give an identical answer, but they will give a relevant one.

App example: A customer submits a support request via Typeform. An AI agent reads the message, decides whether it is a billing query, a technical issue, or a general enquiry, and routes it to the correct channel in Microsoft Teams — without anyone writing complicated rules for every possible question.

Learn more: Zapier Agents · Make.com AI Agents · n8n Advanced AI

API (Application Programming Interface)

What it means: An API is the standardised way that two software applications talk to each other. When your automation platform connects to HubSpot or Xero, it is using their API to send and receive data.

Think of it like this: An API is like a standardised plug socket. You do not need to know how the electrical grid works — you just need to know that your appliance's plug fits the socket. The API is the socket: a published, predictable way for any authorised tool to connect and exchange information.

App example: When Zapier pulls your latest invoices from Xero, it is sending a request to Xero's API saying, "Give me all invoices created today." Xero's API sends back the data in a structured format that Zapier can understand and pass along to the next step.

Approval

What it means: An approval step pauses a workflow and waits for a person to review something and say "yes" or "no" before the automation continues. It is the human checkpoint in an otherwise automatic process.

Think of it like this: On a construction site, certain works cannot proceed until a supervisor signs off on the inspection. The approval step is that signature — the workflow stops, waits for the green light, and only then moves to the next phase.

App example: An employee submits a purchase request via Microsoft Forms. Power Automate sends an approval email to their manager. If the manager approves, the workflow creates a purchase order in QuickBooks. If they reject it, the employee gets a notification explaining why.

Learn more: Power Automate Approvals · Zapier Human in the Loop

Array

What it means: An array is simply a list of items stored together. Instead of one value (like a single email address), an array holds multiple values in a specific order (like a list of ten email addresses).

Think of it like this: A single letter in a pigeonhole is a value. An array is the entire row of pigeonholes — each slot holds one item, and they are numbered so you can find any specific one.

App example: When you pull all line items from a Shopify order, you get an array — a list of products, each with its own name, quantity, and price. Your workflow can then loop through the array to create a separate row in Google Sheets for each product.

Authentication

What it means: Authentication is how your automation platform proves to an app that it has permission to access your data. It is the digital equivalent of showing your ID badge at the front desk.

Think of it like this: Imagine arriving at a client's secure building. You show your ID at reception, they check you are on the visitor list, and they give you a pass that works until 5pm. Authentication works the same way — your platform presents credentials, the app checks them, and access is granted (often with an expiry time).

App example: When you connect your Gmail account to Make.com, Google asks you to sign in and confirm that Make.com is allowed to read and send emails on your behalf. That handshake is the authentication step. Once complete, Make.com can use your Gmail in any scenario without asking again (until the permission expires or is revoked).

Automation

What it means: Automation is the practice of making tasks happen by themselves, without a person manually performing each step. In the context of business software, it means connecting your apps so that data flows between them and actions are taken automatically based on rules you set.

Think of it like this: Before automation, your office processes are like a hand-written ledger — every entry, every cross-reference, every update done by a person. Automation is upgrading to a system where the ledger updates itself: figures flow in, calculations happen, and reports appear — all without anyone picking up a pen.

App example: Every Friday at 5pm, your automation pulls all completed tasks from Asana, summarises them in a formatted list, and posts a weekly progress update to a Slack channel. No one has to write the update — it just appears.

B

Binary Data

What it means: Binary data is any file content that is not plain text — images, PDFs, spreadsheets, audio files, and so on. When your workflow handles an attachment or generates a document, it is working with binary data.

Think of it like this: Text data is like a typed memo — you can read every word at a glance. Binary data is like a sealed parcel — you know something is in there, but you need the right tool (an image viewer, a PDF reader) to open it and make sense of the contents.

App example: A client uploads a signed contract as a PDF via DocuSign. Your automation picks up that binary file and stores it in a specific Google Drive folder, then sends a Slack notification to your accounts team with a link to the document.

Blueprint (Make.com)

What it means: A blueprint in Make.com is the complete design of a scenario (workflow) saved as a portable file. You can export a blueprint and import it into another Make.com account, giving someone an exact copy of your automation without rebuilding it from scratch.

Think of it like this: Blueprints work exactly like architectural blueprints. An architect draws up the plans for a building once, and any qualified builder can use those plans to construct an identical structure somewhere else. A Make.com blueprint is the automation equivalent — the complete set of instructions for building a workflow.

App example: Your automation partner builds a scenario that syncs HubSpot contacts with Mailchimp subscribers. They export the blueprint and send it to you. You import it into your own Make.com account, connect your own HubSpot and Mailchimp accounts, and you are up and running in minutes.

Learn more: Make.com Scenarios

Branch / Branching

What it means: Branching is when a workflow splits into two or more separate paths based on a condition. Each branch handles a different situation, and the workflow follows whichever path matches the data it is processing.

Think of it like this: Picture a railway junction where a points operator directs trains onto different tracks depending on their destination. Branching is the same idea — your data arrives at a decision point, and the workflow sends it down the correct track based on the rules you have set.

App example: A new lead fills in a form on your website. If their company size is over 50 employees, the workflow sends their details to your enterprise sales team in HubSpot. If it is under 50, it routes them to your SME team instead. Same trigger, different paths.

Learn more: Zapier Paths · Make.com Routers · n8n IF Node · Power Automate Conditions

Bundle (Make.com)

What it means: A bundle is Make.com's term for a single packet of data as it moves through a scenario. Every time a module processes information, it handles it as one or more bundles — each bundle is one "unit" of data (one email, one row, one record).

Think of it like this: If your workflow is a postal sorting office, each bundle is a single letter. The sorting office processes letters one at a time — reading the address, deciding where it goes, and sending it on. Bundles work the same way: each one is handled individually as it passes through each module.

App example: You have a Make.com scenario that watches a Google Sheets spreadsheet for new rows. Each new row arrives as a separate bundle. If three rows are added simultaneously, three bundles are processed — each one creating its own contact in HubSpot.

Learn more: Make.com Modules

C

Child Flow

What it means: A child flow is a smaller, self-contained workflow that gets called by a larger "parent" workflow. It lets you break complex processes into manageable, reusable pieces.

Think of it like this: When a general contractor runs a building project, they do not do the electrics, plumbing, and plastering themselves — they call in specialist subcontractors. Each subcontractor (child flow) does their specific job and reports back. The main contractor (parent flow) coordinates the whole operation.

App example: Your main Power Automate flow processes new customer orders. When it reaches the invoicing step, it calls a child flow that handles all the invoicing logic — pulling prices, applying discounts, generating the invoice in QuickBooks, and emailing it to the customer. The parent flow waits for the child flow to finish, then moves on to updating the CRM.

Learn more: Power Automate Child Flows · Zapier Sub-Zaps

Cloud Flow

What it means: A cloud flow is a workflow that runs entirely in the cloud — on the platform's servers — without needing any software installed on your computer. Most automations you will build are cloud flows.

Think of it like this: The difference between a cloud flow and a desktop flow is like the difference between a cloud-based accounting system (Xero) and software installed on a single office PC. The cloud version runs from anywhere, for anyone, at any time — no one needs to leave a particular computer switched on.

App example: A cloud flow in Power Automate watches for new emails in your shared Outlook mailbox. When one arrives from a VIP client, it posts an alert to your Microsoft Teams channel and creates a follow-up task in Planner — all happening on Microsoft's servers, whether your computer is on or off.

Community Node

What it means: A community node is a pre-built integration or function for n8n that was created by a member of the community rather than the n8n team. It extends what n8n can connect to, without you needing to build anything from scratch.

Think of it like this: The n8n team provides a well-stocked standard toolkit. Community nodes are like specialist tools shared by fellow tradespeople — someone has already built the jig you need, tested it, and made it available for others to use.

App example: n8n does not have a built-in node for a niche project management tool your company uses, but a community member has published one. You install the community node, and suddenly n8n can read and write data to that tool just like it does with Slack or Google Sheets.

Learn more: n8n Community Nodes

Concatenate

What it means: Concatenation is simply joining two or more pieces of text together into one. It is a common task in automation when you need to combine separate data fields into a single string.

Think of it like this: You have a first name on one label and a surname on another. Concatenation is sticking them together onto one label with a space in between. Simple as that.

App example: Your Typeform collects a first name and a last name in separate fields. Before creating a contact in HubSpot, you concatenate them — "Jane" + " " + "Smith" = "Jane Smith" — so the full name appears in a single field.

Concurrency

What it means: Concurrency is how many instances of your workflow can run at the same time. If your automation is triggered by incoming emails and five emails arrive simultaneously, concurrency determines whether all five are processed at once or queued up one after another.

Think of it like this: If your workshop has one lathe, you can only turn one part at a time — that is single concurrency. Install five lathes and you can work on five parts simultaneously. Higher concurrency means more tasks handled in parallel.

App example: Your Make.com scenario processes incoming orders from Shopify. During a flash sale, 50 orders arrive in a minute. With concurrency set to 1, they queue up and process one by one. Increase the concurrency, and multiple orders are processed simultaneously — getting invoices into Xero faster.

Condition

What it means: A condition is a yes-or-no test that your workflow checks before deciding what to do next. It is the "if" in "if this, then that" — a decision point based on your data.

Think of it like this: A condition is like a quality control checkpoint on a factory line. The inspector checks: "Does this part meet the specification?" If yes, it moves to packing. If no, it is sent back for rework. The condition sets the criteria; the workflow follows the appropriate path.

App example: An invoice arrives from Xero. Your workflow checks: "Is the total greater than £5,000?" If yes, it sends an approval request to the Finance Director via email. If no, it automatically marks it as approved and logs it in Google Sheets.

Learn more: Power Automate Conditions · n8n IF Node

Connection

What it means: A connection is the saved link between your automation platform and one of your apps. Once you create a connection (by signing in and granting permission), the platform can access that app's data in any workflow without asking you to sign in again.

Think of it like this: Setting up a connection is like giving a trusted colleague a key to your filing cabinet. You do it once, they can access what they need whenever required, and you can revoke the key at any time.

App example: You create a connection between Make.com and your Xero account. Now, any scenario you build can read invoices, create contacts, or update payments in Xero — all using that single, saved connection.

Connector

What it means: A connector is a pre-built integration that lets your automation platform work with a specific app. Each platform comes with hundreds (or thousands) of connectors — one for HubSpot, one for Slack, one for Google Sheets, and so on. The connector handles all the technical details of talking to that app.

Think of it like this: A connector is like a travel adapter. You are plugging your UK appliance (your workflow) into a foreign socket (the app). The adapter handles the differences in voltage and pin layout so everything works seamlessly. Each country (app) needs its own adapter (connector).

App example: Zapier's Xero connector lets you trigger workflows when a new invoice is created, search for existing contacts, or create new payments — all without you needing to know anything about Xero's technical interface.

PlatformWhat they call itZapierApp / IntegrationMake.comApp / Modulen8nNodePower AutomateConnector

Copilot

What it means: Copilot is Microsoft's AI assistant built into Power Automate (and other Microsoft products). It lets you describe what you want your automation to do in plain English, and it builds or suggests the flow for you.

Think of it like this: Copilot is like having a knowledgeable junior draughtsperson sitting beside you. You describe what you need — "I want a flow that emails me when a SharePoint file changes" — and they sketch out the design for you to review and approve.

App example: Instead of building a Power Automate flow step by step, you type: "When a new row is added to my Excel spreadsheet in SharePoint, send me a Teams message with the row details." Copilot generates the flow structure, and you just review it, connect your accounts, and switch it on.

Cron

What it means: Cron is a way of defining a recurring schedule using a standardised format. When you see "cron expression" or "cron schedule," it simply means a precise way of telling your automation exactly when to run — every Monday at 9am, every hour, the first of every month, and so on.

Think of it like this: A cron schedule is like programming a central heating timer — you set exactly when it should switch on and off, and it repeats that pattern faithfully. The format is just a compact way of writing "every weekday at 8:30am" or "the 1st and 15th of each month at noon."

App example: You set up an n8n workflow with a cron trigger to run every weekday at 6pm. It pulls all tasks marked "Done" in Asana that day and posts a summary to Slack — your automated daily wrap-up.

Custom Connector

What it means: A custom connector lets you connect your automation platform to an app that does not already have a pre-built connector. You provide the technical details of the app's interface, and the platform creates a reusable connector from them.

Think of it like this: If a standard connector is a travel adapter you buy off the shelf, a custom connector is one that an electrician wires up specially because the socket is non-standard. It does the same job — it just requires a bit more setup because the app is not one of the common ones.

App example: Your company uses a specialist quoting tool that Zapier does not natively support. You (or your automation partner) build a custom connector that talks to the quoting tool's interface, and now Zapier can pull quotes, update statuses, and trigger workflows based on quote approvals — just like it does with any built-in app.

Learn more: Power Automate Custom Connectors

D

Data Store

What it means: A data store is a simple, built-in database provided by your automation platform. It lets you save, retrieve, and update small amounts of data without needing a separate database or spreadsheet.

Think of it like this: A data store is like a shared logbook kept on the factory floor. Anyone on the team can look up the last entry, add a new one, or update an existing record. It is not a full warehouse management system — it is a quick-reference log for the information your workflows need between runs.

App example: You build a Make.com scenario that processes leads from a Typeform. Before creating a contact in HubSpot, it checks the data store to see if that email address has already been processed. If it has, it skips the duplicate. If not, it adds the email to the data store and creates the contact.

PlatformWhat they call itZapierZapier TablesMake.comData Storen8nNo built-in equivalent (use an external database or Google Sheets)Power AutomateDataverse / SharePoint list

Learn more: Make.com Data Stores · Zapier Tables

Date Formatting

What it means: Date formatting is the process of converting a date from one format to another. Different apps store dates in different ways — "25/02/2026," "2026-02-25," "February 25, 2026" — and your automation needs to translate between them so each app gets the format it expects.

Think of it like this: Date formatting is like converting between metric and imperial measurements on an engineering drawing. The underlying value is the same — a 25mm bolt is still a 25mm bolt — but the drawing for the American client needs it labelled in inches. You are translating the presentation, not the substance.

App example: HubSpot records a deal's close date as "2026-02-25." Your workflow sends a congratulatory Slack message to the sales team, but you want it to read "25 February 2026" instead of the raw format. A date formatting step handles the conversion.

Learn more: Zapier Date/Time Formatting · n8n Luxon Date Handling

Dead Letter Queue

What it means: A dead letter queue is where failed messages or tasks end up when your automation cannot process them after repeated attempts. Instead of losing the data, it is parked in a holding area so you can investigate and fix the problem later.

Think of it like this: In a postal sorting office, if a letter cannot be delivered — wrong address, damaged label, no recipient found — it goes to the "dead letter office." Staff there investigate, try to work out where it should go, and either redirect it or return it to sender. A dead letter queue works the same way for automation failures.

App example: Your workflow tries to create a contact in Salesforce, but the API is temporarily unavailable. After three retry attempts, the failed request is sent to a dead letter queue. You receive a daily summary email listing all queued failures so you can reprocess them once Salesforce is back online.

Default Value

What it means: A default value is a pre-set fallback that your workflow uses when an expected piece of data is missing. Instead of failing because a field is empty, the automation fills in a sensible substitute.

Think of it like this: On a construction site, if the specified paint colour has not been confirmed yet, the decorator uses the agreed default — brilliant white — so work is not held up. When the final colour choice comes through, it can be changed. The default keeps things moving.

App example: Your Typeform collects a "Company Name" field, but some respondents leave it blank. Instead of creating a HubSpot contact with an empty company, your workflow fills in "Not provided" as the default value — making it easy to filter and chase up later.

Delay

What it means: A delay step pauses your workflow for a set amount of time before continuing to the next action. It is a deliberate wait built into the process.

Think of it like this: After pouring concrete, you do not start building on it immediately — you wait for it to cure. A delay is that planned waiting period: the automation knows it needs to pause before moving to the next step.

App example: A customer makes a purchase through Shopify. Your Zap waits 7 days (the delay), then sends a follow-up email via Mailchimp asking for a review. The delay ensures the customer has had time to receive and use the product before you ask for feedback.

Learn more: Zapier Delays · n8n Wait Node

Desktop Flow

What it means: A desktop flow is an automation that runs on your local computer rather than in the cloud. It can interact with applications that do not have cloud-based connections — clicking buttons, filling in forms, copying data between windows — by mimicking what a person would do with a mouse and keyboard.

Think of it like this: A cloud flow is like an automated email system — it works behind the scenes on a server. A desktop flow is more like a robotic arm on a factory floor — it physically operates the machinery (your desktop software) in the same way a person would, just faster and without fatigue.

App example: Your accounts team manually enters supplier invoices from a legacy desktop application into QuickBooks every week. A Power Automate desktop flow opens the legacy app, reads the invoice data, switches to QuickBooks, and enters it automatically — turning two hours of data entry into a five-minute unattended process.

Learn more: Power Automate Desktop Flows

DLP (Data Loss Prevention)

What it means: Data Loss Prevention is a set of policies that control which types of data your automations are allowed to move between apps. It prevents sensitive information from accidentally flowing to the wrong place.

Think of it like this: DLP is like the security classification system on a construction project. Certain documents are marked "Confidential — Client Only" and cannot leave the secure project room. DLP policies work the same way — they define which data is allowed to travel where, and block anything that breaks the rules.

App example: Your IT team sets a DLP policy in Power Automate that prevents any flow from sending customer financial data from your Dynamics 365 system to non-Microsoft apps. If someone tries to build a flow that pushes invoice data to a personal Dropbox, the policy blocks it before the flow can even run.

Learn more: Power Automate DLP

Dynamic Content

What it means: Dynamic content is data from a previous step in your workflow that you insert into a later step. Instead of typing a fixed value, you reference a value that changes each time the workflow runs — a customer's name, an invoice total, a date.

Think of it like this: A mail merge is the perfect analogy. You write one template letter, but each printed copy is personalised with the recipient's name, address, and account details pulled from your database. Dynamic content is the automation equivalent — one workflow template, but the data changes every time it runs.

App example: Your workflow sends a Slack message when a new deal closes in HubSpot. The message template says: "New deal closed: [Deal Name] worth [Deal Amount] with [Company Name]." Each bracketed placeholder is dynamic content — pulled from that specific deal's data in HubSpot.

E

Environment

What it means: An environment is a separate workspace within your automation platform where you can build, test, and manage workflows independently. Most organisations use at least two: a development (or test) environment and a production (live) environment.

Think of it like this: In manufacturing, you have a prototyping workshop and a production floor. You test and refine new designs in the workshop before they go anywhere near the production line. Environments work the same way — you build and test in one space, and only deploy to the live environment once everything is proven.

App example: Your automation team builds a new flow in their development environment in Power Automate. They test it thoroughly with sample data. Once they are confident it works correctly, they promote it to the production environment where it runs on real customer data.

Learn more: Power Automate Managed Environments · n8n Source Control

Error Handler

What it means: An error handler is a specific step or module in your workflow that defines what happens when something goes wrong. Instead of the entire automation crashing, the error handler catches the problem and follows a predefined recovery plan.

Think of it like this: On a construction site, you do not just hope the generator works — you have a documented procedure for what happens if it fails: switch to backup power, notify the site manager, log the incident. An error handler is that documented procedure for your automation.

App example: Your Make.com scenario creates invoices in Xero. You add an error handler that says: "If Xero returns an error, send me a Slack message with the error details, save the failed data to a Google Sheets for manual review, and continue processing the remaining invoices." The workflow does not grind to a halt — it handles the problem gracefully.

Learn more: Make.com Error Handling · n8n Error Handling · Power Automate Error Handling

Error Handling

What it means: Error handling is the overall strategy and set of techniques you use to deal with failures in your automations. It covers everything from retrying failed steps, to sending alerts, to logging problems for later review.

Think of it like this: Error handling is your entire site safety plan — not just one fire extinguisher, but the complete set of procedures: fire drills, evacuation routes, first aid protocols, and incident reporting. It is how you prepare for the things that will go wrong, because in any system, something eventually will.

App example: Your automation syncs contacts between HubSpot and Salesforce every hour. Your error handling strategy includes: automatic retries if the connection times out, an email alert to you if it fails three times in a row, and a fallback that logs the unsynced contacts in Google Sheets so no data is lost — even on a bad day.

Learn more: Make.com Error Handling · n8n Error Handling · Zapier Autoreplay · Power Automate Error Handling

Execution

What it means: An execution is a single run of your workflow from start to finish. Each time a workflow is triggered and processes data through its steps, that counts as one execution. Most platforms track executions because they relate to your usage and billing.

Think of it like this: If your workflow is a recipe, an execution is one time you actually cook the dish. You can have the recipe written down forever, but it only counts when you get the ingredients out and make it. Each time you cook = one execution.

App example: Your Zap triggers every time a new row is added to a Google Sheets. On Monday, 15 new rows are added, so the Zap runs 15 times — that is 15 executions. Each execution consumes tasks (Zapier) or operations (Make.com) from your plan's monthly allowance.

PlatformWhat they call itZapierZap run / TaskMake.comScenario execution / Operationsn8nWorkflow executionPower AutomateFlow run

Expression

What it means: An expression is a short formula or piece of logic you write inside a workflow step to transform or calculate data on the fly. It is similar to writing a formula in a spreadsheet cell.

Think of it like this: A spreadsheet formula like =A1*1.2 (to add 20% VAT) is an expression. You are not writing a software programme — you are just telling the system to do a quick calculation or transformation with the data it already has.

App example: In Power Automate, you write an expression to convert a price from dollars to pounds: multiply the "price" field by 0.79. Or in n8n, you use an expression to extract just the domain name from an email address — turning "jane@acmeconsulting.co.uk" into "acmeconsulting.co.uk."

Learn more: n8n Expressions · Power Automate Expression Functions

F

Field

What it means: A field is a single piece of information within a record — like "Email Address," "Company Name," or "Invoice Total." When you map data between apps, you are connecting fields from one app to fields in another.

Think of it like this: If a database record is a row in a filing cabinet, a field is one of the labelled boxes on the form in that file — "Name," "Address," "Date of Birth." Each field holds one specific type of information.

App example: When your workflow creates a new contact in HubSpot from a Typeform submission, you map the "Email" field from Typeform to the "Email address" field in HubSpot, the "Company" field to the "Company name" field, and so on.

Filter

What it means: A filter is a rule that lets your workflow continue only when specific conditions are met. If the data does not match the filter criteria, the workflow stops for that particular item — it is simply ignored, not processed.

Think of it like this: A filter is like a sieve in a gravel plant. Everything gets poured in, but only the pieces of the right size pass through. The rest are set aside. Your workflow's filter does the same with data — only records that meet your criteria get processed.

App example: Your Zap triggers on every new email to your support inbox. But you only want to create a HubSpot ticket for emails from clients on your VIP list. A filter step checks whether the sender's domain matches your VIP list — if it does, the Zap continues. If not, it stops there.

Learn more: Zapier Filters · Make.com Filters

Find and Replace

What it means: Find and replace lets you search for a specific piece of text within your data and swap it for something else — exactly like the find-and-replace function in a word processor.

Think of it like this: You have printed 200 copies of a proposal and realise the client's name is misspelt throughout. Find and replace is like having an assistant who goes through every copy, finds every instance of "Acme Consulting" and replaces it with "Acme Consultancy" — instantly and without missing any.

App example: Data from your web form includes phone numbers in different formats — some with spaces, some with dashes, some with brackets. A find-and-replace step strips out all the non-numeric characters so every phone number is in a clean, consistent format before it hits your CRM.

Formatter (Zapier)

What it means: Formatter is Zapier's built-in tool for transforming text, numbers, and dates within a Zap. It can capitalise text, split names, convert currencies, reformat dates, and perform dozens of other common data clean-up tasks — all without any code.

Think of it like this: The Formatter is like a prep station in a commercial kitchen. Before ingredients go into the dish (your destination app), they need to be washed, peeled, chopped, and measured. The Formatter does that prep work for your data.

App example: A lead's name arrives from Typeform as "jane smith" (all lowercase). Zapier's Formatter capitalises it to "Jane Smith" before creating the contact in HubSpot — so your CRM stays clean and professional.

Learn more: Zapier Formatter Text

Function

What it means: A function is a built-in tool within your automation platform that performs a specific calculation or transformation on your data — like rounding a number, converting text to uppercase, or calculating a date 30 days from now.

Think of it like this: Functions are like the preset buttons on a scientific calculator. You do not need to know the maths behind them — you press "√" and it gives you the square root. Each function takes an input, does something specific to it, and gives you the result.

App example: In Make.com, you use a function to calculate the due date of an invoice: take the invoice creation date and add 30 days. The function handles all the calendar logic (month lengths, leap years) so you do not have to think about it.

Learn more: Make.com Custom Functions · Power Automate Expression Functions

G

GraphQL

What it means: GraphQL is a way of requesting data from an app where you specify exactly what information you want, rather than getting a fixed bundle of everything the app offers. It is an alternative to the more common REST approach.

Think of it like this: Ordering with a REST API is like ordering a set meal at a restaurant — you get the starter, main, and dessert whether you want them all or not. GraphQL is like ordering à la carte — you pick precisely the dishes (data fields) you want, nothing more, nothing less. It is more efficient when you only need specific information.

App example: Your workflow needs only the name and email of contacts from a project management tool. With GraphQL, you request just those two fields — instead of downloading the entire contact record with 50 fields you do not need. This makes the workflow faster and uses less of your platform's data allowance.

H

HTTP Request

What it means: An HTTP request is the most basic way to send or fetch data from any app or service on the internet. When there is no pre-built connector for the app you need, an HTTP request step lets you communicate with it directly.

Think of it like this: Pre-built connectors are like using a telephone directory — you look up the number, dial, and the call is handled for you. An HTTP request is like dialling the number manually and speaking the right language yourself. It gives you more flexibility, but you need to know the correct "phone number" (URL) and what to say (the request format).

App example: Your company uses a niche quoting tool that Zapier does not have a built-in connector for, but the tool has a published web interface. You use Zapier's HTTP Request action (via webhooks) to send a request to the quoting tool, pull back the latest quote data, and feed it into a Google Sheets for your sales team.

Learn more: n8n HTTP Request Node · Zapier Webhooks

I

Instant Trigger

What it means: An instant trigger fires your workflow the moment an event happens — in real time, with no delay. Unlike polling triggers that check for changes periodically, instant triggers are notified immediately.

Think of it like this: An instant trigger is like a smoke alarm — it goes off the moment it detects smoke. A polling trigger is more like a security guard who walks the building every 15 minutes to check for problems. Both catch the issue, but the smoke alarm catches it straight away.

App example: When someone fills in your Typeform, an instant trigger fires your Zap immediately — creating a HubSpot contact and sending a Slack notification within seconds. There is no delay because Typeform actively notifies Zapier the moment the form is submitted.

PlatformWhat they call itZapierInstant triggerMake.comInstant module / Webhookn8nWebhook triggerPower AutomateAutomated trigger (instant)

iPaaS (Integration Platform as a Service)

What it means: iPaaS is the category of cloud-based tools that let you connect different software applications and automate workflows between them — without writing code or managing servers. Zapier, Make.com, n8n, and Power Automate are all iPaaS platforms.

Think of it like this: An iPaaS is like a telephone exchange for your business software. In the old days, a switchboard operator connected calls between any two phones in the building. An iPaaS does the same thing for your apps — it sits in the middle and routes data between them so they can "talk" to each other.

App example: You use HubSpot for sales, Xero for accounting, and Slack for communication. An iPaaS platform connects all three: when a deal is marked as won in HubSpot, the iPaaS creates an invoice in Xero and posts a celebration message to Slack — all automatically, all without anyone switching between tabs.

Iterator

What it means: An iterator takes a collection of items (a list or an array) and processes them one at a time. It is the opposite of an aggregator — instead of grouping things together, it separates them out for individual handling.

Think of it like this: An iterator is a quality inspector on a production line. A box of 50 components arrives. Instead of inspecting the box as a whole, the inspector takes out each component one by one, examines it, and decides what happens to it individually.

App example: A Shopify order contains five line items (products). An iterator in Make.com splits the order into five individual items. Each one is then checked: if the item is in stock, it is sent to the warehouse system; if not, it triggers a back-order email to the supplier.

Learn more: Make.com Iterators

J

JSON (JavaScript Object Notation)

What it means: JSON is the standard format that most apps use to send data to each other. It is a structured way of organising information using labels and values — like a very tidy inventory list that machines can read easily.

Think of it like this: JSON is like a well-organised parts list on an engineering drawing. Each part has a label (name, quantity, material, cost) and a value next to it. Everything is structured consistently so anyone reading it knows exactly what each item is. JSON does the same thing for data — it gives every piece of information a clear label and value.

App example: When Zapier requests a contact from HubSpot, HubSpot sends back the data in JSON format — something like: name is "Jane Smith," email is "jane@acme.co.uk," company is "Acme Consulting." You never need to read the raw JSON yourself — your automation platform translates it into friendly, clickable fields.

L

LangChain

What it means: LangChain is an open-source framework that helps developers build applications powered by large language models (the technology behind tools like ChatGPT). In the context of automation, some platforms use LangChain under the bonnet to enable AI-powered workflow steps.

Think of it like this: If an AI model is the engine, LangChain is the gearbox and drive shaft — the components that translate the engine's raw power into useful motion (practical applications). You do not interact with LangChain directly any more than you interact with your car's gearbox — you just benefit from what it enables.

App example: n8n uses LangChain-compatible nodes to let you build AI-powered workflows. You could create a workflow where a customer support email comes in, an AI node (powered by LangChain under the bonnet) reads the email, generates a draft reply, and saves it in Gmail for your team to review before sending.

Learn more: n8n Advanced AI

Line Items

What it means: Line items are the individual entries within a larger record — typically the separate products or services listed on an invoice, quote, or order. Handling line items in automation often requires loops or iterators because each item needs to be processed individually.

Think of it like this: A purchase order is a single document, but the line items are the individual rows within it — 100 metres of cable, 50 junction boxes, 12 distribution boards. Each line item has its own description, quantity, and price. Your automation often needs to handle each one separately.

App example: A customer places an order in Shopify with three products. Your Make.com scenario needs to create a matching invoice in Xero with those same three line items. An iterator splits the products out, and your scenario builds each line item in the Xero invoice individually.

Lookup Table

What it means: A lookup table is a simple reference list that your workflow uses to translate one value into another. It works like a two-column table: look up the input in the left column, and get the corresponding output from the right column.

Think of it like this: A lookup table is like a price list pinned to the workshop wall. When someone says "Part number A-4521," you run your finger across the row and find the price is £12.50. The lookup table does the same thing automatically — it matches an input to a stored value.

App example: Your web form collects a country name ("United Kingdom"), but your CRM expects a country code ("GB"). A lookup table in your Zap maps "United Kingdom" to "GB," "United States" to "US," and so on — converting every submission instantly.

Loop

What it means: A loop is a workflow step that repeats an action multiple times — once for each item in a list. It processes item one, moves to item two, then item three, and so on until the list is complete.

Think of it like this: A loop is like a bricklayer building a wall. They lay one brick, move to the next position, lay another brick, move again — repeating the same action until the row is complete, then starting the next row. Each repetition follows the same pattern with a different brick (data item).

App example: You have a Google Sheets with 30 overdue invoices. Your workflow loops through each row: for every overdue invoice, it sends a personalised chasing email via Gmail with the client's name, invoice number, and amount owed. Thirty rows, thirty emails — all handled by one loop.

PlatformWhat they call itZapierLooping by ZapierMake.comIterator / Repeatern8nLoop Over Items (Split In Batches)Power AutomateApply to Each

Learn more: Zapier Looping · Make.com Iterators · n8n Loop Over Items · Power Automate Apply to Each

M

Managed Environment

What it means: A managed environment is an administrative setting (primarily in Power Automate and the broader Power Platform) that gives your IT team extra controls over who can build automations, which connectors they can use, and how workflows are monitored. It is governance for your automation practice.

Think of it like this: A managed environment is like a controlled workspace in a factory — only authorised personnel can access certain machinery, safety protocols are enforced automatically, and supervisors can see everything that is happening on the floor. It does not stop work from getting done; it just makes sure it is done safely and to standard.

App example: Your IT team enables managed environments in Power Platform. Now, every new flow built by someone in the marketing team is automatically checked against company policies, certain high-risk connectors are blocked, and the IT team gets a dashboard showing all active flows and their owners.

Learn more: Power Platform Managed Environments

Manual Trigger

What it means: A manual trigger starts a workflow only when a person deliberately clicks a button or runs it by hand. Unlike automatic triggers that fire on their own, a manual trigger gives you full control over when the workflow runs.

Think of it like this: An automatic trigger is like a sprinkler system — it kicks in by itself when it detects heat. A manual trigger is a fire extinguisher — someone has to walk over, pick it up, and use it deliberately. Both are useful; the right choice depends on the situation.

App example: You have built a Power Automate flow that generates a monthly management report by pulling data from Xero and HubSpot into a formatted Excel spreadsheet. You do not want it running automatically — you want to review the timing and context first. So you set a manual trigger and run it yourself on the last Friday of each month.

Mapping

What it means: Mapping is the process of connecting a data field from one step of your workflow to a field in another step. When you "map" the email address from your Typeform submission to the email field in HubSpot, you are telling the automation where each piece of data should go.

Think of it like this: Mapping is like wiring a patch panel in a recording studio. You connect output 1 from the mixing desk to input 1 on the recorder, output 2 to input 2, and so on. Each wire (mapping) tells a specific piece of data where it should travel to and plug in.

App example: Your workflow creates a new Xero contact from a HubSpot deal. You map HubSpot's "Company name" field to Xero's "Contact name" field, HubSpot's "Email" to Xero's "Email address," and HubSpot's "Phone" to Xero's "Phone number." Each mapping ensures the right data lands in the right place.

Markup Conversion

What it means: Markup conversion is the process of converting text from one formatting style to another — for example, turning Markdown (the simple formatting used in tools like Notion and Slack) into HTML (the formatting used on websites and in emails), or vice versa.

Think of it like this: Markup conversion is like translating a hand-drawn engineering sketch into a formal CAD drawing. The content is the same — the same dimensions, the same specifications — but the format changes to suit the destination.

App example: Your team writes project updates in Notion using Markdown formatting. Your workflow needs to send those updates as a formatted HTML email via Gmail. A markup conversion step translates the Markdown headings, bold text, and bullet points into HTML so the email looks polished and professional.

MCP (Model Context Protocol)

What it means: Model Context Protocol is an emerging standard that allows AI models (like ChatGPT or Claude) to connect directly to external tools and data sources in a structured, secure way. In the context of automation, MCP lets AI agents access your business apps and data during a conversation or workflow.

Think of it like this: Without MCP, asking an AI a question about your business data is like phoning a consultant who has never seen your files — they can only give generic advice. With MCP, it is like that consultant having secure, read-only access to your filing cabinet — they can look up the specific information they need to give you a precise, relevant answer.

App example: An AI agent in Make.com uses MCP to connect to your HubSpot CRM. When a team member asks the AI chatbot, "What is the status of the Acme project?" the agent can look up the actual deal record in HubSpot and reply with current, accurate information — rather than guessing.

Learn more: Make.com MCP

N

Note: Same concept, different name. You will notice that Zapier, Make.com, n8n, and Power Automate often use completely different terms for the same thing — a workflow is a "Zap," a "Scenario," a "Workflow," or a "Flow" depending on which platform you use. The comparison tables throughout this glossary are your Rosetta Stone.

No-Code / Low-Code

What it means: No-code means you can build automations entirely using visual tools — dragging, dropping, clicking, and configuring — without writing any programming code. Low-code means mostly visual, but with the option to add small pieces of code for advanced tasks. Most iPaaS platforms are no-code for basic workflows and low-code for complex ones.

Think of it like this: No-code is like flat-pack furniture with an Allen key — the manufacturer has done the engineering, and you just follow the instructions and assemble it. Low-code is like the same flat-pack, but with a few custom modifications: you might need a drill for one step, but you do not need to be a carpenter. Full-code is building the furniture from raw timber.

App example: In Zapier, you can build a complete automation — "When I get an email in Gmail with an attachment, save it to Google Drive and notify me in Slack" — without writing a single line of code. That is no-code. In n8n, you might write a short expression to format a date in a specific way — that is low-code.

O

OAuth

What it means: OAuth is the industry-standard method that apps use to grant your automation platform secure access to your data without you having to share your password. When you "Connect your Google account" and a pop-up asks you to "Allow access," that is OAuth in action.

Think of it like this: OAuth is like a hotel key card system. Instead of giving the maintenance team a copy of your room key (your password), the hotel issues them a temporary access card that only works for specific rooms at specific times. If you check out (revoke access), the card stops working immediately. Your master key (password) is never shared.

App example: When you connect your Xero account to Make.com, you are redirected to Xero's website where you sign in and click "Allow." Xero gives Make.com a secure token (the key card) that lets it access your invoices and contacts. Your Xero password is never shared with Make.com.

Operation (Make.com)

What it means: An operation is Make.com's unit of measurement for usage and billing. Every time a module in your scenario successfully processes a bundle of data, that counts as one operation. Your Make.com plan includes a monthly allowance of operations.

Think of it like this: Operations are like the units on your electricity meter. Every time you switch on a light (run a module), the meter ticks over. At the end of the month, your bill reflects how many units you have used. Planning your scenarios efficiently is like being energy-efficient — you get more done without burning through your allowance.

App example: Your scenario has three modules: watch for new Google Sheets rows, create a HubSpot contact, and send a Slack notification. Each time a new row is processed, that is 3 operations (one per module). If 100 rows are added in a month, you use 300 operations.

Learn more: Make.com Credits and Operations

P

Parse

What it means: Parsing is the process of extracting structured information from unstructured or messy data. When your automation receives a block of text, an email, or a file and needs to pull out specific pieces of information (a name, a date, an amount), that extraction is called parsing.

Think of it like this: Parsing is what a post room worker does with incoming mail. A pile of envelopes arrives. The worker opens each one, reads the contents, identifies the sender, the reference number, and the department it should go to, and routes it accordingly. They are "parsing" the mail — extracting the useful information from the raw package.

App example: Your business receives purchase orders as PDF email attachments. Your automation parses each PDF — extracting the PO number, line items, quantities, and total amount — and feeds that structured data into your accounting system. What was a document becomes usable data.

Polling

What it means: Polling is when your automation platform repeatedly checks an app at regular intervals to see if anything new has happened. It is the opposite of instant triggers (which are notified immediately). With polling, there is always a short delay between the event occurring and the workflow noticing it.

Think of it like this: Polling is like a security guard doing rounds — they check each door every 15 minutes. If someone broke in at 10:01, the guard will not know until their 10:15 round. An instant trigger, by contrast, is a burglar alarm that alerts the guard the moment the door opens.

App example: Zapier's free plan polls most apps every 15 minutes. If a new row is added to your Google Sheets at 9:02, Zapier might not notice until 9:15. On a paid plan, polling is faster (every 1–2 minutes), but it is still not instant. For truly real-time responses, you need an instant trigger or webhook.

Process Mining

What it means: Process mining is a technique that analyses your existing business processes — by looking at data from your IT systems — to discover how work actually flows through your organisation. It reveals bottlenecks, inefficiencies, and variations from the intended process, helping you decide what to automate.

Think of it like this: Process mining is like attaching GPS trackers to every job on your factory floor. Instead of relying on how you think work flows through the building, you see how it actually moves — including all the unexpected detours, waiting times, and bottlenecks. The data tells you where to focus your improvement efforts.

App example: Before building any automations, your operations team uses Power Automate's process mining tools to analyse how purchase orders move through your organisation. They discover that 60% of the delay is caused by manual approval bottlenecks — so they know exactly where to focus their automation effort.

Learn more: Power Automate Process Mining

R

RAG (Retrieval-Augmented Generation)

What it means: Retrieval-Augmented Generation is a technique that makes AI responses more accurate by first searching a knowledge base for relevant information, then feeding that information to the AI model so it can generate a well-informed answer. Instead of the AI relying solely on its training data, it retrieves specific, up-to-date facts from your own documents.

Think of it like this: Imagine asking a consultant a question. Without RAG, they answer from memory — which might be outdated or incomplete. With RAG, they first pull the relevant files from your cabinet, read the latest data, and then give you an answer grounded in your actual business information. Same consultant, much better answer.

App example: You build an internal chatbot using n8n and an AI model. When a team member asks, "What is our refund policy for project delays?" the RAG system searches your company's policy documents (stored in Google Drive), retrieves the relevant paragraphs, and the AI generates a clear, accurate answer based on your actual policy — not a generic guess.

Learn more: n8n Advanced AI

Regex (Regular Expression)

What it means: A regular expression is a pattern-matching tool that lets you search for, extract, or validate text based on a specific pattern rather than an exact match. It is like a search function on steroids — instead of searching for "invoice," you can search for "any word followed by a number" or "any email address."

Think of it like this: Normal search is like looking for a specific book title in a library catalogue. Regex is like asking the librarian, "Find me any book where the title starts with 'Project' and ends with a year." It is pattern-based rather than exact-match.

App example: Incoming emails contain reference numbers in different formats — "REF-12345," "Ref: 12345," "Reference 12345." A regex pattern in your workflow can find and extract the number regardless of the format, so your automation can match it to the correct record in your CRM.

REST API

What it means: REST is the most common architectural style for APIs on the web. When you hear "REST API," it simply means a standardised way for apps to communicate over the internet using simple, well-understood methods for reading, creating, updating, and deleting data.

Think of it like this: REST is like a standardised set of hand signals on a construction site. "Come here," "Stop," "Lift," "Lower" — everyone understands them regardless of language. REST uses similarly standardised commands (GET, POST, PUT, DELETE) that any web application can understand. It is the common language of the internet.

App example: When your automation reads contacts from HubSpot, it sends a GET request to HubSpot's REST API. When it creates a new invoice in Xero, it sends a POST request. The method names tell the API exactly what kind of action is being performed — read or write, fetch or create.

Retry

What it means: A retry is when your workflow automatically attempts a failed step again after a short pause. If an app is temporarily unavailable or a network hiccup occurs, a retry can resolve the issue without any human intervention.

Think of it like this: A retry is like redialling a phone number that was engaged. You wait a moment, try again, and it usually connects on the second or third attempt. Your automation does the same thing — it waits, tries again, and only raises an alert if it fails multiple times.

App example: Your workflow tries to create an invoice in Xero, but Xero's servers are momentarily slow and return a timeout error. Instead of failing outright, the workflow waits 30 seconds and tries again. On the second attempt, it succeeds — and no one even knew there was a problem.

Router

What it means: A router is a step that splits your workflow into multiple paths, sending data down different routes based on conditions you define. Unlike a simple branch (which is typically one yes-or-no decision), a router can create many paths simultaneously.

Think of it like this: A router is like a postal sorting machine in a mail centre. Letters come in on one conveyor belt, and the machine reads the postcode on each one, directing it to the correct output chute — one for local, one for national, one for international. Each letter goes to the right destination without manual sorting.

App example: New leads arrive from your website form. A router in Make.com sends them down different paths based on industry: engineering leads go to one HubSpot pipeline, construction leads to another, and consulting leads to a third. Each path can have its own follow-up sequence tailored to that industry.

PlatformWhat they call itZapierPathsMake.comRoutern8nIF Node / Switch NodePower AutomateCondition / Switch

Learn more: Zapier Paths · Make.com Routers

RPA (Robotic Process Automation)

What it means: Robotic Process Automation is technology that automates repetitive tasks by mimicking what a human does at a computer — clicking, typing, copying, pasting, and navigating between applications. Unlike iPaaS (which connects apps through their built-in interfaces), RPA interacts with the screen the way a person would.

Think of it like this: iPaaS is like connecting two machines with a conveyor belt — data flows directly between them through purpose-built channels. RPA is more like a robotic arm sitting at a desk, using the same keyboard and mouse that a person would. It is especially useful for legacy software that does not have modern connection points.

App example: Your accounts team spends two hours a day copying data from a 20-year-old desktop system into a spreadsheet. The old system has no modern interface to connect to. A Power Automate desktop flow (RPA) opens the old application, navigates its menus, copies the data, and pastes it into Excel — exactly as a person would, but in minutes instead of hours.

Learn more: Power Automate Desktop Flows

S

Schedule

What it means: A schedule trigger starts your workflow at a specific, recurring time — every morning at 8am, every Monday at noon, the 1st of every month, and so on. The workflow runs on a clock rather than being triggered by an event.

Think of it like this: A scheduled trigger is your automation's alarm clock. Just as you set an alarm to wake you at 6:30 every weekday, you set a schedule to run your workflow at a fixed time. It fires whether or not anything new has happened — it is time-based, not event-based.

App example: Every Monday at 9am, a scheduled workflow pulls all open deals from HubSpot, calculates the total pipeline value, and posts a summary to the sales team's Slack channel. The weekly sales overview appears like clockwork — no one has to prepare it.

Learn more: Power Automate Scheduled Flows

Self-Hosting

What it means: Self-hosting means running the automation platform on your own servers (or your own cloud infrastructure) rather than using the vendor's hosted service. n8n is the most prominent iPaaS platform that offers self-hosting. It gives you full control over your data, security, and infrastructure — but requires technical expertise to set up and maintain.

Think of it like this: Using a vendor's hosted platform is like renting a serviced office — the landlord handles the utilities, maintenance, and security. Self-hosting is like owning your own building — you have total control, but you are also responsible for fixing the boiler when it breaks. Most businesses prefer the serviced office; some need the control that ownership provides.

App example: A financial services company with strict data residency requirements self-hosts n8n on their own UK-based servers. All automation data stays within their infrastructure, never passing through a third-party cloud service — meeting their regulatory obligations.

Learn more: n8n Self-Hosting

Split Text

What it means: Split text is an action that takes a single piece of text and divides it into separate parts based on a character you specify — a comma, a space, a line break, and so on.

Think of it like this: Splitting text is like cutting a length of timber at marked points. You have one long plank (the text), and you saw it at each pencil mark (the separator character) to create individual pieces.

App example: A form submission contains a "Products interested in" field with the value "CRM, Accounting, Payroll" — all in one text string separated by commas. A split text step turns this into three separate items: "CRM," "Accounting," and "Payroll." Your workflow can then create a separate HubSpot deal for each product interest.

SSO (Single Sign-On)

What it means: Single Sign-On is a system that lets your team members sign in once (usually to a central identity provider like Microsoft or Google) and then access multiple apps without signing in separately to each one. In the context of automation platforms, SSO means your team can access Make.com or Power Automate using their existing company login.

Think of it like this: Without SSO, entering your office building is like having a separate key for the front door, the lift, your office, the meeting room, and the kitchen. With SSO, you have one ID badge that opens every door. One credential, access to everything.

App example: Your company uses Microsoft 365 with SSO enabled. When a team member navigates to Power Automate, they are already signed in — no separate username or password needed. If they leave the company and their Microsoft account is deactivated, their access to Power Automate (and all other connected apps) is revoked automatically.

Step

What it means: A step is a single action or operation within a workflow. Each step does one thing — checking a condition, transforming data, creating a record, sending a message. Your complete workflow is a series of steps executed in order.

Think of it like this: A step in a workflow is like a step in a recipe. "Preheat the oven" is step one. "Mix the dry ingredients" is step two. Each step is a discrete instruction that must be completed before moving to the next. Your workflow follows the same logic.

App example: Your Zap has four steps: (1) Trigger — new form submission in Typeform. (2) Create a contact in HubSpot. (3) Create an invoice in Xero. (4) Send a Slack notification. Four steps, one workflow, fully automatic.

Sub-Workflow

What it means: A sub-workflow is a separate, self-contained workflow that can be called from within another workflow. It lets you break complex automations into smaller, reusable pieces — and use the same sub-workflow in multiple places without duplicating your work.

Think of it like this: A sub-workflow is like a subroutine in a factory process. The main assembly line calls upon specialist stations — welding, painting, testing — when needed. Each specialist station (sub-workflow) does its specific job and sends the result back to the main line.

App example: You have three different workflows that all need to send a formatted PDF report via email. Instead of building the "generate PDF and email" logic three times, you create it once as a sub-workflow. Each of your three main workflows calls the same sub-workflow when it needs to send a report.

PlatformWhat they call itZapierSub-ZapMake.comScenario called via webhookn8nSub-workflowPower AutomateChild flow

Learn more: Zapier Sub-Zaps · Power Automate Child Flows

T

Task (Zapier)

What it means: A task is Zapier's billing unit. Every time an action step in your Zap successfully runs, that counts as one task. (Triggers and filters do not count as tasks.) Your Zapier plan includes a monthly task allowance, and your usage is measured in tasks consumed.

Think of it like this: Tasks are like stamps in a franking machine. Every letter you send (action you run) uses one stamp. Your monthly plan comes with a set number of stamps, and when they run out, you either wait until next month or buy more.

App example: Your Zap has one trigger and two actions: new form submission (trigger — free), create a HubSpot contact (action — 1 task), send a Slack message (action — 1 task). Each form submission uses 2 tasks. If you get 500 submissions a month, that is 1,000 tasks.

Task Mining

What it means: Task mining uses software to observe and record how people perform tasks at their computers — which apps they open, what they click, what they type, and in what order. The data reveals repetitive patterns that are good candidates for automation.

Think of it like this: Task mining is like a time-and-motion study on your office processes. Instead of a clipboard-wielding analyst standing behind someone's chair, software quietly observes the task, records every step, and presents a clear map of how the work is actually done — including all the inefficiencies.

App example: Your operations team suspects that processing supplier invoices takes too long, but they are not sure exactly why. They use Power Automate's task mining to record the process across several team members. The analysis reveals that everyone manually copies the same three data fields between two apps — a perfect candidate for automation.

Learn more: Power Automate Task Mining

Timezone Conversion

What it means: Timezone conversion is the process of adjusting a date and time value from one timezone to another. When your automation pulls a timestamp from an app in one timezone and needs to display or store it in another, a timezone conversion step handles the adjustment.

Think of it like this: If you have a conference call scheduled for "3pm Eastern Time" and you are in London, you need to convert it to GMT to know when to dial in. Timezone conversion in automation does the same thing — it adjusts timestamps so that times are accurate for the people and systems that need them.

App example: Your Calendly meetings are booked in various timezones by international clients. Your workflow converts all meeting times to UK time before posting them to your team's Google Calendar and Slack channel — so no one has to do timezone maths in their head.

Trigger

What it means: A trigger is the event that starts your workflow. It is the "when this happens" part of the automation. Without a trigger, a workflow is just a plan sitting idle — the trigger is what sets everything in motion.

Think of it like this: A trigger is like the sensor on a conveyor belt that starts the sorting machine. As soon as a parcel hits the sensor, the machine springs into action — scanning, sorting, routing. Without the sensor (trigger), the machine just sits there. The trigger detects the event and says, "Go."

App example: "When a new contact is created in HubSpot" is a trigger. Once that event happens, the workflow kicks off — perhaps creating a matching record in Xero, sending a welcome email via Mailchimp, and posting a notification to Slack.

PlatformWhat they call itZapierTriggerMake.comTrigger modulen8nTrigger nodePower AutomateTrigger

Learn more: Power Automate Triggers

Trim

What it means: Trim removes extra spaces (or other unwanted characters) from the beginning and end of a text value. It is a common data clean-up step that prevents problems caused by invisible leading or trailing spaces.

Think of it like this: Trim is like sanding the rough edges off a piece of timber before it goes into the finished product. The extra spaces are the splinters — invisible at first glance, but they cause problems when you try to fit the piece into the next step (matching records, comparing values, sending emails).

App example: A customer types " jane@acme.co.uk " (with accidental spaces before and after) into your web form. Without trimming, your CRM might treat it as a different email address from "jane@acme.co.uk" — creating a duplicate contact. A trim step strips the spaces so the email matches correctly.

V

Variable

What it means: A variable is a named container that holds a value your workflow needs to remember and use later. You can set a variable at one point in your workflow and reference it at any later step — like jotting a note on a sticky pad and referring back to it later.

Think of it like this: A variable is like a labelled jar on a workshop shelf. You put something in the jar (a value), stick a label on it (the variable name), and any time you need that item later, you know exactly which jar to reach for.

App example: At the start of your Power Automate flow, you set a variable called "ApprovalStatus" to "Pending." As the flow progresses through different approval steps, it updates the variable to "Approved" or "Rejected." At the end, the flow checks the variable's final value to decide whether to proceed with the purchase order or notify the requester of the rejection.

Learn more: Power Automate Variables

Vector Store

What it means: A vector store is a specialised database designed to work with AI. It stores information as numerical representations (vectors) that allow an AI model to quickly search for and find the most relevant pieces of information — even when the exact words do not match. It powers features like semantic search and Retrieval-Augmented Generation (RAG).

Think of it like this: A traditional database is like a filing cabinet where you can only find documents by their exact label. A vector store is like having a librarian who understands meaning — you say, "I need something about late project deliveries," and they find the relevant document even though its title is "Schedule Overrun Procedures." It searches by meaning, not just keywords.

App example: You upload your company's entire operations manual, process documents, and policy PDFs into a vector store connected to n8n. When an employee asks your internal chatbot, "How do we handle a client complaint?" the vector store finds the most relevant paragraphs from across all your documents — even if none of them contain the exact phrase "client complaint."

W

Webhook

What it means: A webhook is a way for one app to send a real-time notification to your automation platform the moment something happens. Instead of your platform repeatedly checking the app for updates (polling), the app proactively pushes the data to you as soon as the event occurs.

Think of it like this: A webhook is like a site foreman radioing the office the moment concrete is poured — instant, proactive communication. Polling, by contrast, is like the office ringing the site every hour to ask, "Has the concrete been poured yet?" Webhooks are faster, more efficient, and put less strain on both sides.

App example: When a customer completes a payment through Stripe, Stripe sends a webhook to your Make.com scenario with the payment details — instantly. The scenario creates an invoice in Xero and sends a thank-you email via Gmail, all within seconds of the payment being made. No waiting, no polling, no delay.

PlatformWhat they call itZapierWebhooks by ZapierMake.comWebhooks / Custom Webhookn8nWebhook nodePower AutomateHTTP Request trigger

Learn more: Zapier Webhooks · Make.com Webhooks · n8n Webhook Node

Workflow

What it means: A workflow is the complete automated process from start to finish — the trigger, the actions, the conditions, and everything in between. It is the full set of instructions that tells your automation platform what to do, in what order, and under what circumstances.

Think of it like this: A workflow is like a manufacturing process map. It shows the complete journey of a product from raw materials to finished goods — every station, every inspection, every decision point. Your automation workflow does the same for data: it maps the complete journey from trigger to final output.

App example: Your complete workflow: (1) A new lead fills in a Typeform. (2) The workflow checks if their email already exists in HubSpot. (3) If yes, it updates the existing contact. If no, it creates a new one. (4) It assigns the lead to a sales rep based on their region. (5) It sends a welcome email via Mailchimp. (6) It posts a notification to Slack. That entire chain is one workflow.

PlatformWhat they call itZapierZapMake.comScenarion8nWorkflowPower AutomateFlow

Z

Zap (Zapier)

What it means: A Zap is Zapier's name for a workflow — the complete automated process you build, consisting of a trigger and one or more actions. Every automation you create in Zapier is called a Zap.

Think of it like this: "Zap" is simply Zapier's brand name for what other platforms call a scenario, workflow, or flow. It is like how some people say "Hoover" when they mean vacuum cleaner — different name, same fundamental thing.

App example: You build a Zap with three steps: (1) Trigger — a new row is added to a Google Sheets. (2) Action — create an invoice in Xero with the row's data. (3) Action — send a Slack message to the finance channel confirming the invoice was created. That complete chain is one Zap.

What Next?

You have made it through every term in the glossary — well done. You now understand the language of workflow automation better than most people in the industry. But knowing the terminology is just the starting point. The real value comes from applying it: connecting your HubSpot to your Xero, automating your onboarding process, or building the reporting workflow you have been doing manually every Friday afternoon.

If you are ready to stop reading about automation and start benefiting from it, flowmondo can help. We specialise in building practical, reliable automations for medium and midmarket businesses— the kind of businesses that need things to work properly, not just look clever.

Whether you need a simple two-step Zap or a multi-platform integration across your entire tech stack, we will design it, build it, test it, and hand it over with documentation your team can actually understand.