Here are a list of Syncari terms and Industry terms to help orient you with our platform and data jargon.
Syncari Terms
Synapse
| What is it | What does it mean | Why does it matter |
|---|---|---|
|
This is a stateful integration connection to external sources. See list of Synapses |
We use Synapses to connect and authenticate to external systems, discovering the entities and data types of those fields so that we can read or write information to those systems. | A Synapse standard lets us connect to data in any system, and enables us to read and write records according to the connected user. |
Entity
| What is it | What does it mean | Why does it matter |
|---|---|---|
| An object (table) that exists in Syncari | A way to represent an external table or a new object that represents a view of multiple tables | We use an entity for each thing in Syncari we are trying to represent, e.g. Account |
Field
| What is it | What does it mean | Why does it matter |
|---|---|---|
| A column header or field for an entity | A column of data from the original source or a column added in Syncari | Fields in Syncari may map to the source or destination or be used only within Syncari to keep track of information in the data hub |
Schema
| What is it | What does it mean | Why does it matter |
|---|---|---|
|
The way to describe the fields, data types, and relationship of an Entity. See Schema Studio |
This describes the data model for an entity, including relationships of data | A schema lets you know what data needs to be present in a record and how records relate to records in another entity. |
Pipeline
| What is it | What does it mean | Why does it matter |
|---|---|---|
|
The movement of data from a source Synapse into Syncari and potentially into a destination Synapse. These are all managed from Sync Studio |
Every time we detect a change, we run a series of entity and field pipelines based on the pipeline logic | This is the mechanism to move data through Syncari. Entity pipelines run on an entity; field pipelines run on a field for that entity; source pipelines run when data is received from a source; destination pipelines run after data is processed in Syncari. |
Left side vs Right side of the Pipeline
| What is it | What does it mean | Why does it matter |
|---|---|---|
|
Left Side includes source node(s) and all of the nodes and functions that are on the left side of the pipeline, before the central Syncari node Right Side includes destination node(s) and all of the nodes and functions that are on the right side of the pipeline, after the central Syncari node |
Data coming in from the left side will eventually land in the central Syncari node, and you'll be able to see this data in Data Studio. Data that gets sync'd out on the right side of the pipeline goes out to the destination(s) | This is an important distinction because Syncari sits in the middle and statefully syncing your data between source(s) from the left side and destination(s) on the right side of the data pipelines. |
Node
| What is it | What does it mean | Why does it matter |
|---|---|---|
| Nodes are pipeline blocks that either take data from a source, transform or make logical decisions based on data, trigger an action, or send data to a destination | A node is an element in a pipeline where things happen. Nodes can calculate data, ask external systems for information, or make logical decisions. | The node is the basic building block of a pipeline, accepting data from previous nodes, sending data to future nodes, or starting an external action. |
Function
| What is it | What does it mean | Why does it matter |
|---|---|---|
| A way to change information in a node in a standard way | We use functions to transform information, e.g. to capitalize a string, calculate an aggregate, count something, or look up data from another Syncari entity | Functions are the way we create logic in Syncari inside a node. |
Token
| What is it | What does it mean | Why does it matter |
|---|---|---|
|
A shorthand way to refer to information that’s calculated. |
We use a token, e.g. {{previous}}, as a value that will be replaced with a computed value when executed | Tokens let you write a standard value representing an eventual value that will be populated like a variable when run |
Action
| What is it | What does it mean | Why does it matter |
|---|---|---|
| A way to end a pipeline by taking an action on an external system | Actions let you send data to an external system or retrieve data or kick off a process, e.g. “Add to campaign” | Actions extend Syncari to other systems. With Custom Actions, write your own extensions to pipelines. |
Source
| What is it | What does it mean | Why does it matter |
|---|---|---|
| A way to add data from a Synapse to a Syncari entity | The source node connects Synapse fields mapped in a schema to a Syncari Entity or Syncari fields in an entity | A source node brings in data from an object in a Synapse into a Syncari entity where we can act on it. |
Destination
| What is it | What does it mean | Why does it matter |
|---|---|---|
| A way to update data in a Synapse from a Syncari entity | The destination node connects Synapse fields mapped in a schema in a Syncari entity | A destination node pushes data from a Syncari entity to an object in a Synapse |
Sync
| What is it | What does it mean | Why does it matter |
|---|---|---|
| Beginning a cycle to request data, receive data, run source pipelines, update core nodes, run destination pipelines | Based on the configuration of the authenticated user connected to the Synapse and any filters, bring data in from the Synapse, compare this data to the schema in Syncari, and trigger pipeline logic based on changes | This is the core functionality of Syncari - the ability to watch for changes, detect changes, and apply these changes to a stored copy of the data for an object. Then, based on the logic, apply that data to downstream systems if needed. |
Sync Direction
| What is it | What does it mean | Why does it matter |
|---|---|---|
| Syncs may be one way or bidirectional | A one way sync flows into Syncari from a system or out from Syncari to a system. A bidirectional sync flows both directions. | Designing a data flow requires knowing which systems need to be updated in which order and confirming connected users have the requisite permissions |
Incremental Sync
| What is it | What does it mean | Why does it matter |
|---|---|---|
| A sync cycle run since the last sync sycle, happening in an ongoing cadence | We check for changes that happened since the last sync cycle, approximately 1-5 minutes ago. When we find these changes, we compare against the data in Syncari entities and apply the changes | Reviewing changes is faster than copying the entire record, and allows us to catch up if the sources lag. |
Re-Sync
| What is it | What does it mean | Why does it matter |
|---|---|---|
| A sync cycle run for a subset of records, usually by the last time changed | Re-sync forces the pipelines to run since a watermark (date stamp) | We use this to catch up after a lapse in data or to re-run data through a pipeline |
Node configuration
| What is it | What does it mean | Why does it matter |
|---|---|---|
| Each node in every pipeline has specific configuration details. The configuration for a Source Node, a Destination Node, and a Syncari Node is distinct. Syncari Node configuration is optional. |
In an entity pipeline, a source node requires you to select an entity and a sync direction, along with optional timing of the sync. A source node in a field pipeline requires selecting a field from that entity. In an entity pipeline, a destination pipeline requires you to select an entity and a sync direction, along with a configuration on when to accept deletes (data governance on the synapse that controls deletes). A destination node in a field pipeline requires you to select the field you’re writing to, any default value, and whether you should send an empty value |
Node configuration is a critical step in setting up your pipelines to match the input and output logic you expect in Syncari. To make this clearer, you might write a flow chart to identify the source data, when you need it to flow through to be compared and/or transformed, and in what conditions you need it to be sent to the destination. Building a flow chart also gives you a handy checklist for quality testing during a live test or simulated test. |
Live Test
| What is it | What does it mean | Why does it matter |
|---|---|---|
| A live test runs records through a draft entity pipeline so that you can see the results of your work |
Using an ID from a source system, you select a record or records to run through a pipeline to confirm the logic is working as expected. This test also triggers other pipelines as needed. This test performs actual work on end systems (source and/or destination) so it needs to be used with caution |
Live tests give you a way to test your logic before the pipeline is published, or to investigate edge cases in your pipeline logic so that you can resolve and publish a new working pipeline. Live tests create records in the transaction log if records are changed from the test. |
Simulated Test
| What is it | What does it mean | Why does it matter |
|---|---|---|
| A simulated test lets you run an Entity or a Field pipeline with temporary data that you enter so that you can troubleshoot the logic of your pipeline |
Simulated tests show you the results of your logic without actually committing the changes to Syncari. Simulated tests are a diagnostic tool during development. |
Simulated tests do not make changes on connected Systems from Syncari. These tests do not create transactions that can be viewed in the logs. |
Dedupe
| What is it | What does it mean | Why does it matter |
|---|---|---|
| Finding duplicate records in set of records | Creating a set of conditions to identify records that logically mark duplicates. A duplicate record may be identified by a key (like email) or a composite (email + last name) | Selecting duplicates is the first step to deciding whether these duplicates should be merged. Duplicate rules vary by companies. |
Merge
| What is it | What does it mean | Why does it matter |
|---|---|---|
| The act of combining one or more duplicate records | Merging records means that you select among a set of duplicate records and decide which record will survive and which record will be deleted. You may also need to decide which field from each record has a higher priority. | Merging is destructive and must be done carefully to achieve an expected result. There are some situations where records should never be merged, so these records should not be allowed into the deduplication process. Use the “report only” merge process in Syncari to get a preview of merges without actually executing the action |
Data Authority
| What is it | What does it mean | Why does it matter |
|---|---|---|
| Data authority indicates the best source of data for an individual field or entity | If you have conflicting information, select a winner that will always have the best data | When you have multiple competing values in fields that pertain to the same object, you’ll want to know which system is agreed to be the System of Truth for that record (or field) |
Draft/Publish
| What is it | What does it mean | Why does it matter |
|---|---|---|
| When you start creating things in Syncari (a field, a schema, a pipeline, a Quick Start) they are marked with the status of “Draft”. When you publish them, they are marked “Published.” | We don’t have versioning in Syncari - we have “draft” which is the default status, and “published” which is the currently published version of a thing. | New items (an update to a pipeline, etc) need to move through draft state into published before they will work in a published system. |
Industry Terms
Data Synchronization
| What is it | What does it mean | Why does it matter |
|---|---|---|
| Data Synchronization is the task to match data in one system to like data in another system | When data changes in system 1, field 1, we want another system with similar data to get an updated record |
We create a “distributed source of truth” when we take the valid values from the system that owns the record and copy that data to another system that references that record but doesn’t own it. Example: an invoice system holds the legal name of a company. When that legal name is updated, a LegalName field in the CRM Account must be updated to have the new information |
ELT
| What is it | What does it mean | Why does it matter |
|---|---|---|
| Extract. Load. Transform | ETL transforms data on a separate processing server, while ELT transforms data within the data warehouse itself. ETL does not transfer raw data into the data warehouse, while ELT sends raw data directly to the data warehouse. |
Most data platforms or data systems use a version of this process to move data. Syncari does ELT because we do transforms on the datastore |
ETL
| What is it | What does it mean | Why does it matter |
|---|---|---|
| Extract. Transform. Load. (ETL) | This is a common acronym in the data industry meaning Extract (move data from one system to another system), Transform (change the data, correcting and enriching it needed), and Load (move it to another system) | Most data platforms or data systems use a version of this process to move data. |
Reverse ETL
| What is it | What does it mean | Why does it matter |
|---|---|---|
| A way to identify and send data from one system to another system on a scheduled or event-driven basis | Putting data back into frontline applications, usually from a database or data warehouse |
This is an easy way to provide basic integration to systems that aren’t connected directly Example: when a account becomes closed won, to update customer success system with a field marked “Gold Customer” |
Data Hub
| What is it | What does it mean | Why does it matter |
|---|---|---|
| The ability to create a copy of operational data in Syncari so that we get a single view of the data in our company | Because we sit at the center of many systems, we get a consolidated view of the customer 360 degree view | A data hub gives you the ability to see data across systems and make inferences based on the combined data |
Unification
| What is it | What does it mean | Why does it matter |
|---|---|---|
| Creating a combined record based on a common primary key | The ability to map records in one system to records in another system |
A unified record provides a single view across systems of all of the information related to that record Example: a “360 degree view” of the customer |
Data Orchestration
| What is it | What does it mean | Why does it matter |
|---|---|---|
| A term to describe the process for moving data in a structured way | Every time data arrives from a system, we handle it the same way. | When multiple integrations interact with the same entity in a system, you need to be able to control the changes so they combine and don’t overwrite each other |
Transformation
| What is it | What does it mean | Why does it matter |
|---|---|---|
| The process of changing data from one form into another | Data in one system does not always arrive in the correct format for another system. In addition, data may need to be enriched through external means or corrected if it is in error | Transforming data lets you standardize data, improve data integrity, make it more accurate, and change its shape for analytics, entity comparison, or other purposes. |
Automation
| What is it | What does it mean | Why does it matter |
|---|---|---|
| Setting up a process so that it happens automatically when the conditions are right | Once you automate the transformation of data or create workflows that run based on events, you don’t have to repeat the action manually | With automation, changes to records happen every time or cause an error. Automation improves your ability to follow through with a process. |
Standardization
| What is it | What does it mean | Why does it matter |
|---|---|---|
| The process of aligning data to an expected standard | Many times when you receive data it will be in the wrong format, or have unexpected errors. Having a standard to apply against this data makes it more likely that your automations will run successfully | Setting a minimum viable standard for data makes it possible to find the existing and future errors in your data and by extension, in your processes |
Data Governance
| What is it | What does it mean | Why does it matter |
|---|---|---|
| The process of managing the structure of data fields and values for operational data | Identifying which systems are the owners of which pieces of data, and putting controls in place to manage changes to this information | Data changes force other kinds of changes in the organization and need to be done in a controlled way |
Data Warehouse
| What is it | What does it mean | Why does it matter |
|---|---|---|
| A data warehouse is a data management system that stores current and historical data from multiple sources for easier insights and reporting. Data warehouses are typically used for business intelligence (BI), reporting and data analysis. | This is typically a central repository for a company's data. It's most used for building data models and BI reporting. | It's one central spot to house all a company's data. |
Data Lake
| What is it | What does it mean | Why does it matter |
|---|---|---|
| A data lake is a storage repository that holds raw data until it is needed for analytics applications. A data lake uses a flat architecture to store data, primarily in files or object storage. | It's a centralized repository that allows for the storage and analysis of large volumes of structured, semi-structured, and unstructured data. It is designed to hold raw, unprocessed data in its native format, without any predefined schema or organization. | A data lake addresses the need for storing diverse data types in a centralized repository and offers the agility and flexibility required for comprehensive data exploration and analysis. |
Data Lakehouse
| What is it | What does it mean | Why does it matter |
|---|---|---|
| A data lakehouse combines elements of the data warehouse with those of the data lake. Data lakehouses use data structures from a data warehouse and management features from data lakes, which are typically more cost-effective for data storage. | A data lakehouse is a unified platform that combines the strengths of a data lake and a data warehouse, providing a single solution for storing, processing, and analyzing data. It enables organizations to seamlessly integrate raw and processed data, simplifying data management, accelerating analytics, and gaining valuable insights from diverse data sources. | A data lakehouse offers the advantages of both a data lake and a data warehouse, providing a unified platform for data storage and analytics. By integrating raw and processed data and leveraging modern technologies, organizations can achieve efficient data processing, seamless data exploration, and significant time and cost savings. |