Symbl’s topic model is based on the internal conversation structure of how concepts are interrelated in a discussion. This goes beyond traditional topic classification which is tethered to its reliance on frequency, probability distribution and a supervised training algorithm. 

Before we dive further into Symbl’s advanced topic modeling capabilities, let’s take a look at traditional topic modeling, in order to understand how far we have come.

What is Traditional Topic Modeling?

Traditional topic modeling is a natural language processing (NLP) task for detecting keywords that have significance in a given document. Unlike Symbl’s advanced topic modeling, traditional topic modeling was specifically built for processing sets of documents, not for spoken conversations. Even though majority of topic modeling is applied to just about anything today, from standalone docs to chatbot conversations, the most used approach is unsupervised models best suited to analyzing multiple documents, detecting word patterns within them, and automatically clustering word groups and similar expressions that best characterize the entire set of documents.

Traditional topic modeling performs mostly keyword-based analysis like inverse normalization (doing the reverse sorting of keywords). This more rudimentary approach delivers results based on complex mathematical systems that work well when there is a clear structure and distribution of information in a document. For example, when analyzing news articles or formal documents.

Advanced topic modeling goes beyond documents and provides topic analysis for spoken conversations. This works with both real-time and off-line conversations.  

Consider a free-flowing conversation where someone initiates a thread on a topic and then switches to another topic during the same conversation. Non-context-aware topic modeling and topic classification systems will underperform here, reacting to the unstructured nature of the conversation data as data sparsity and ignore the order and semantic relationship between words because context doesn’t exist in those systems without intervention.

Each time a context switch happens in the conversation, Symbl’s topic algorithm can detect the change in the context and extract the most important topics out of it.

Using Added Intelligence to improve Topic Modeling

The topics algorithm provides a framework for users to calibrate and exactly model the relationship among the concepts. The analysis of certain fundamental features of the conversation graph provide the ability to abstract and derive the most relevant topics.

As mentioned earlier, when you try to use a traditional topic modeling for spoken conversations, that pattern no longer exists. There is no flow of information nor a coherent pattern you can mathematically model like can in documents. So you need a more sophisticated system that doesn’t rely on the normality of the document and the distribution of the way the words are used. To do this you want to understand the goal of that conversation – what concepts are they talking about and how they are talking about them. 

What Sets Symbl Topic Modeling Apart?

Symbl works off of a graphed based modeling system that analyzes the dialog and conversation, converting it into a graph of a conversation. This graph shows the relationship between different concepts, how they interact with each other. 

Graph analyses on top of this layer identify which combination of which concepts (which phases of that concept) have higher and higher significance. The graph favors the concept that has more diverse links in the conversational patterns.

Symbl not only identifies the top keywords in a conversation but also assigns a contextual score to them based on the graph intelligence that model’s the structure of conversation. You can see this scoring of keyword ranking in the Topics API response.

Parent Topics

Parent Topics are the highest level of abstraction of discussion and key aspects of discussion that the speakers talked and expanded their discussion on in the meeting. You can see ParentTopics of conversation in the Topics API response.

Scope

Scope of a topic defines the sentences and the information in the conversation that is directly linked to the topic of discussion. You can see the scope of the topic in the Topics API response.

This is the baseline engine that we use for Symbl’s topic modeling, but we can take this topic modeling and create different variations of it. 

Flat and Hierarchical Topics

One variation is called flat topic API, which consists of  just a list of topics. 

Going a level deeper, instead of flattening that topic, we can pick up the top hierarchy of the topic to create hierarchical topics. 

Abstract Topics

Another problem with the traditional topic modeling system is that it doesn’t generalize. It only looks at that document and that conversation, therefore all it has to work with are the words inside that document..

To solve this problem Symbl came up with abstract topics. The idea is to take all the elements of a graph and use deep learning to generalize it and find the abstract concepts that map to the specific thing being talked about. That way, a person that was not in that conversation can easily grasp what the conversation was about. 

Custom Topics

Custom topics are essentially just a layer that allows users to bias the way we analyze a graph. By allowing assigning topic detection to specific keywords, the Symbl API analyzes and changes the calibrating function to favor topics higher in the hierarchy list. That way you can introduce inherent bias into the deep-learning model. 

Next Steps

With Symbl, you have the capability to customize topics with your own topic vocabulary. You can also use trackers to find specific things in the conversation. Learn more about trackers here.

Ready to try Symbl.ai? Get started with a free account.

Visit our documentation page to learn more about Symbl topics. 

The post What are Topics and Topic Modeling? appeared first on Symbl.ai.

We have also introduced our Web SDK to speed up development of web applications that can understand and generate real-time insights for streaming audio data.

Finally, we are doubling down on our commitment to our developer community by introducing a free developer plan that includes free monthly credits of up to 1,000 minutes and 10,000 words of text files. 

New Developer Platform Experience

Once you are ready to log in to the Symbl developer platform, you will notice an improved onboarding experience. New users will especially find this new guided onboarding helpful, as you can make your first API call in less than a minute. You can also easily process audio and text conversations, and extract intelligence with a few clicks.

Once signed in to the developer platform, you will still see your Authentication (App ID and App Secret), as well as your current balance. But now you will notice links to explore new features, our prebuilt trackers library. And on the left you will find our navigation bar, which now includes two new items: Playground and API Explorer. 

Playground

Playground is where new users will want to go once they sign in. Here developers can see key features in action before even having to make a single API call on their own.

This view shows the three conversation types you can process: streaming, async and telephony. Using sample conversations, it shows what data becomes available using Symbl, including summary, topics, transcription with sentiment analysis, as well as conversation analytics. You can also provide your own conversations or speak into the mic and process a conversation in real-time. Users will have access to insights, which include actions, follow ups, questions and trackers.

After you process a conversation, your session will be saved in the Playground, where you can access or share both a visual view as well as a code view of those sessions. 

API Explorer

In the API Explorer, we expose you to all the Conversation APIs we have available. The Conversation API provides a REST API interface for getting your processed speech-to-text data and conversational insights. Once you choose an API, you will be able to see the steps involved, including obtaining an access token, processing a conversation, checking job status, and generating intelligence. Once you run a request, an access token is generated in the response section. The code snippets generated after each process can also be copied and used in your applications.

New Docs Experience

You may notice our documentation has undergone a major facelift. On the surface, our docs experience is brighter and more spaced out, creating a visual welcome space to help get you to the areas you are most interested in quickly. But what’s under the hood is even more impressive.

New Get Started Guide

We have revamped our Get Started section of our docs, to help you quickly get your App ID and App Secret and quickly get started using our APIs and features.

Interactive API Reference

Not only will you be able to learn about our APIs, you are now able to sign in to the Symbl.ai platform through the API reference and make calls right in our Docs page. Drop your App ID and App Secret in the API reference, and the page returns an access token. 

You can then take that access token and immediately use it to submit a file. Just drop the access token into the interface, bring over an audio url, and retrieve your conversation ID and job ID. The API calls are logged, whether successful or not. Pulling up the history you can now find the associated conversation ID.

Our docs also provide an updated Postman guide and introduce new visual elements to help with different learning types. In order to get you using our APIs as SDKs as quickly and easily as possible, we have also introduced the Interactive API Reference.

Introducing Web SDK

We recently announced the Beta version of our Web SDK, and are now excited to announce its general release with even more functionality. Symbl’s WebSDK enables you to programmatically integrate real-time conversation intelligence capabilities for sentiment analysis, topic modeling, intent analysis, and live captioning into your applications. 

Symbl’s Web SDK is an open-source kit that streamlines the development process for developers building applications with streaming audio using Symbl.ai’s Streaming API or Subscribe API. It does this by taking care of the WebSocket connections within the Symbl.ai platform, in line with the RFC 6455 standardized approach. 

WebSDK supports JavaScript and TypeScript languages and lets you build web apps that can understand and generate real-time insights for streaming audio data.

And to automatically resolve connection issues, Web SDK now supports reconnection. If reconnectOnError is set to true in the SymblConfig, Web SDK attempts to reconnect after a break in connection. 

Free Developer Plan

To empower the dev community, we are excited to offer a generous free plan – no credit card or calls with sales required to get started or go live. With 1,000 mins of audio and video conversation data and 10,000 words for  text files  – not to mention unlimited usage of our intelligence APIs on the developer platform – you can literally build and deploy product apps at no cost.

In addition to the free monthly credits, developers can process conversations in real-time or async on all channels. They can also process up to five conversations simultaneously.

When you upgrade to a pro account, your free credits are still available for you to use. Only when you exhaust your free minutes, will you be charged on a pay-as-you-go basis.

Symbl Dev Community 

In order to help developers on their journey we also have introduced a community forum in addition to our public Slack channel. The Symbl Community is a discussion forum where builders can collaborate, share knowledge and best practices, and track previously made requests that sometimes get lost in Slack messages.  

Advanced Conversation Understanding AI for your Apps

Some cloud vendors may do a good job offering generic ML-driven features, but do not offer models that are robust enough for complex tasks like understanding unstructured human to human conversations  data generated from video, audio, and chat. 

Symbl.ai platform users can easily explore and test out summarization, sentiments, topics, intents, and more–no credit card required. 

Symbl is on a mission to make it easy for developers to integrate advanced conversation understanding AI into the applications they are building, and all of the updates mentioned in this post are the result of that commitment. 

Get started building with Symbl today.

The post Announcing New Symbl.ai Platform Experience, Interactive Docs, Web SDK and More appeared first on Symbl.ai.

What is Sentiment Analysis?

Sentiment analysis (also known as opinion mining) is the process of helping users understand human thoughts and feelings in all types of data. ​​Sentiment analysis tools interpret that general feeling – or sense of an object or a situation – using natural language processing (NLP). To do this, machine learning (ML) algorithms systematically identify, extract, quantify, and study affective states and subjective information. 

Sentiment analysis detects underlying positive, negative or neutral sentiment in text, voice, and video conversations. 

Businesses can leverage this data for a variety of uses. These include shaping sales and marketing plans, evaluating social media posts, improving crisis management and brand strength, and translating digital PR into tangible actions.

What are the Different Types of Sentiment Analysis?

Document-Level Sentiment Analysis

Document-level Sentiment Analysis reviews text and determines whether it has a positive or negative sentiment. It supports any sentiment-bearing text and determines the overall opinion  of the document. 

Sentiment analysis at the document level assumes that each document expresses opinions on a single entity.

Sentence-Level Sentiment Analysis

Sentence level sentiment analysis determines if each sentence has expressed an opinion. This  level distinguishes the objective sentences expressing factual information and subjective  sentences expressing opinions. 

This time of sentiment analysis first identifies if the  sentence has expressed an opinion or not, and then assesses the polarity of that opinion. 

Aspect-Based Sentiment Analysis

Aspect based sentiment analysis refers to categorizing opinions by aspect and identifies the sentiment related to each 

First, a system identifies the attitude targets mentioned in a given sentence. This process is known as aspect extraction. Once these aspects are identified, a system determines the attitude associated with each target in a process known as aspect-level sentiment analysis.

Rule-based strategies that leverage predefined text classifiers are a common technique for aspect extraction. A variety of approaches have been developed to understand the relationship between attitude targets and their context.

Contextual-Based Sentiment Analysis

Context based analysis is used to recognize cues and enhance other types of sentiment analysis. Contextual sentiment analysis refers to the way words change their meaning with context. The same word or phrase can be positive, neutral, or negative, depending on other words in the sentence.

Sentiment is also strongly influenced by background knowledge. People do not express commonsense knowledge that they expect anyone to know. Understanding this implicit knowledge is vital. 

Topic-Based Sentiment Analysis

Topics are key drivers of conversations. In fact, they’re the most important keywords or phrases used. Topic level sentiment analysis breaks down a message into topic chunks and then assigns a sentiment score to each topic. Sentiment analysis on topics determines whether the topics resulting from the conversation are positive, negative, or neutral. 

The topics algorithm provides a framework for the user to calibrate and precisely model the relationship among the concepts and understand the semantics used in conversations. Not all sentiment analysis tools offer topic-based sentiment analysis.

Sarcasm Analysis

Sarcasm and irony are highly prevalent in everyday conversation, which makes sarcasm analysis a critical area of focus for successful sentiment analysis systems.

Research into sarcasm analysis has historically focused on sentence-level understanding. A variety of approaches have been tested focusing on sentence-level features, such as detecting incongruity between the sentiment expressed by different words within a sentence.

Bias in Sentiment Analysis

Whether it’s used in customer care, market research, or reputation management, sentiment analysis typically handles data from a wide variety of demographic backgrounds. With that in mind, it’s critical to remove bias that can introduce error into sentiment analysis.

Bias is frequently introduced into sentiment analysis systems through word embeddings. That refers the underlying representation that results when words and phrases are mapped to a vector space for use by sentiment analysis systems.

Since bias can be easily introduced into sentiment analysis systems, identifying effective de-biasing methods is an emerging area of study. This is an important focus for future development for sentiment analysis.

How to Choose the Best Sentiment Analysis API

When looking for the right API, check for contextual understanding to get the most accurate insights. Also look for how the sentiment analysis is built for audio or video content rather than generic text content. The right sentiment analysis API should provide the performance and extensibility you need to achieve your business objectives. It should also give you the ability to surface useful information in real-time.

Symbl.ai uses deep learning models for sentiment analysis. It’s created on top of existing language models with existing language data to detect sentiments at the sentence level. It works in conjunction with our topic modeling system to scope the segments of conversation by contextual coherence of that segment and assign that sentiment to the related topic. 

We take the overall sentiment and calculates it based on topic. Symbl can generate topic level sentiment, offering contextual relevance of the sentiment as well as context of how the topic is set in the conversation. And given our advanced method for sentence level and topic level analysis, you can derive very accurate document level analysis. This sets Symbl apart from other document level analysis tools.

READ MORE: Using aspect-based sentiment analysis for voice and video conversation with Symbl.ai

How Does Symbl’s Sentiment Analysis Work?

Symbl uses the transformer based deep learning architecture for sentiment analysis. Our sentiment analysis architecture is first fine tuned on conversational data. From the multi-modal aspect, we use topics to detect scopes in which the said topic is being talked about. We also consider the sentiment of these specific paragraphs in order to compute the overall sentiment of the conversation.

To analyze Sentiment, Symbl.ai combines the Conversation API’s Speech-to-Text messages (usually sentences) with Conversation Topics. For a given conversation, the Topics algorithm analyzes each message and provides a sentiment intensity / polarity score (-1.0 to +1.0) and suggested type (positive, neutral, negative).

Why Use the Symbl.ai Platform for Sentiment Analysis?

Symbl.ai’s Sentiment API offers aspect-based sentiment analysis performed on real-time messages. It also offers polarity values that you can freely define and adjust after testing. 

Most importantly, users have access to integrations that bring a human-level understanding to different contexts without upfront training data or custom classifiers. And they get access to other valuable conversation analytics, including speaker ratio, talk time, silence, pace, and overlap. 

Supported Channels 

Symbl.ai’s APIs can be used on both asynchronous audio, video or text data as well as streaming audio or video content. Symbl generates real-time sentiments over WebSocket protocol using Symbl’s Streaming API. You can also get sentiment analysis on recorded conversations by processing using Async APIs and extracting the sentiments on sentences and topics using Conversation API.  

For step-by-step instructions on how to implement Symbl.ai Sentiment Analysis visit our docs page.

What Can You Do with Sentiment Analysis?

• Social audio content listening – in day-to-day monitoring, or around a specific event such as a product launch.
• Analyzing video survey responses for a large-scale research program.
• Processing employee feedback in a large organization through meetings.
• Identifying very unhappy customers so you can offer closed-loop follow up.
• See where sentiment trends are clustered in particular groups or regions.
• Competitor research – checking your approval levels against comparable business.
• Use for compliance, risk management and data governance.
• To measure engagement and empathy in internal and external conversations. 

Next Steps

Symbl Trackers in conjunction with sentence level sentiment also offer a very powerful and flexible way to do zero shot analysis. This gives the user the power to define the aspect/feature and then perform the analysis. 

Learn more about Symbl Trackers.

Ready to try Symbl.ai? Get started with a free account.

READ MORE: Learn how to configure your new Symbl.ai developer account to analyze recorded calls for sentiments with cURL commands.

The post Sentiment Analysis Explained appeared first on Symbl.ai.

Consider customer service exchanges. Some are audio, such as calls to your service center. Others use functions like chat. These exchanges contain a lot of information that can boost sales or assist marketing. The problem is that because of the large volume and unstructured nature of these “conversations,” it takes great effort to capture and understand how they provide information for process improvement. Conversation intelligence fixes this problem.

The ability to derive meaning from conversations includes information that can assist people in their jobs. As with any analytic process, the tools you need and the technology you use depend on the questions you are trying to answer and the decisions you’re trying to make. The technology must fit the problem.

This article explores the available features in conversation intelligence use cases, and how Symbl.ai make its easy to implement conversation intelligence in your organization.

Leveraging Conversation Intelligence Technology

Conversation intelligence must work in all the conversational channels of your business – whether files (audio, video, or text), streaming video, or by phone. It must connect easily to capture all of your relevant communication channels. You need a platform that connects through APIs or SDKs to capture your conversations and incorporate them into your applications.

Once you have captured a conversation, you want the ability to determine the features of the conversation. These features go beyond speech-to-text, selecting the topics, topic hierarchies’ scope, and conversation sentiment. 

There is also valuable analytics beyond content to help you understand and act on conversation intelligence. These include:

• Talk time and speaker ratio — the total time and percentage of total time used by a given speaker.
• Silence — the time when none of the speakers was saying anything.
• Pace — the speed, in words per minute, that a person spoke.
• Overlap in conversation — the percent and total time a speaker spoke over another speaker.

Each of these elements indicates the dynamics of the conversation. For example, suppose the conversation was between a new salesperson and a potential customer. In that case, these statistics could assist in coaching the salesperson to listen more or not talk over the customer. 

To understand this information, you want to identify the entities and intent of the conversation: who participated, the company, location, and role in the conversation. This information adds another layer of understanding to the conversation intelligence statistics.

How Does Conversation Intelligence Inform Sales, Marketing, E-Commerce, and Customer Experience Strategy?

Conversation intelligence provides the information needed to improve the variety of interactions your business has. While sales and marketing effectiveness measures tell you the results, conversation intelligence provides insight into the all-important “why” and gives you the information you need to improve the process. Listening to customers rather than just talking to them can make a huge difference in sales. Conversation intelligence provides the information to improve your processes.

Let’s dive into a few examples.

Sales

Conversation intelligence allows you to analyze call performance in real-time and understand critical events in the conversation. You can determine the reason for calling and the handler’s response if it is an inbound call. Understanding why customers are calling allows you to see product defects or areas of customer disappointment. 

Are the call center employees adhering to the script? Areas of the script that result in silence may need to be adjusted. Are your employees not listening and talking over the customer? Are there places in the call where the customer gets annoyed (negative sentiment)? Understanding these call details can help you improve your call center processes and retain customers.

Every sales team has its top performers. The question is what they’re doing differently than the average salesperson. Collecting and analyzing conversation intelligence information using a tool like Symbl allows you to understand the conversational strengths of your top salespeople and compare them to the rest of the team, increasing revenue and sales intelligence. 

What are their talk topics, duration of the call, and talk-to-listen-to ratios? This information can detect behavioral differences and provide coaching points to improve the entire team’s performance. You can also use conversation intelligence to track and compare changes in performance, thereby validating the coaching effectiveness. 

Sales conversations change over time. For example, what effect does a new market competitor have on your sales? Conversation intelligence lets you know when new topics change your sales conversations, allowing you to react. In the case of a new competitor, you want your salespeople to have a ready answer to challenges rather than responding with silence. You can uncover new factors affecting your sales by analyzing the topic and concern trends.

Marketing

Conversation intelligence also has applications in marketing. Understanding a competitor’s approach to the market can help shape your marketing messages. You can apply conversation intelligence to analyze the content of competitors’ webinars, and by using topic identification, you can understand the messaging your competitors are using. By analyzing the content of the QA session, you can pick up on the listener’s areas of interest and concern. 

If you use focus groups or have conversations with key customers, you can use conversation intelligence to identify win and loss factors and areas of customer empathy. With Symbl, you can use conversation intelligence to identify trending topics across various conversational forms, whether social media, chat, webinars, or customer service. Symbl supports conversation intelligence for events, webinars, and even social media apps. These all have data you can analyze to reveal trends, areas of concern, or general questions. 

For example, are comments on a new product trending positive or negative? Is the community asking for more information, such as details on features or buying options? Are there complaints that you must address? Conversation Intelligence’s ability to capture and analyze the social media conversation provides vital insights to improve marketing and sales. 

E-commerce

Many of your e-commerce applications have your customers complete the entire transaction online. Sometimes they feel they have to talk to someone over the phone to get the necessary information to complete the sale. In other cases, they abandon their transaction and go to a competitor. Conversation intelligence can provide insight into what happened.

E-commerce transactions generate a high volume of conversational information, including inquiries about product features, requests, or complaints. With a conversation intelligence tool like Symbl, you can increase revenue and sales intelligence by analyzing customer interactions with the website as a conversation. Additionally, you can use conversation analytics on the large volume of transactions to determine customer sentiment, identify areas of high customer interest, or interpret other feedback generated by the transaction. 

Customer Experience

Customer experience includes all aspects of a customer’s interaction with your enterprise, including everything from product quality and ease of use to advertising, sales, and customer service. You can view them as aspects of having a conversation with your customer, and every interaction provides a way to capture a customer’s present and future business.

The impact of conversational intelligence is not just in its ability to improve sales or customer service interaction alone. It’s about improving the customer experience. 

For example, by analyzing a customer service call from a customer experiencing an issue, you can determine its root cause, such as an unclear instruction. By correcting the instructions, you can remove the need to call and thereby improve customer experience.

Conversation intelligence provides information that can improve the performance of the individuals involved in the conversation and remove the root cause of the problem. Using a tool like Symbl to boost your strategy helps to create a more robust customer care response. 

Taking Your Conversation Intelligence Strategy to the Next Level with Symbl.ai

You can implement all the capabilities and conversation intelligence use cases mentioned above by using Symbl.ai. Symbl.ai gives you the complete package, from connectivity to analytics. It provides developers with the tools to embed conversation Intelligence into your applications and boost your business. Symbl.ai has a wide range of APIs and SDKs, making it easy for developers to convert audio, video, or chat data. Tools within Symbl.ai take this data and provide you with summaries, intents, topics, and sentiments that you can incorporate directly into your applications. Symbl.ai uses artificial intelligence models to give the context-awareness necessary to understand a conversation. 

Conversation Intelligence requires connectivity to various sources, the algorithms to understand it, and the tools to act on it. Symbl.ai allows you to embed conversation intelligence into your existing applications easily. For more information, visit Symbl.ai and create a free trial account.

The post Conversation Intelligence Technology: Understanding the Use Cases appeared first on Symbl.ai.

“We’ve worked with hundreds of early-stage B2B tech companies who are on-mission to transform the way we live and work,” said Paige Robinson, Founder & CEO of Will Reed. “We believe the most successful companies are those like Symbl, who are committed to building human-first cultures that offer meaningful work and support the full employee experience.” 

In November 2021 Symbl.ai announced a $17 million Series A funding round to accelerate product development and substantially grow our engineering and leadership teams. Symbl also expanded sales and marketing to meet the growing demand for our offering. 

“We are very excited about this recognition and what it says about our growing company, especially two years in a row. We have made every effort to invest in our people, and it is a continued focus for the company. This award just proves to me that we are building the right team, who all share a similar vision and focus.” 

-Symbl.ai CEO Surbhi Rathore

Visit top100bywillreed.com to watch Surhbhi share more about Symbl’s mission, vision, culture, and growth plans. 

The post Symbl.ai Named Top 100 Early-Stage Company to Work for in 2022 appeared first on Symbl.ai.

WebSockets are primarily used for two-way communication between a server and a client, such as a web server and a browser. WebSockets keep a connection open between the server and the client so information can be sent back and forth between the two. The WebSocket protocol, which most browsers support, defines a standardized way for this communication to take place, usually by opening a connection over the Transmission Control Protocol (TCP).

Webhooks work slightly differently. Commonly used for one-way server-to-server communication, webhooks use regular HTTP to send requests from one server to another. A source service can send requests containing data to a destination service using a webhook URL provided by the destination service. The destination service then must then implement logic to accept the HTTP request and process it accordingly.

Now that you understand how webhooks and WebSockets work, let’s delve into their key differences and use cases so you know when to use webhooks versus WebSockets.

Webhooks Versus WebSockets

The primary difference between Webhooks and WebSockets is the method of communication each uses. Because webhooks are primarily used for one-way communication, while WebSockets are primarily used for two-way communication between a client and server, each communication method is best suited to a particular kind of task.

Webhooks are useful when you want to be notified about an event occurring in a particular system, like a status change or an update to a specific entity. You can provide the system with a webhook URL and choose which events you’re interested in. This event in the source system can then trigger an HTTP request to be sent to the given webhook URL. The consumer builds a service with HTTP endpoints that can accept the incoming requests made by the webhook service, parse the request, and process any data. The communication flows one way — from source to destination — and the socket is closed after each request.

WebSockets are useful when two-way communication is required, usually between a client and server, but also for server-to-server communication. The WebSocket is set up on the server and the client then connects to it. The initial request from the client is an HTTP handshake request, which is upgraded to a WebSocket connection if it’s valid. Both client and server can send messages via the WebSocket and each must implement message listeners, called when a message is received. This allows the message to be processed and, if necessary, a response to be made. The connection remains open for as long as required and is closed upon completion. 

The main advantage of webhooks is that the client can send messages back to the server. However, because WebSockets open a connection between a specific client and server, they make the server stateful (as it now contains the active connection), and therefore more difficult to scale.

Let’s look at some use cases for each to understand when and why you would use one over the other.

Webhooks Use Case

Webhooks are great for keeping track of changes in an external service, like tracking added or modified transactions in an account. They’re also helpful to both the service providing the webhook and the downstream consumer. Additionally, webhooks are inherently stateless as they don’t need to keep an open connection, so scaling them is far easier. Webhook providers can cater to much larger volumes of subscribers and scale in either direction to meet demand. 

Webhooks are also simpler for the consumer, with no need to manage connections and state information. This makes them especially useful for services that regularly publish updates to many different consumers. And if you’re tracking sales on an online marketplace store, or donations to a charity page with millions of donors, webhooks let you process these updates in real-time.

You may think that WebSockets are the best choice for a chatbot, as communication goes both ways. However, to be able to meet scaling demands, larger businesses should consider webhooks — especially if they handle large numbers of users in simultaneous chat sessions.

WebSockets Use Case

There are instances where two-way communication is required between a client and a server so that data can be shared in real-time. 

For example, consider building a web document editing service that allows multiple people to edit and update a document in real-time via their browsers. This would be a great use case for WebSockets. If two users have the same document open in their browser, both will be connected to the server via a WebSocket. When User A makes a change, it’s sent from the client to the server before being relayed back to User B, updating their browser and showing the edit. This exchange can continue, with both users submitting changes, the server processing them, and the changes being relayed back in real-time. 

Though it’s possible to achieve this real-time communication with webhooks, there would likely be latency issues with such a large number of API calls to send and receive updates. Using a WebSocket helps to eliminate latency and delays.

Conclusion

Webhooks and WebSockets are both great tools for real-time conversation analytics and real-time communication over the Internet. However, their different methods of communication make webhooks and WebSockets best suited to different uses. If you require one-way communication between two servers and scaling is important, you should use webhooks. If you need two-way communication between a client and server that are constantly exchanging data in real-time, you’ll find that WebSockets will better meet your needs.

If webhooks suit your needs best, just remember to take care when implementing them. A webhook is effectively a reverse API, so you should carefully consider what information is being exposed. Signing and encrypting payloads ensures that data can’t be intercepted and viewed by anyone other than the expected party. With your webhooks secured, you’ll be able to safely scale as needed, experiencing their benefits with the knowledge that your data is safe.

The post When to Use Webhooks vs. WebSockets appeared first on Symbl.ai.