Making sense of composable CDPs: What martech buyers need to know MarTech compliance and privacy the right data An exhaustive approach to identity resolution. Significant data collection and, therefore, data quality effort. An extremely complex data integration implementation.

Making sense of composable CDPs: What martech buyers need to know

Companies disclosing environmental impacts through CDP now account for more than two-thirds of global market capitalization.

CDP secures environmental disclosures from a record 23,000 companies

Using LLMs in data analysis costs less than 1 per cent as much as hiring a human analyst while turning in comparable performance, according to a recent study by Damo Academy, the in-house research ...

BlackRock: AI tools like ChatGPT are a ‘revolution’ in investment analysis, able to pull nuance from vast data sets

Here are the questions used for this analysis, along with responses, and its methodology. In an era where every click, tap or keystroke leaves a digital trail, Americans remain uneasy and uncertain about their personal data and feel they have little ...

How Americans View Data Privacy

Customer Data Platforms or CDPs. Their job is quite simply to take information from across different channels and collate it in one place. These platforms such as Segment and mParticle have been ...

Customer Data Platforms: Off-The-Shelf vs Composable

Difficult initial “MVP” deployments, as enterprises struggle to deal with core issues of data availability and quality, as well as identity resolution blockers. Stunted roll-outs, where enterprises lack the resources to broaden a CDP footprint beyond a ...

Customer data debt: The hidden obstacle to CDP success

This report covers a research time span from 2018 to 2028, and presents a deep and comprehensive analysis of the global Customer Data Platforms (CDP) market, with a systematical description of the ...

Customer Data Platforms (CDP) Market 2023 to Hit Statistical Significance Value by 2030

making data analysis accessible to everyone. With more than 350 pre-built integrations and AI-powered support for custom data sources, CorralData is committed to helping businesses make data-driven decisions and stay focused on the metrics that matter.

CorralData, Ready Set Rocket, and Baruch College to Host AI & Data Conference in New York

But who are the biggest creators? The influencer-marketing firm Captiv8 ran a recent analysis studying the top Amazon Live creators. To collect this data, Captiv8 analyzed total follower count ...

The top Amazon Live influencers selling products, according to new data analysis

Additionally, he discussed its potential to pinpoint pain points and bottlenecks in these networks by employing benchmarking and comparative analysis. Xiaolin Wen concluded that on-chain data ...

Enhancing Web3.0 security through on-chain data analysis: Insights from industry leaders

mParticle

mparticle

Growth marketers - this one is for you. mParticle & HackerNoon are excited to host a . Here’s your chance to win money from a whopping $12,000 prize pool! It could be any story on . We highly encourage you to write on the following topics: Growth Marketing Writing Contest #growth-marketing Product Management Customer Data Platforms (CDP) Growth Marketing Data Engineering Data Management The contest starts on September 1st! Prizes Awarded Each Month 1000 USD 1st Prize: 600 USD 2nd Prize: 400 USD 3rd Prize: Remember: The sooner you submit your first story, the higher your chances of winning! About mParticle mParticle is a real-time AI customer data platform that powers your entire marketing stack with high-quality customer data. Request for a demo today! The Growth Marketing Writing Contest Rules & Guidelines Must be 18+ to enter Story content can be any story on . We encourage you to write on product management, customer data platform, growth marketing, data engineering, and data management. #growth-marketing Must . create a HackerNoon account Who Can Enter the Contest? Anyone over the age of 18. There are no location restrictions. Can I Write Under a Pen Name? Yes! You can use your real name on your HN profile, a fake name, or even create a persona to write under. How Long Will the Contest Run? September 1 - September 30, 2022 Round 1: October 1 - October 31, 2022 Round 2: November 1 - November 30, 2022 Round 3: December 1 - December 31, 2022 Round 4: January 1 - January 31, 2023 Round 5: February 1 - February 28, 2023 Round 6: Can I submit more than one entry to the contest? Of course! Each story submission shall be considered a new entry into the writing contest. How are the winners selected? Each month, we’ll take the top 10 story submissions that receive the most eyeballs (real humans, not bots!). Those 10 stories will be voted on by HackerNoon staff, and the 3 stories with the most votes will win. Can I win more than one prize? Yes! If you submit multiple stories, each has the chance to win. Nominees are selected by total traffic each month and then voted on by our team. If 2 of your submitted stories that month are in the top that month, both of them are eligible to win a prize. Meaning you could potentially bag both first and second place in the contest if 2 of your stories make it to the top 10. If I win a prize in round 1, can I still submit another entry into round 2? Absolutely! Please feel free to enter each round of the competition. Wish y’all the best of luck! Looking forward to reading great content on . #growth marketing

The Growth Marketing Writing Contest by mParticle and HackerNoon

El concurso de redacción sobre marketing de crecimiento de mParticle y HackerNoon

Le concours d'écriture sur le marketing de croissance par mParticle et HackerNoon

mParticle और HackerNoon द्वारा ग्रोथ मार्केटिंग राइटिंग प्रतियोगिता

mParticle と HackerNoon によるグロース マーケティング ライティング コンテスト

O concurso de redação de marketing de crescimento por mParticle e HackerNoon

Cuộc thi Viết về Tiếp thị Tăng trưởng của mParticle và HackerNoon

mParticle 和 HackerNoon 举办的增长营销写作比赛

As a developer, being tasked with integrating a third-party system into your products can feel like standing at the base of a giant mountain. This isn't just any mountain, however. To reach the summit, you’ll need to navigate through a Pac-Man-like maze that takes you up, down, left, and right along a winding route filled with all sorts of hurdles and setbacks. Sure, you’ll get rewarded with some apples in the form of a successful API call or a passing unit test. But as soon as you turn the corner, the unforgiving ghost of vague documentation looks to stymie your progress. You probably didn’t become a developer to spend your time climbing weird Pac-Man mountains. More likely, it was the thrill of building useful products and features that drew you to the job. Of course, there will always be the need to work on projects that direct some of your time and attention away from the really interesting and impactful work. When the need to integrate vendor APIs into your application does arise, it’s best to be equipped with resources that eliminate the frustrating guesswork that all too often goes hand-in-hand with this task. You may be under the impression that adopting a CDP is an arduous process that will drain your engineering team of its lifeblood for months. This certainly does not have to be the case––especially with the aid of a tool built by developers and for developers that ensures the implementation process is straightforward and headache-free. This is exactly the role that our , , and sample applications fulfill. In the rest of this article, we’ll dive deeper into why we built these sample apps, and how they can help you reduce the time it takes to implement mParticle and realize value from your data. Web iOS Android What are the mParticle Sample Applications? The new mParticle sample apps are fully functional projects that illustrate how mParticle data collection works in the context of three major platforms—Web, iOS, and Android. Each of these projects uses a fictional eCommerce store called to demonstrate how our SDKs capture user events and identities in a production context. HiggsMart Here is what HiggsMart looks like running locally: Why did we build these sample apps? When it comes to understanding any third-party system, there is no better teacher than hands-on experience. In each app, we’ve implemented the mParticle SDK to collect data events based on common triggers in the user flow, such as viewing a product page, adding an item to a cart, customizing an order, and completing a purchase. By inspecting the code behind these events in your IDE or browser developer tools, you can quickly learn best practices for implementing event collection with mParticle that you can apply directly to your own projects. Quickly learn how and when to use different mParticle events Examining and experimenting with the code behind these projects will help you quickly understand the different event types that the mParticle SDK provides, and how to implement these events in your own applications. Since the events in these applications were designed to closely mirror those in real-world projects, in many cases you can even copy these events directly into your own application. useEffect(() => {
    // Renders an initial cart view when the page loads
    mParticle.logPageView('View My Cart', {
        number_of_products: numberOfProducts,
        total_product_amounts: subTotal,
    });
}, []); By providing these in-context code snippets that you can copy/paste directly into your applications, we intend to help you implement data collection more quickly and seamlessly. Of course, our documentation is still there for you to fine-tune these events to meet the needs of your specific use cases. But beginning with working code means starting the race that much closer to the finish line. The sooner your events are written, deployed, and working in production, the sooner your teams will realize value from your data, and the sooner you can move on to core engineering initiatives. See the complete flow of data from collection to activation In addition to using these sample apps to bootstrap event implementation, we are also piloting a new onboarding tutorial that allows you to send data from the HiggsMart Web interface directly to your mParticle instance, examine events as they enter the Live Stream in real time, then forward these events to an output destination of your choosing. This tutorial is available to new participants in our Accelerator Program— about the program and see if your startup qualifies for free access to mParticle. learn more The first step in the tutorial is to generate API credentials for your workspace, then add the API key to your sample app. This establishes the sample app as a data input source for mParticle, at which point the user events you perform in the sample app will be visible in Live Stream in real time. Following this, the tutorial guides you through the additional steps you would take to complete the flow of data, including: Establishing a sample output source for your data (in this case, a Webhook) Connecting your input to your output and configuring this connection Sending more test events from the sample app to mParticle Viewing outbound events in the mParticle Live Stream Seeing forwarded events arrive in your output destination This onboarding tutorial will quickly familiarize you with the aspects of the mParticle UI that pertain to implementing and testing event collection. Additionally, since the tutorial also covers how mParticle facilitates each aspect of your data’s lifecycle, you will clearly see how the data consumers within your organization will interact with the data events that your event collection code is forwarding to mParticle. Getting started To get started with the mParticle sample apps: 1: Visit the repositories for the , , or projects Web iOS Android 2: Fork or clone the repo of your choice 3: Follow the instructions provided in the README. for a hands-on introduction to CDPs using these Web, iOS, and Android Sample Apps on May 10th, 2022! Sign up

Implement a CDP with Ease Using mParticle's Sample Applications

This sample application allows you to see mParticle data events and attributes displayed in an eCommerce UI as you perform them and experiment with implementing an mParticle data plan yourself. Effective cross-functional data planning is a great way to ensure data quality throughout the data lifecycle and avoid time-intensive debugging for data engineers. In this and , respectively, we covered how to create data plans using the mParticle Data Plan Builder and leverage mParticle’s Live Stream to debug data errors before they reach your downstream systems. article video However, between developing a data plan and validating incoming data lies a crucial step that falls squarely on the shoulders of engineers: Implementing data collection. Without effective tools to aid in implementation, writing data collection code can be an error-prone time sink for technical teams. In many scenarios, developers find themselves needing to translate a static data plan into
production code. This not only diverts engineering cycles away from core objectives––it increases the likelihood of data collection errors as well. It is easy to imagine how a hyphen in the data plan could inadvertently be written as an underscore in the code, for example. Once this error is discovered further along in the ETL pipeline, data engineers will be called upon to track down its origin and correct the source code. Luckily, mParticle’s Data Planning tools and templates eliminate these problems by giving developers the resources to implement data collection efficiently, accurately, and consistently across platforms. In this article, we’ll use this as our development environment to demonstrate implementing an mParticle data plan using the and the . Additionally, we’ll talk through some of the choices we made in creating the sample app itself, which we hope you'll find to be a helpful playground as you learn more about implementing data collection. eCommerce sample application Data Plan Builder Snippet tool Implement Data collection with mParticle’s Data Planning Tools Step 1: Create a data plan Before we implemented data collection, we needed a data plan to work from. Using mParticle’s (which is one of six vertical-specific templates that our Data Plan Builder provides), we instantly had a set of retail-specific identities, events, and attributes defined for us to start with. retail data plan template Step 2: Turn the data plan into JSON From there, we could use the Data Plan Builder’s built-in script to automatically translate the data plan (as defined in the tabs of the Google Sheet) into machine-readable JSON: Step 3: Automatically generate event collection code With this JSON in hand, we used the mParticle to automatically generate the data collection code that we would implement in our application. This Snippet tool’s interface immediately represents each event in the data plan as a discrete code block that can be transported directly into your source code: Snippet tool In this case we used the “Web SDK” option (which outputs JavaScript). As you can see in the example above, the interface also supports code generation for iOS (Swift and Objective-C) as well as Android (Kotlin and Java). Since the Snippet tool generates this code directly from the JSON representing the data plan, it eliminates many of the errors that commonly arise when manually translating data plans into code––for example, misspelling attribute names or using incorrect data types. Additionally, having a tool capable of generating data collection code across major platforms from the same JSON source addresses another significant challenge of implementing a data plan: Maintaining consistently across data inputs. This can be especially challenging at large enterprises with a variety of touchpoints, where it is common to have separate engineering teams handling Web, iOS, Android and OTT products. Step 4: Implement events throughout the application While the Snippet tool does the work of generating the event collection code, we still have to implement these events throughout the application using the appropriate user-based triggers to fire them. Many of our commerce events in this application are tied to users viewing or customizing products. Since we’re using a component-based architecture with React, our application has one template, , that we use to dynamically display each individual product to the user. Therefore, we only had to implement each product-related event once on this page for it to work for each product the site displays. ItemView.js For example, let’s look at how we implemented the Product Detail event, which sends three attributes, id, name, and categories, to mParticle: useEffect(() => {
     // isBrowser refers to a variable that checks whether the window object is defined
    if (isBrowser) {      
      let customAttributes = {
        id: props.pageContext.content.id,
        name: props.pageContext.content.name, 
        categories: props.pageContext.content.categories
      }
      // Log event to mParticle 
      window.mParticle.logPageView("Product", customAttributes)
      // Display event properties in UI	
      toast(
        <ToastSuccess
          eventName="Product"
          eventCategory="Screen View"
          product={customAttributes.name}
        ></ToastSuccess>,
        {
          position: toast.POSITION.TOP_RIGHT,
          className: "success-toast",
        }
      )
    }
  }) Since we wanted this event to run each time the user views a new product, we placed it inside the useEffect hook, which will automatically run each time the component renders. Inside useEffect, we initialize an object called customAttributes which we use to store the event attributes: id, name, and categories. We then pass these attributes as the second argument to logPageView, which forwards this event along to mParticle. How We Built This Data Planning Sample App What tools did we use? We bootstrapped our data planning sample app with this , which provides an out-of-the-box retail user interface built on React, Next.js, and Gatsby. Since most developers working on eCommerce products today leverage these tools or similar component-based frameworks, we felt this stack would be ideal for showcasing this process. Jamstack starter template Displaying events in the user interface As you , you will see notifications appear in the top-right corner of the as you take actions including: browse through the live deployment of this user interface Viewing different product pages Customizing the color/configuration of products Adding or removing products from your cart Selecting a username Changing your marketing consent status Purchase the items in your cart To add this feature, we used , an awesome and easily customizable React package for adding notifications to a user interface. Each time we log an event in our sample app, we make a call to toast, passing in our custom-made notification component as the first argument, and an object containing some styling information as the second: React Toastify toast(
        <ToastSuccess
          eventName="Product"
          eventCategory="Screen View"
          product={customAttributes.name}
        ></ToastSuccess>,
        {
          position: toast.POSITION.TOP_RIGHT,
          className: "success-toast",
        }
      ) Experiment With This Sample App Using Your Own mParticle Workspace We invite you to experiment with and the and explore how you can easily build and implement data plans that capture the customer events and user identities that matter to your business. this application Data Plan Builder Tools/Resources discussed in this post: Data plan template (Generic) Data plan template (Retail) Snippet tool eCommerce sample app (repository) eCommerce sample app (live) React Toastify Check out on how to protect data quality from collection to activation! more resources

A Guide to Implementing an mParticle Data Plan in an eCommerce App

It seems as if time flew - the final results announcement of the Growth Marketing Writing Contest by HackerNoon and mParticle is here! We thank mParticle for sponsoring this amazing contest for the HackerNoon community! During these 6 months, we published 200+ growth marketing stories generating around half a million reads and months of reading time! The Growth Marketing Writing Contest: Final Round Nominations & Winners For selecting the top 10 nominations, we picked all the stories with the #growth-marketingtag on HackerNoon, published in February 2023. Then we chose the top stories using 60:30:10 weightage respectively to: Number of hours read
The number of people reached
The freshness of the content Top 10 Nominations How to Use ChatGPT for Marketing by @julietde
AI and Chill: 7 Growth Marketing Tools to Streamline Your Workflow by @khamisihamisi
Marketing with a Handicap: How to Promote Your Product on a Budget by @correspondentone
How I Met My Girlfriend by Running Ads on Facebook: A Growth Marketer’s Playbook by @0xjack
How to write SaaS comparison pages that beat the competition by @electriccopy
AI in Marketing Ensures the Survival of Artists Who ‘Think Different’ by @nilanganray
Growth Trends in Tech and Retail Industries by @anisimovapo
Nobody Buys an NFT Just for Art by @sergey-baloyan
23 Free Digital Tools That Will Save You Thousands Of Hours by @khamisihamisi
The Secret Plugin for ChatGPT3 by @alfredodecandia #Growth Marketing: Final Round Winners In the first place, we have: https://hackernoon.com/how-i-met-my-girlfriend-by-running-ads-on-facebook-a-growth-marketers-playbook?embedable=true I probably had over 200+ submissions from interested prospects. I probably did Zoom calls with about 10 different women. And out of that, I met 4 in person. And the last woman I met, we’re still together.
Sorry, no picture of her… yet. She said she’ll let me drop a press release if we get married.
Of all the growth hacks I’ve implemented over the years, this is by far my best. After all, what’s the value of finding love? Priceless. That was a fun read, @0xjack! Congratulation on winning $1000! The second position is won by: https://hackernoon.com/how-to-write-saas-comparison-pages-that-beat-the-competition?embedable=true “Focusing on a best-fit prospect segment doesn't always make sense to people who don't have a marketing background. If I wanted to increase my chances of landing a customer, would I want to target as broad a market as possible? The reality is the exact opposite. The broader your focus, the more difficult it is to connect with prospects and convince them that your solution is the best one for them.” Congratulations,@electriccopy! You have won $600!In the third place, we have: https://hackernoon.com/how-to-use-chatgpt-for-marketing?embedable=true On the one hand, it seems that they can completely replace marketing specialists, but it is still worth considering them as a tool to achieve goals. In any case, someone should set the tasks for the neural network, and this will be a person, at least for now. Today, neural networks can help solve a number of tasks facing marketers, thereby facilitating their work, saving money and creating high-quality results. Yay, @julietde, well deserved! You have won $400! Congratulations again to all the nominations and winners. Visit contests.hackernoon.com today to view all the current and upcoming contests! These contests are for everyone! Share your stories and win $$$! It seems as if time flew - the final results announcement of the Growth Marketing Writing Contest by HackerNoon and mParticle is here! We thank mParticle for sponsoring this amazing contest for the HackerNoon community! During these 6 months, we published 200+ growth marketing stories generating around half a million reads and months of reading time! Growth Marketing Writing Contest by HackerNoon mParticle The Growth Marketing Writing Contest: Final Round Nominations & Winners For selecting the top 10 nominations, we picked all the stories with the #growth-marketingtag on HackerNoon, published in February 2023. Then we chose the top stories using 60:30:10 weightage respectively to: Number of hours read The number of people reached The freshness of the content Number of hours read The number of people reached The freshness of the content Top 10 Nominations How to Use ChatGPT for Marketing by @julietde AI and Chill: 7 Growth Marketing Tools to Streamline Your Workflow by @khamisihamisi Marketing with a Handicap: How to Promote Your Product on a Budget by @correspondentone How I Met My Girlfriend by Running Ads on Facebook: A Growth Marketer’s Playbook by @0xjack How to write SaaS comparison pages that beat the competition by @electriccopy AI in Marketing Ensures the Survival of Artists Who ‘Think Different’ by @nilanganray Growth Trends in Tech and Retail Industries by @anisimovapo Nobody Buys an NFT Just for Art by @sergey-baloyan 23 Free Digital Tools That Will Save You Thousands Of Hours by @khamisihamisi The Secret Plugin for ChatGPT3 by @alfredodecandia How to Use ChatGPT for Marketing by @julietde How to Use ChatGPT for Marketing @julietde AI and Chill: 7 Growth Marketing Tools to Streamline Your Workflow by @khamisihamisi AI and Chill: 7 Growth Marketing Tools to Streamline Your Workflow @khamisihamisi Marketing with a Handicap: How to Promote Your Product on a Budget by @correspondentone Marketing with a Handicap: How to Promote Your Product on a Budget @correspondentone How I Met My Girlfriend by Running Ads on Facebook: A Growth Marketer’s Playbook by @0xjack How I Met My Girlfriend by Running Ads on Facebook: A Growth Marketer’s Playbook @0xjack How to write SaaS comparison pages that beat the competition by @electriccopy How to write SaaS comparison pages that beat the competition @electriccopy AI in Marketing Ensures the Survival of Artists Who ‘Think Different’ by @nilanganray AI in Marketing Ensures the Survival of Artists Who ‘Think Different’ @nilanganray Growth Trends in Tech and Retail Industries by @anisimovapo Growth Trends in Tech and Retail Industries @anisimovapo Nobody Buys an NFT Just for Art by @sergey-baloyan Nobody Buys an NFT Just for Art @sergey-baloyan 23 Free Digital Tools That Will Save You Thousands Of Hours by @khamisihamisi 23 Free Digital Tools That Will Save You Thousands Of Hours @khamisihamisi The Secret Plugin for ChatGPT3 by @alfredodecandia The Secret Plugin for ChatGPT3 @alfredodecandia #Growth Marketing: Final Round Winners In the first place, we have: https://hackernoon.com/how-i-met-my-girlfriend-by-running-ads-on-facebook-a-growth-marketers-playbook?embedable=true https://hackernoon.com/how-i-met-my-girlfriend-by-running-ads-on-facebook-a-growth-marketers-playbook?embedable=true I probably had over 200+ submissions from interested prospects. I probably did Zoom calls with about 10 different women. And out of that, I met 4 in person. And the last woman I met, we’re still together. Sorry, no picture of her… yet. She said she’ll let me drop a press release if we get married. Of all the growth hacks I’ve implemented over the years, this is by far my best. After all, what’s the value of finding love? Priceless. I probably had over 200+ submissions from interested prospects. I probably did Zoom calls with about 10 different women. And out of that, I met 4 in person. And the last woman I met, we’re still together. Sorry, no picture of her… yet. She said she’ll let me drop a press release if we get married. Of all the growth hacks I’ve implemented over the years, this is by far my best. After all, what’s the value of finding love? Priceless. That was a fun read, @0xjack ! Congratulation on winning $1000! @0xjack The second position is won by: https://hackernoon.com/how-to-write-saas-comparison-pages-that-beat-the-competition?embedable=true https://hackernoon.com/how-to-write-saas-comparison-pages-that-beat-the-competition?embedable=true “Focusing on a best-fit prospect segment doesn't always make sense to people who don't have a marketing background. If I wanted to increase my chances of landing a customer, would I want to target as broad a market as possible? The reality is the exact opposite. The broader your focus, the more difficult it is to connect with prospects and convince them that your solution is the best one for them.” “Focusing on a best-fit prospect segment doesn't always make sense to people who don't have a marketing background. If I wanted to increase my chances of landing a customer, would I want to target as broad a market as possible? The reality is the exact opposite. The broader your focus, the more difficult it is to connect with prospects and convince them that your solution is the best one for them.” Congratulations, @electriccopy ! You have won $600! In the third place, we have: @electriccopy https://hackernoon.com/how-to-use-chatgpt-for-marketing?embedable=true https://hackernoon.com/how-to-use-chatgpt-for-marketing?embedable=true On the one hand, it seems that they can completely replace marketing specialists, but it is still worth considering them as a tool to achieve goals. In any case, someone should set the tasks for the neural network, and this will be a person, at least for now. Today, neural networks can help solve a number of tasks facing marketers, thereby facilitating their work, saving money and creating high-quality results. On the one hand, it seems that they can completely replace marketing specialists, but it is still worth considering them as a tool to achieve goals. In any case, someone should set the tasks for the neural network, and this will be a person, at least for now. Today, neural networks can help solve a number of tasks facing marketers, thereby facilitating their work, saving money and creating high-quality results. Yay, @julietde , well deserved! You have won $400! @julietde Congratulations again to all the nominations and winners. Visit contests.hackernoon.com today to view all the current and upcoming contests! These contests are for everyone! Share your stories and win $$$! contests.hackernoon.com

The Growth Marketing Writing Contest: Round 6 Results Announced!

গ্রোথ মার্কেটিং রাইটিং কনটেস্ট: রাউন্ড 6 এর ফলাফল ঘোষণা করা হয়েছে!

Der Wachstumsmarketing-Schreibwettbewerb: Ergebnisse der sechsten Runde bekannt gegeben!

Concurso de redacción sobre marketing de crecimiento: ¡se anuncian los resultados de la ronda 6!

Concours d'écriture sur le marketing de croissance : les résultats de la 6e ronde sont annoncés !

विकास विपणन लेखन प्रतियोगिता: छठवें दौर के परिणाम घोषित!

グロース マーケティング ライティング コンテスト: 第 6 ラウンドの結果が発表されました!

그로스 마케팅 글쓰기 콘테스트: 6차 결과 발표!

O Concurso de Redação de Growth Marketing: Resultados da 6ª Rodada Anunciados!

Büyüme Pazarlaması Yazma Yarışması: 6. Tur Sonuçları Açıklandı!

Cuộc thi Viết về Tiếp thị Tăng trưởng: Đã công bố Kết quả Vòng 6!

增长营销写作比赛：第 6 轮结果公布！

Heyo Growth Marketers - The round 5 results of the by and HackerNoon are now LIVE! Growth Marketing Writing Contest mParticle Here’s a quick reminder on how the contest works - You can win from $12,000 prize pool by writing a story on . Here are a few topic suggestions shared by mParticle: #growth-marketing Product Management Customer Data Platforms (CDP) Growth Marketing Data Engineering Data Management ! Submit your #growth-marketing story today The Growth Marketing Writing Contest: Round 5 Nominations & Winners For selecting the top 10 nominations, we picked all the stories with the #growth-marketingtag on HackerNoon, published in January 2022. Then we chose the top stories using 60:30:10 weightage respectively to: Number of hours read The number of people reached The freshness of the content Here are the top 10 nominations: by 6 More Growth Marketing Strategies to 10x Your Startup in 2023 @mrkrystallis by Press Mentions For Your Brand: A Growth Marketer's How-To @darragh by 10 Reasons Why I Sucked At Growth Marketing @nebojsa.todorovic by How to Use No-Code Machine Learning to Optimize your HackerNoon Articles @giorgiob by How to Unlock Product Market Fit and Rapid Growth for Startups @loringuyen by 7 Tips for Product Managers in Web3 @hunais by Behavioral Economics: Growth Hackers' Secret Weapon @aghafasih by How to Achieve 'Product of the Day' on Product Hunt Using a Trello Board @artkulakov by If Growth Marketing Is So Terrible, Why Don't Statistics Show It? @requiredcrx by Why Web3 and NFT Gaming Can Be the Biggest Trend of 2023 @sergey-baloyan In the first place, we have: https://hackernoon.com/press-mentions-for-your-brand-a-growth-marketers-how-to?embedable=true Quick Tips Building Relationships With Journalists: Don’t be self-promotional about ideas for a story; they hate that. Add-value: Follow them on socials and share and interact with content you like and can add value to. It’ll help you stand out. Engage online: Find out what they need for their story or story ideas; be helpful and add-value. Listen more than you talk. Listen: Congratulations, . You have won $1,000! @darragh In the second place, we have: https://hackernoon.com/6-more-growth-marketing-strategies-to-10x-your-startup-in-2023?embedable=true Lastly, here's an extra Growth Marketing Strategy you should never forget… Never stop testing! Constant testing is the backbone of any successful growth marketing strategy. Just make sure that your analytics are set up right so that you can analyze both your successes & failures equally and it won't be long till you discover a strategy with hockey stick growth potential that's worth doubling down on. Congratulations, . You have won $600!!! @mrkrystallis The story scoring the third position is: https://hackernoon.com/10-reasons-why-i-sucked-at-growth-marketing?embedable=true I used to be a humble and happy writer who one day had the Gordon Gekko “moment." You know what I'm talking about, don't you? I thought it was my revelation, but it turned out it was my damnation. I wanted more, plain and simple. It wasn't enough for me to be just a writer. Oh no, how about a writer-marketer?! More shiny titles, more clients. Sounds like a solid business growth plan. What could possibly go wrong? Well, it didn't take me too long to realize one simple growth marketing truth: Every marketer can be a writer; but not every writer can be a marketer. Marketing takes and asks so much more than just writing. So, I didn't get a dime for my rhyme. Well-deserved ! You have won $400. @nebojsa.todorovic Congratulations again to all the nominations and winners. Visit today to view all the current and upcoming contests! These contests are for everyone! Share your stories and win $$$! contests.hackernoon.com

The Growth Marketing Writing Contest: Round 5 Results Announced!

গ্রোথ মার্কেটিং রাইটিং কনটেস্ট: রাউন্ড 5 এর ফলাফল ঘোষণা করা হয়েছে!

Der Wachstumsmarketing-Schreibwettbewerb: Ergebnisse der fünften Runde bekannt gegeben!

Concurso de redacción sobre marketing de crecimiento: ¡se anuncian los resultados de la ronda 5!

Concours d'écriture sur le marketing de croissance : les résultats de la 5e ronde sont annoncés !

ग्रोथ मार्केटिंग राइटिंग कॉन्टेस्ट: राउंड 5 के परिणाम घोषित!

The Growth Marketing Writing Contest: 第 5 ラウンドの結果が発表されました!

그로스 마케팅 글쓰기 콘테스트: 5차 결과 발표!

O Concurso de Redação de Growth Marketing: Resultados da 5ª Rodada Anunciados!

Büyüme Pazarlaması Yazma Yarışması: 5. Tur Sonuçları Açıklandı!

Cuộc thi Viết về Tiếp thị Tăng trưởng: Đã công bố Kết quả Vòng 5!

增长营销写作比赛：第 5 轮结果公布！

Hackers, Assemble! The round 4 results of the by mParticle and HackerNoon are live! Growth Marketing Writing Contest Here’s a quick reminder on how the contest works - You can win from $12,000 prize pool by writing a story on . Here are a few topic suggestions shared by mParticle: #growth-marketing Product Management Customer Data Platforms (CDP) Growth Marketing Data Engineering Data Management . Submit your #growth-marketing story today The Growth Marketing Writing Contest: Round 4 Nominations & Winners For selecting the top 10 nominations, we picked all the stories with the #growth-marketingtag on HackerNoon, published in December 2022. Then we chose the top stories using 60:30:10 weightage respectively to: Number of hours read The number of people reached The freshness of the content Here are the top 10 nominations: by How to Build a Multi-Million Dollar Business from Scratch Without Investing in Marketing @antagonist by How I Made Programming Flashcards A Thing @tomaszs by What Product Team Structure is Right for You? @lazutkina by The Pointless TaskMaster: Why Product Management Might be a BS Job @eduardo.mignot by Why Web3 Projects Fail With Growth Marketing: 3 Fundamental Mistakes @web3mops by How to Find X When Your Product + X = Success @correspondentone by Growth Marketing for Dev Tools: Everything You Need to Know @annamarsx by How to Build a Community for your Blockchain Project @pavlokarapinka by Growth Loops; the Business Essential @theweb3diary by The Future of Growth Marketing: Predictions and Trends for 2023 and Beyond @chinechnduka The clear winner in first place with 40% of the votes is: https://hackernoon.com/how-i-made-programming-flashcards-a-thing?embedable=true Don't get me wrong, it's great that there are so many ways to learn programming. However, many people get lost in all this. Should I learn through a course, go to a bootcamp or buy a book? What language to learn? Everyone responds differently. If you add to this the authors of texts and courses who pour water instead of writing specifically and clearly, entering the world of programming seems very chaotic and difficult. That's why I found flashcards attractive, as opposed to the trend that discourages programming. You can fit very little information on a flashcard. However, this is still not interesting enough to reach for the flashcards. That's why I enriched them with sci-fi fantasy illustrations. In this way, they are beautiful and evoke Stoneheart or Magic The Gathering. Congratulations, @ , your storytelling is lit 🔥🔥🔥. You have won $1,000! tomaszs In the second place, we have: https://hackernoon.com/growth-loops-the-business-essential?embedable=true The growth loop for Netflix begins with providing a tailored viewing experience. Netflix is able to recommend new content and adapt its catalog to each individual user by utilizing data from customer viewing patterns and preferences. Customers are encouraged to explore more material and engage more deeply with the platform as a result of this. Congratulations, . You have won $600. @theweb3diary The story scoring the third position is: https://hackernoon.com/how-to-build-a-community-for-your-blockchain-project?embedable=true Forget about useless checklists from 2017 - they don't work anymore. Communicate with your audience, spark them with your ideas, and always ask what they want to see on your channels. If you don't know how to move or you get stuck at some stage, it doesn't matter. Any difficulties are solvable, and, as we have already demonstrated above, there are plenty of tools for promotion. Well-deserved ! You have won $400. @pavlokarapinka Congratulations again to all the nominations and winners. Visit today to view all the current and upcoming contests! These contests are for everyone! Share your stories and win $$$! contests.hackernoon.com

The Growth Marketing Writing Contest: Round 4 Results Announced!

গ্রোথ মার্কেটিং রাইটিং কনটেস্ট: রাউন্ড 4 এর ফলাফল ঘোষণা করা হয়েছে!

Der Wachstumsmarketing-Schreibwettbewerb: Ergebnisse der vierten Runde bekannt gegeben!

Concurso de redacción sobre marketing de crecimiento: ¡se anuncian los resultados de la cuarta ronda!

Concours d'écriture sur le marketing de croissance : résultats de la 4e ronde annoncés !

ग्रोथ मार्केटिंग राइटिंग कॉन्टेस्ट: राउंड 4 के नतीजे घोषित!

グロース マーケティング ライティング コンテスト: 第 4 ラウンドの結果が発表されました!

그로스 마케팅 글쓰기 콘테스트: 4차 결과 발표!

Concurso de Redação de Growth Marketing: Resultados da 4ª Rodada Anunciados!

Büyüme Pazarlaması Yazma Yarışması: 4. Tur Sonuçları Açıklandı!

Cuộc thi Viết về Tiếp thị Tăng trưởng: Đã công bố Kết quả Vòng 4!

增长营销写作比赛：第 4 轮结果公布！

Keeping the holiday spirit alive, here we are with more prizes for our fantastic community 🎁 🎁 Round 3 results of the by mParticle and HackerNoon are now live! Growth Marketing Writing Contest For those who are reading about this for the first time - mParticle & HackerNoon are excited to host for marketers and product managers. Growth Marketing Writing Contest Here’s your chance to win money from a whopping $12,000 prize pool! It could be any story on . We highly encourage you to write on #growth-marketing Product Management, Customer Data Platforms (CDP), Growth Marketing, Data Engineering, and Data Management. mParticle is a real-time AI customer data platform that powers your entire marketing stack with high-quality customer data. Request for a demo today! The Growth Marketing Writing Contest: November 2022 Nominations & Winners For selecting the top 10 nominations, we picked all the stories with the tag on HackerNoon, published in November 2022. Then we chose the top stories using 60:30:10 weightage respectively to: #growth-marketing Number of hours read The number of people reached The freshness of the content Here are the top 10 nominations: by Growth Marketing Vs Performance Marketing @dailyabay by 4 Steps to Build a Web Analytics Measurement Plan @mikhailkirilin by Nudge Theory Best-Practices For Business (With Examples) @darragh by 18 AI Marketing Softwares Your B2B Needs to Try Today @darragh by Secrets to Growth Marketing Data Engineering – Even in This Down Economy @manish8227 by 5 Steps to Increase Your Sales Department’s Efficiency @alexsolo by How Latin American Startups Can Make Their Own User Data Their Growth Superpower @marialejads by FTX Raised $1bn With This Pitch Deck in 2021. Let’s Review It! @sergey-baloyan by An Intro to Blockchain/Web3 Marketing in 2023 @dariavolkova by How to Use AI for Your B2B Marketing @darrag The winner in first place is: https://hackernoon.com/nudge-theory-best-practices-for-business-with-examples?embedable=true People value what they currently possess more than what they don’t yet own. We’re more motivated by potential losses than potential gains. For example, a majority of people will exert more effort to prevent the loss of $10,000 than to acquire $10,000 because the satisfaction lost is more overpowering than the satisfaction gained with the same amount. Wow , you have your three stories in the top 10 this month. Well-deserved win! You have won $1,000! @darragh In the second place, we have another excellent writeup: https://hackernoon.com/ftx-raised-$1bn-with-this-pitch-deck-in-2021-lets-review-it?embedable=true The first thing that catches your eye clearly is . There are no complicated backgrounds, animations, or illustrations. Just a light background and text. Nothing distracts from the content. Nothing should. the minimalistic design Psychologically, it is easier to make the reader focus on the things he needs to be focused on. In investment Pitch Decks, it's the numbers. And there will be numbers, lots of numbers. Congratulations, - you’ve won the second position and $600 in the growth marketing writing contest. @sergey-baloyan In the third place, we have: https://hackernoon.com/how-latin-american-startups-can-make-their-own-user-data-their-growth-superpower?embedable=true LatAm’s founders’ resilience is unmatched and the increasing adoption of tech solutions is a clear indicator of greater potential in the region and businesses’ willingness to adapt to the current digitization era. Early-stage startup data analytics tools help fuel investments and attract greater partnerships that consequently help the population uplift its quality of life. By showing well-informed data-based business strategies local founders and interested VCs are paving the way toward sustainable growth. Yay, ! You have won $400. @marialejads We will contact all the winners shortly. Want to know more about all the running and upcoming writing contests? Visit today! See y’all in round 4 of the growth marketing writing contest next month. contests.hackernoon.com

The Growth Marketing Writing Contest: Round 3 Results Announced!

Concurso de redacción sobre marketing de crecimiento: ¡se anuncian los resultados de la ronda 3!

Concours d'écriture sur le marketing de croissance : les résultats du troisième tour sont annoncés !

विकास विपणन लेखन प्रतियोगिता: तीसरे दौर के परिणाम घोषित!

グロース マーケティング ライティング コンテスト: 第 3 ラウンドの結果が発表されました!

O Concurso de Redação de Growth Marketing: Resultados da 3ª Rodada Anunciados!

Cuộc thi Viết về Tiếp thị Tăng trưởng: Đã công bố Kết quả Vòng 3!

增长营销写作比赛：第三轮结果公布！

Growth Marketers, Assemble! The second round results of the are here! Growth Marketing Writing Contest For those who are reading about this for the first time - mParticle & HackerNoon are excited to host a for growth marketers and product managers. Growth Marketing Writing Contest Here’s your chance to win money from a whopping $12,000 prize pool! It could be any story on . We highly encourage you to write on #growth-marketing Product Management, Customer Data Platforms (CDP), Growth Marketing, Data Engineering, and Data Management. mParticle is a real-time AI customer data platform that powers your entire marketing stack with high-quality customer data. Request for a demo today! The Growth Marketing Writing Contest October 2022 Nominations & Winners For selecting the top 10 nominations, we picked all the stories with the tag on HackerNoon, published in October 2022. Then we chose the top stories using 60:30:10 weightage respectively to: #growth-marketing Number of hours read The number of people reached The freshness of the content Here are the top 10 nominations: by Product Marketer vs. Growth Marketer vs. Content Marketer: How to Choose Your First Marketing Hire @oluwasegunoyebode by 16 Growth Marketing Hacks For Crypto Projects & Start-Ups @darragh by How Salesforce’s Partnership With Meta Gives Rise To Conversational Commerce @malcolmx by Growth Hacking Can Be the Difference Between Life and Death @scott-d.-clary by Microsoft: A Worthy Competitor Or A Killer Of Startups? @malcolmx by The Three Questions Every Entrepreneur Should Ask Themselves @scott-d.-clary by Why Visual Marketing is Still the Best Way to Get 69% of Customers @claidai by Why Professions Are Adding Analytics to Their Skillsets @jacquic by What is Growth Marketing? @juxtathinka by Product-Led Strategy That Guarantees Business Growth in Any Industry @samuelj The Growth Marketing Contest Round 2 Winners With 20k+ reads and a tag, the winner in the first place is: #hackernoon-top-story https://hackernoon.com/product-marketer-vs-growth-marketer-vs-content-marketer-how-to-choose-your-first-marketing-hire?embedable=true ‘Sounds like it's gonna be SEO spam but surprising turned out to be a well-researched article’ - HackerNoon Editor Congratulations, , you have won $1,000! @oluwasegunoyebode The second place is won by: https://hackernoon.com/growth-hacking-can-be-the-difference-between-life-and-death?embedable=true “Great interview and an interesting case study” - HackerNoon Editor Yay, ! You have won $600! @scott-d.-clary Last but not least, we have the following story in the third place: https://hackernoon.com/why-visual-marketing-is-still-the-best-way-to-get-69percent-of-customers?embedable=true ‘Data, and numbers!’ - HackerNoon Editor Congratulations, , you have won $400 and third place in the growth marketing writing contest! @claidai We will contact all the winners shortly. Want to know more about all the running and upcoming writing contests? Visit today! See y’all next month. contests.hackernoon.com

The Growth Marketing Writing Contest: Round 2 Results Announced!

Concurso de redacción sobre marketing de crecimiento: ¡se anuncian los resultados de la segunda ronda!

Concours d'écriture sur le marketing de croissance : résultats du deuxième tour annoncés !

विकास विपणन लेखन प्रतियोगिता: दूसरे दौर के परिणाम घोषित!

グロース マーケティング ライティング コンテスト: 第 2 ラウンドの結果が発表されました!

O Concurso de Redação de Growth Marketing: Resultados da 2ª Rodada Anunciados!

Cuộc thi Viết về Tiếp thị Tăng trưởng: Đã công bố Kết quả Vòng 2!

增长营销写作比赛：第二轮结果公布！

Growth marketers - the wait is OVER. The first round results announcement of the is now LIVE! Growth Marketing Writing Contest For those who are reading about this for the first time - mParticle & HackerNoon are excited to host a . Here’s your chance to win money from a whopping $12,000 prize pool! It could be any story on . We highly encourage you to write on Product Management, Customer Data Platforms (CDP), Growth Marketing, Data Engineering, and Data Management. Growth Marketing Writing Contest #growth-marketing mParticle is a real-time AI customer data platform that powers your entire marketing stack with high-quality customer data. Request for a demo today! The Growth Marketing Writing Contest September 2022 Nominations & Winners For selecting the top 10 nominations, we picked all the stories with the tag on HackerNoon, published in September 2022. Then we chose the top stories using 60:30:10 weightage respectively to: #growth-marketing Number of hours read The number of people reached The freshness of the content Here are the top 10 nominations: by How to Be Systematic When Testing Marketing Hypotheses @refocus by Why Your Business Requires Data Driven Growth-Marketing? @nemmanuel by Here’s Why Your TikTok Marketing is Failing @Oentoro by Top 5 Growth Hacks for Web3 @annicalin by Every Career Starts with a Leap of Faith: Navigating a Career in Content Creation @scott-d.-clary by Has SEO stopped being relevant in 2022 @minniemururi000 by Monetization Strategies to Help Your App Weather the Economic Downturn @lomitpatel by Why is the Documentation of Tech Products So Hard to Use? (from the User’s Point of View) @summerbud by No-Code AI In Marketing: A Shift In Mindset? @viceasytiger by Build vs. Buy: The AI Martech Conundrum @lomitpatel AND THE WINNERS ARE In the first place, we have: https://hackernoon.com/heres-why-your-tiktok-marketing-is-failing?embedable=true “A very insightful breakdown of what marketers might be doing wrong when approaching TikTok.” - HackerNoon Editor Congratulations @ , you have won $1,000! Oentoro The second place is won by: https://hackernoon.com/how-to-be-systematic-when-testing-marketing-hypotheses?embedable=true “A great post on how marketing is analytical and not all popularity contests and wild ideas that have no way of verifying their efficiency. The writer uses real 1st-hand accounts of their process and how it worked for them.” - HackerNoon Editor Well-deserved ! You have won $600. @refocus Last but not least, we have the following story in the third place: https://hackernoon.com/monetization-strategies-to-help-your-app-weather-the-economic-downturn?embedable=true “A really good breakdown of strategies for app developers in the recession and that's honestly something I’ve never read about anywhere.” - HackerNoon Editor Congratulations, @ , you have won $400 and third place in the growth marketing writing contest! lomitpatel We will contact all the winners shortly. Want to know more about all the running and upcoming writing contests? Visit today! contests.hackernoon.com See y’all next month.

The Growth Marketing Writing Contest: Round 1 Results Announced!

Concurso de redacción de marketing de crecimiento: ¡Resultados de la ronda 1 anunciados!

Concours d'écriture sur le marketing de croissance : résultats du premier tour annoncés !

ग्रोथ मार्केटिंग राइटिंग कॉन्टेस्ट: राउंड 1 के परिणाम घोषित!

グロース マーケティング ライティング コンテスト: ラウンド 1 結果発表!

O Concurso de Redação de Growth Marketing: Resultados da 1ª Rodada Anunciados!

Cuộc thi Viết Tiếp thị Tăng trưởng: Kết quả Vòng 1 được Công bố!

增长营销写作大赛：第一轮结果公布！

Audiences is one of the most flexible, powerful, and heavily leveraged tools within the mParticle platform. mParticle users can define audiences based on any user-related data captured from any source, be it a standard input like a web, iOS or Android app, or a partner feed. Once audiences are created, you can easily forward them to downstream activation partners to power a wide variety of use cases, like driving user engagement and app downloads or visualizing customer journeys, to name just a few. As Audiences is central to the value that customers derive from mParticle, this feature set is a high priority when it comes to performance optimization work. This is why recently underwent a project to overhaul Standard Audiences––a product that extends the functionality of regular Audiences. This article dives into how we conducted research to potential areas of optimization, executed on these opportunities, and evaluated the results. What is Standard Audiences? The feature on the mParticle platform allows you to define and build audiences based on historical user data, which is stored in Amazon S3, mParticle’s long-term data store. Standard Audiences Since Standard Audiences are based on events that customers perform over a long period of time, they are very useful for targeting subsets of users who have demonstrated high lifetime value. How are Standard Audiences calculated? Standard Audience calculations use the same audience engine as real-time audiences. The main difference lies in how the user profile and events data are fed into the audience engine. Real-time audience calculation is a stream processing job that reads real-time user events from a stream. On the other hand, the Standard Audience calculation is a batch job that reads from S3. A typical Standard Audience batch job needs to read many terabytes (up to a few petabytes) of data from S3. As such, it is obvious that we need to cut the whole job into smaller pieces and process them concurrently. A Standard Audience batch job can be represented as a DAG (Directed Acyclic Graph) that consists of many independent tasks, each of which is responsible for calculating audience membership for a small subset of users. In the end, the Standard Audience calculation results are stored in a unique location for each DAG in S3. The task dependencies, task status, and task workload specifications are managed by a home-grown batch job manager, which uses a MySQL database to store the DAG and task data and submits the tasks to AWS Batch for execution. Identify Opportunities for Optimization As the usage of offline audiences and the amount of user data on S3 grew organically, we noticed that some big Standard Audience DAGs consisting of hundreds of thousands of tasks could take six days to finish, which interrupted our customers’ workflows as they had to wait for almost a week for results. We set out to investigate the bottlenecks in the system and learn what we could do to improve performance. We added more metrics and logging to all steps in audience calculation task, and this is what we found: Finding 1: The Logic Behind API Calls Could be Optimized The audience calculation task makes API calls to a user profile service and batch job manager. When there are many concurrent calls, they can become sluggish or return failures. When a task fails, it will be retried up to four times. Since a task-level retry entails starting a task from the beginning, this is an expensive operation. : Solutions Reexamine the API call logic and make only the minimum number of calls necessary. Size the external services/databases properly to handle the workload incurred by the largest Standard Audience DAG. Add retry logic with exponential backoff with jitter to all API calls. Result: Reduced running time and cost by 20%. Finding 2: The Memory Requirement for Tasks Could be Reduced The memory requirement for each individual task (configured in AWS Batch job definition) is 30GB, which was determined by the worst case scenario. Most of the tasks do not need that much memory. We could run many more tasks concurrently on the same ECS cluster if we can reduce the memory requirement per task. : Solutions Reduce the default memory requirement to 7.5GB, and if a task fails because it ran out of memory, retry with 2x of the previously run’s memory requirement (capping at 30GB). Result: A 4x increase in the amount of tasks we could run on the same ECS cluster. A 2x increase in throughput, reducing the total DAG run time and cost by 50%. It is 2x, not 4x because when we run 4x containers using 7.5GB memory each container on the same EC2 instance, it takes them longer to finish and a small percentage of them needs to be retried with more memory. The end result is approximately 2x increase in total throughput. Finding 3: Make the Data-Finding Logic More Efficient The performance is IO bound. Most of the time is spent on retrieving data for the subset of users that an offline audience calculation task is responsible for. Data retrieval consists of two main steps: finding the data and downloading it. Finding data on S3 means ListBucket calls from many concurrent tasks, which is relatively fast but the aggregated time is significant. : Solutions Refactor the data-finding logic such that a small number of S3 ListBucket calls can find the data for all of the audience calculation tasks. Put this logic into a new task that all of the audience calculation tasks depend on. Result: A 30% average reduction in total DAG run time. Finding 4: Partition Sizes Could be Tuned We use EC2 spot instances in the ECS clusters managed by AWS Batch. This means that they could be terminated at any time. On one hand, we want each individual task to finish quickly so that when they get killed due to spot instance termination, we don’t waste much CPU time. On the other hand, we want each individual task to be fairly large because downloading a larger amount of data from S3 in one request incurs less overhead. This means that the partition size for individual tasks should be tuned. : Solutions Finding the optimal individual task partition size through experiments. Result: A 30% reduction in total DAG run time. Now let’s crunch the numbers: (1 - 20%)(1 - 50%)(1 - 30%)(1 - 30%) = 19.6%. This is how we reduced Standard Audience calculation time and cost by 80%. The result is very noticeable. Now most Standard Audience DAGs finish within 10 hours, with the maximum being approximately one day on DAGs from our biggest customers. We continuously invest in the infrastructure that drives our customer's outcomes, which is highly valuable. Also, these problems are really fun to solve, by very talented engineers. This project was at the intersection of Value, Fun, and Talent, which is very exciting! - Lee Morris, Senior Vice President of Engineering Data-Driven Optimizations In the solutions outlined above, we took a data-driven approach to finding the best parameters, such as partition size or initial memory requirement. Specifically, we reran several DAGs with different parameters, extracted the performance metrics, and then compared the results using Jupyter notebooks. We don’t guess. We let the data speak for themselves. Metrics and Monitoring In order to track the performance and cost of the Standard Audience DAGs, we added metrics to track events scanned per DAG, bytes read per DAG, total latency per task, and time spent by audience calculation engine per task, leveraging the our data services team built in house. automated cost reporting We implemented two versions of these metrics, one for internal monitoring and the other for customer facing metrics. For example, for events scanned per DAG, we track the actual total events scanned per DAG for internal monitoring (including events scanned by failed and then retried tasks), and track the total events scanned by successful task runs per DAG for customer billing purposes. If some tasks within a DAG got retried, this is something we should know about, but we should not charge customers for the cost of retries. By analyzing the performance metrics, we were able to establish SLOs for the Standard Audience DAGs. One of the key SLOs is total DAG latency of 24 hours. We set up DataDog monitoring for the SLOs so that we will get alerted if they are exceeded. Future Work After these optimizations, Standard Audience is still a highly IO-bound job. We believe that we can improve the performance further in the following ways. Build better indices for the user event data on S3, reducing the amount of data transfers. Add persistent checkpoints to the Standard Audience calculation task such that it can resume where it left off should it be killed due to spot instance termination. Devise better partitioning schemes to eliminate hot partitions. This would divide the work into more evenly sized partitions. We make it easy to get started and stand behind the value we deliver. That's why we offer an industry-leading, feature-rich, free trial for 30 days. ! Sign up now

An 80% Reduction in Standard Audience Calculation Time

Una reducción del 80% en el tiempo de cálculo de audiencia estándar

Stories, News & Mentions about MPARTICLE

Categories

Trending Topics

Stories, News & Mentions about MPARTICLE

Mparticle’s Stories on HackerNoon

Mparticle's News & Mentions

Categories

Trending Topics