In this project a model of an online consumer panel is proposed. The model was based on observations and empirical analysis.
Then an agent-based artificial panel was developed based on that model. For demostration purposes, the panel was subjected to stress and other modifications to analyse the possible behaviour of an actual panel under similar circunstances.
The presentation concludes with a short analysis of benefits and cost of this approach for online panel companies.
The document discusses retail audits and consumer panels as methods for collecting consumer information and measuring marketing strategy effectiveness.
A retail audit studies what consumers buy by gathering sales data from retailers, including volume, value, share, stock levels, and distribution. A consumer panel studies consumer behavior and attitudes by continuously tracking product purchases and in-store activities of a select group of consumers.
Both methods aim to understand purchasing habits and trends to measure factors like volume and value sales, price performance, distribution and stock levels, helping companies evaluate brand performance versus competitors. However, panels and audits face obstacles like costs, cooperation fatigue over time, and risk of biased or unrepresentative conclusions.
A customer panel is a group of people who agree to regularly participate in research studies for a company. Panels allow companies to get feedback on new products and ideas through short online surveys. Customers join panels to provide input and feel involved with the company. The panel enrollment process begins with customers voluntarily signing up and providing information. Panel members are then periodically invited to complete surveys on topics relevant to their interests to help guide the company's decisions.
This Six Sigma project aimed to reduce the number of outstanding Severity III tickets for an insurance support project. The project defined the problem as an average of 12-13 open tickets per week, risking impact on service level agreements. The critical-to-quality metric was to reduce open tickets to <=7 on average per week. Analysis identified key causes like attitude issues and lack of work planning. Improvement actions were implemented and data collection continued to monitor the sustained reduction in open tickets and increased process capability from 1.75 sigma to 6 sigma, resulting in estimated annual savings of $9,600.
1) The document discusses methods for increasing production efficiency through continuous improvement processes like Lean.
2) It outlines a 6-step process to identify opportunities, test changes, introduce improvements, measure results, and continually develop through additional cycles.
3) The goal is to increase machine efficiency by 25%, labor efficiency by 80%, reduce inventory by 40%, and costs by 24% over three years through engaging staff and eliminating losses.
The document discusses retail audits and consumer panels as methods for collecting consumer information and measuring marketing strategy effectiveness.
A retail audit studies what consumers buy by gathering sales data from retailers, including volume, value, share, stock levels, and distribution. A consumer panel studies consumer behavior and attitudes by continuously tracking product purchases and in-store activities of a select group of consumers.
Both methods aim to understand purchasing habits and trends to measure factors like volume and value sales, price performance, distribution and stock levels, helping companies evaluate brand performance versus competitors. However, panels and audits face obstacles like costs, cooperation fatigue over time, and risk of biased or unrepresentative conclusions.
A customer panel is a group of people who agree to regularly participate in research studies for a company. Panels allow companies to get feedback on new products and ideas through short online surveys. Customers join panels to provide input and feel involved with the company. The panel enrollment process begins with customers voluntarily signing up and providing information. Panel members are then periodically invited to complete surveys on topics relevant to their interests to help guide the company's decisions.
This Six Sigma project aimed to reduce the number of outstanding Severity III tickets for an insurance support project. The project defined the problem as an average of 12-13 open tickets per week, risking impact on service level agreements. The critical-to-quality metric was to reduce open tickets to <=7 on average per week. Analysis identified key causes like attitude issues and lack of work planning. Improvement actions were implemented and data collection continued to monitor the sustained reduction in open tickets and increased process capability from 1.75 sigma to 6 sigma, resulting in estimated annual savings of $9,600.
1) The document discusses methods for increasing production efficiency through continuous improvement processes like Lean.
2) It outlines a 6-step process to identify opportunities, test changes, introduce improvements, measure results, and continually develop through additional cycles.
3) The goal is to increase machine efficiency by 25%, labor efficiency by 80%, reduce inventory by 40%, and costs by 24% over three years through engaging staff and eliminating losses.
1. A/B testing involves splitting users into a test group that sees a new feature and a control group that sees the original version. Users are typically assigned randomly with equal percentages to each group.
2. Metrics are tracked for both groups to measure the impact of the new feature, such as additions to cart or purchases. Significance testing determines if results are likely real or due to chance.
3. Long-term experiments require holding back some users from the test to measure novelty or learning curve effects over time.
Using process thinking to define project scope: how to start in the right placeSamuel Chin, PMP, CSM
Process work can be one of the most difficult types of project to scope, because "process" itself can be anything and everything. But without a clear starting point, your process improvement projects will suffer and you may find yourself solving problems which don't actually move the needle or need to be solved. In this meetup, we discussed how you can apply the concept of process detail layers to your project scoping phase, to ensure that you establish clear boundaries for your work from the outset and are able to focus on the right things.
This document provides an introduction and overview of Six Sigma. It begins with a brief history of Six Sigma, noting its origins at Motorola in 1981 in response to Japanese competition. It then discusses some key Six Sigma concepts, including that it is a highly disciplined process to develop near-perfect products and services, it aims for 3.4 defects per million opportunities, and that it is a philosophy, statistical measurement, business strategy, and project management framework. The document then covers some differences between Six Sigma and traditional business excellence approaches. It also provides definitions of some common Six Sigma terms. Finally, it discusses the Define, Measure, Analyze, Improve, and Control project framework and causes and effects analysis tools used in Six Sigma
QuestBack is a Norwegian company founded in 2000 that provides enterprise feedback management software and services. It has over 3,000 customers in more than 50 countries. The company's vision is to help businesses improve relationships with customers, employees, and other stakeholders by enabling them to collect, analyze, and act on feedback. A key feature is linking feedback collection (asking) with follow up actions (acting) based on the results. The software offers templates, customization options, reporting, and other tools to streamline the feedback process. Pricing is based on an annual subscription starting at 8,500 euros.
This document provides information about a module on managerial economics taught by Professor Almudena Sevilla at Queen Mary University of London. It outlines the module content, structure, learning outcomes, assessment, and resources. The module uses lectures, seminars, cases studies and problems to teach microeconomic concepts and their application to business decision making. It aims to provide students with an economic perspective on management and policy issues. Assessment includes a midterm test, final exam, and case study analyses. Main textbooks and additional readings are listed.
13 0806 webinar q & a financial analysis and planningCleantechOpen
This document contains a summary of questions and answers from webinars on financial analysis and planning for Cleantech Open participants. Key topics discussed include: defining average selling price and monetizing electronic data flows; addressing low-margin but reliable industries; outsourcing manufacturing versus capital expenditures; estimating costs and revenues for software and pre-revenue companies; and using financial advisors. Mentors provided guidance on getting feedback, meeting deadlines, and formatting responses.
The SID (SOC Intermediary Desk) acts as a bridge between first and second tier support teams in the SOC (Security Operations Center) environment. It aims to handle all incidents and cases within 24 hours and 12 hours respectively. This is to ensure that important cases are prioritized and unnecessary work is not passed to the SOC teams. Implementing the SID is expected to improve customer experience metrics like FCR, NPS, and reduce process times based on a VOC analysis showing current poor ratings. It will combat current challenges by properly filtering and prioritizing cases. Its success is forecast to increase positive feedback and decrease negative feedback over time.
This presentation was given at the Sustainable Brands Africa Conference in May 2016. It provides case studies and lessons learnt of conducting numerous impact assessments. It also provides advice of how to conduct impact assessments, what indicators to consider and how to determine return on investment
Australian speech pathologists in private practice: reduce stress by building...Speechiesinbusiness
This document provides guidance to Australian speech pathologists on establishing a basic compliance system to reduce stress and non-attendance issues. It recommends appointing a compliance officer and following 11 steps: 1) identify key issues and stakeholders, 2) establish foundations and scope, 3) identify risks, 4) understand regulations, 5) implement policies and training, 6) set escalation processes, 7) regularly update the system. It also uses a case study to discuss using compliance systems to analyze non-attendance data, communicate policies, and monitor clients. The overall goal is to provide a organized, evidence-based approach to reduce chaos and firefighting.
Estimate and Measure. Minimize work, maximize value. Part 2Shiftup
Discover in this deck different output and outcome metrics, have an overview of popular impact metrics and get a link to an estimation exercise.
Want to attend our next webinar? Become a Shiftup Explorer: https://shiftup.work/product/explorer-agility-innovation-qualification-program/
Seedcamp is an organization that provides mentorship and seed funding to early-stage technology startups in Europe. It receives over 1,500 applications each year and invests in about 20-25 companies. The document discusses Seedcamp's process of selecting startups and providing lean startup methodology training through mini bootcamps and mentorship. It also provides examples of startups in Seedcamp's portfolio that have successfully implemented customer development and iterative product building.
Unleashing the Power of User Experience (UX): Bridging the Gap between Market...Anurag Pandey
This document discusses various topics related to user experience (UX) design including the difference between marketers and creators, components of user experience like trust and culture, conducting user interviews, latent needs, UX journey mapping, and Norman's model of emotional design. It provides explanations of UX research models like fishbone diagrams and the importance of understanding users through their experience journey. Interview tips are provided along with explanations of various UX design tools like empathy mapping. Resources for learning more about UX design are listed at the end.
This document provides an overview of results-based accountability (RBA) and the RBA 101-OBA 101 training. It includes slides on distinguishing between results, indicators, and performance measures. It outlines the "20-60-20 rule" and presents examples of population-level and program-level results, indicators, and performance measures. The document also describes the Turn the Curve exercise and provides templates for documenting its outcomes. Key aspects of the exercise are to establish a baseline, discuss causes and potential solutions, and identify low-cost action steps.
The document describes a project to predict customer churn and upgrade opportunities using product usage data. A random forest model was developed to classify customers as either churned or upgraded and not churned within one year. The model achieved AUCs of 0.764 and 0.882 respectively. The company was able to gain insights into important factors like early user activity signals and increase monthly recurring revenue over 100% through targeting upgrades. Further improvements to the model were proposed, including additional product and user data.
This document discusses the importance of financial modeling for startups. It explains that financial models can help visualize business plans, identify key metrics that impact revenue, and understand the path to profitability. The document provides examples of modeling different revenue streams like ads, subscriptions, and marketplaces. It emphasizes testing assumptions with real data and updating models based on tracking key performance indicators like acquisition, activation, retention, and revenue. The overall message is that financial modeling provides a compass for startups to evaluate hypotheses and guide business decisions with an understanding of what metrics most directly impact growth and success.
Agile Portugal 2015 - Agile: The Power of I(n)terationNuno Rafael Gomes
Agile: The Power of I(n)teration, an introduction :-)
The Power of i(n)teration in Scrum: Cultural Flow, Compound Interest, Knowledge and Value.
– Agile Manifesto and Agile Origins: Lean and Systems Thinking.
– A powerful concept: Cultural Flow.
– The Gap between Culture and Patterns and how to address it? Theory of Constraints (TOC) and some real life examples from the trenches.
– Agile Values and Interaction & Iteration. Compound Interest. Another powerful concept: Compound Knowledge. An example.
– Retrospective :-)
This document provides information about an upcoming Grant Writing Masterclass on August 16th 2019 presented by Lorraine Acheson. The masterclass will provide guidance to attendees on writing successful applications for R&D grant funding, covering topics such as: types of available funding, assessing project suitability for different grants, eligibility criteria, searching for funding opportunities, and answering assessors' questions to impress reviewers. The goal is to help attendees develop competitive applications that clearly demonstrate how their projects align with funders' priorities and represent good value for money.
The document discusses the systems development life cycle (SDLC) process for planning, creating, testing, and deploying information systems. It describes the main stages of the SDLC as preliminary analysis, systems analysis, systems design, development, integration and testing, acceptance and deployment, maintenance, and evaluation. It also discusses problems with the traditional sequential SDLC model such as long development cycles and difficulty accommodating changes. The incremental waterfall model is presented as an alternative that develops the system in smaller incremental releases.
End-to-end pipeline agility - Berlin Buzzwords 2024Lars Albertsson
We describe how we achieve high change agility in data engineering by eliminating the fear of breaking downstream data pipelines through end-to-end pipeline testing, and by using schema metaprogramming to safely eliminate boilerplate involved in changes that affect whole pipelines.
A quick poll on agility in changing pipelines from end to end indicated a huge span in capabilities. For the question "How long time does it take for all downstream pipelines to be adapted to an upstream change," the median response was 6 months, but some respondents could do it in less than a day. When quantitative data engineering differences between the best and worst are measured, the span is often 100x-1000x, sometimes even more.
A long time ago, we suffered at Spotify from fear of changing pipelines due to not knowing what the impact might be downstream. We made plans for a technical solution to test pipelines end-to-end to mitigate that fear, but the effort failed for cultural reasons. We eventually solved this challenge, but in a different context. In this presentation we will describe how we test full pipelines effectively by manipulating workflow orchestration, which enables us to make changes in pipelines without fear of breaking downstream.
Making schema changes that affect many jobs also involves a lot of toil and boilerplate. Using schema-on-read mitigates some of it, but has drawbacks since it makes it more difficult to detect errors early. We will describe how we have rejected this tradeoff by applying schema metaprogramming, eliminating boilerplate but keeping the protection of static typing, thereby further improving agility to quickly modify data pipelines without fear.
1. A/B testing involves splitting users into a test group that sees a new feature and a control group that sees the original version. Users are typically assigned randomly with equal percentages to each group.
2. Metrics are tracked for both groups to measure the impact of the new feature, such as additions to cart or purchases. Significance testing determines if results are likely real or due to chance.
3. Long-term experiments require holding back some users from the test to measure novelty or learning curve effects over time.
Using process thinking to define project scope: how to start in the right placeSamuel Chin, PMP, CSM
Process work can be one of the most difficult types of project to scope, because "process" itself can be anything and everything. But without a clear starting point, your process improvement projects will suffer and you may find yourself solving problems which don't actually move the needle or need to be solved. In this meetup, we discussed how you can apply the concept of process detail layers to your project scoping phase, to ensure that you establish clear boundaries for your work from the outset and are able to focus on the right things.
This document provides an introduction and overview of Six Sigma. It begins with a brief history of Six Sigma, noting its origins at Motorola in 1981 in response to Japanese competition. It then discusses some key Six Sigma concepts, including that it is a highly disciplined process to develop near-perfect products and services, it aims for 3.4 defects per million opportunities, and that it is a philosophy, statistical measurement, business strategy, and project management framework. The document then covers some differences between Six Sigma and traditional business excellence approaches. It also provides definitions of some common Six Sigma terms. Finally, it discusses the Define, Measure, Analyze, Improve, and Control project framework and causes and effects analysis tools used in Six Sigma
QuestBack is a Norwegian company founded in 2000 that provides enterprise feedback management software and services. It has over 3,000 customers in more than 50 countries. The company's vision is to help businesses improve relationships with customers, employees, and other stakeholders by enabling them to collect, analyze, and act on feedback. A key feature is linking feedback collection (asking) with follow up actions (acting) based on the results. The software offers templates, customization options, reporting, and other tools to streamline the feedback process. Pricing is based on an annual subscription starting at 8,500 euros.
This document provides information about a module on managerial economics taught by Professor Almudena Sevilla at Queen Mary University of London. It outlines the module content, structure, learning outcomes, assessment, and resources. The module uses lectures, seminars, cases studies and problems to teach microeconomic concepts and their application to business decision making. It aims to provide students with an economic perspective on management and policy issues. Assessment includes a midterm test, final exam, and case study analyses. Main textbooks and additional readings are listed.
13 0806 webinar q & a financial analysis and planningCleantechOpen
This document contains a summary of questions and answers from webinars on financial analysis and planning for Cleantech Open participants. Key topics discussed include: defining average selling price and monetizing electronic data flows; addressing low-margin but reliable industries; outsourcing manufacturing versus capital expenditures; estimating costs and revenues for software and pre-revenue companies; and using financial advisors. Mentors provided guidance on getting feedback, meeting deadlines, and formatting responses.
The SID (SOC Intermediary Desk) acts as a bridge between first and second tier support teams in the SOC (Security Operations Center) environment. It aims to handle all incidents and cases within 24 hours and 12 hours respectively. This is to ensure that important cases are prioritized and unnecessary work is not passed to the SOC teams. Implementing the SID is expected to improve customer experience metrics like FCR, NPS, and reduce process times based on a VOC analysis showing current poor ratings. It will combat current challenges by properly filtering and prioritizing cases. Its success is forecast to increase positive feedback and decrease negative feedback over time.
This presentation was given at the Sustainable Brands Africa Conference in May 2016. It provides case studies and lessons learnt of conducting numerous impact assessments. It also provides advice of how to conduct impact assessments, what indicators to consider and how to determine return on investment
Australian speech pathologists in private practice: reduce stress by building...Speechiesinbusiness
This document provides guidance to Australian speech pathologists on establishing a basic compliance system to reduce stress and non-attendance issues. It recommends appointing a compliance officer and following 11 steps: 1) identify key issues and stakeholders, 2) establish foundations and scope, 3) identify risks, 4) understand regulations, 5) implement policies and training, 6) set escalation processes, 7) regularly update the system. It also uses a case study to discuss using compliance systems to analyze non-attendance data, communicate policies, and monitor clients. The overall goal is to provide a organized, evidence-based approach to reduce chaos and firefighting.
Estimate and Measure. Minimize work, maximize value. Part 2Shiftup
Discover in this deck different output and outcome metrics, have an overview of popular impact metrics and get a link to an estimation exercise.
Want to attend our next webinar? Become a Shiftup Explorer: https://shiftup.work/product/explorer-agility-innovation-qualification-program/
Seedcamp is an organization that provides mentorship and seed funding to early-stage technology startups in Europe. It receives over 1,500 applications each year and invests in about 20-25 companies. The document discusses Seedcamp's process of selecting startups and providing lean startup methodology training through mini bootcamps and mentorship. It also provides examples of startups in Seedcamp's portfolio that have successfully implemented customer development and iterative product building.
Unleashing the Power of User Experience (UX): Bridging the Gap between Market...Anurag Pandey
This document discusses various topics related to user experience (UX) design including the difference between marketers and creators, components of user experience like trust and culture, conducting user interviews, latent needs, UX journey mapping, and Norman's model of emotional design. It provides explanations of UX research models like fishbone diagrams and the importance of understanding users through their experience journey. Interview tips are provided along with explanations of various UX design tools like empathy mapping. Resources for learning more about UX design are listed at the end.
This document provides an overview of results-based accountability (RBA) and the RBA 101-OBA 101 training. It includes slides on distinguishing between results, indicators, and performance measures. It outlines the "20-60-20 rule" and presents examples of population-level and program-level results, indicators, and performance measures. The document also describes the Turn the Curve exercise and provides templates for documenting its outcomes. Key aspects of the exercise are to establish a baseline, discuss causes and potential solutions, and identify low-cost action steps.
The document describes a project to predict customer churn and upgrade opportunities using product usage data. A random forest model was developed to classify customers as either churned or upgraded and not churned within one year. The model achieved AUCs of 0.764 and 0.882 respectively. The company was able to gain insights into important factors like early user activity signals and increase monthly recurring revenue over 100% through targeting upgrades. Further improvements to the model were proposed, including additional product and user data.
This document discusses the importance of financial modeling for startups. It explains that financial models can help visualize business plans, identify key metrics that impact revenue, and understand the path to profitability. The document provides examples of modeling different revenue streams like ads, subscriptions, and marketplaces. It emphasizes testing assumptions with real data and updating models based on tracking key performance indicators like acquisition, activation, retention, and revenue. The overall message is that financial modeling provides a compass for startups to evaluate hypotheses and guide business decisions with an understanding of what metrics most directly impact growth and success.
Agile Portugal 2015 - Agile: The Power of I(n)terationNuno Rafael Gomes
Agile: The Power of I(n)teration, an introduction :-)
The Power of i(n)teration in Scrum: Cultural Flow, Compound Interest, Knowledge and Value.
– Agile Manifesto and Agile Origins: Lean and Systems Thinking.
– A powerful concept: Cultural Flow.
– The Gap between Culture and Patterns and how to address it? Theory of Constraints (TOC) and some real life examples from the trenches.
– Agile Values and Interaction & Iteration. Compound Interest. Another powerful concept: Compound Knowledge. An example.
– Retrospective :-)
This document provides information about an upcoming Grant Writing Masterclass on August 16th 2019 presented by Lorraine Acheson. The masterclass will provide guidance to attendees on writing successful applications for R&D grant funding, covering topics such as: types of available funding, assessing project suitability for different grants, eligibility criteria, searching for funding opportunities, and answering assessors' questions to impress reviewers. The goal is to help attendees develop competitive applications that clearly demonstrate how their projects align with funders' priorities and represent good value for money.
The document discusses the systems development life cycle (SDLC) process for planning, creating, testing, and deploying information systems. It describes the main stages of the SDLC as preliminary analysis, systems analysis, systems design, development, integration and testing, acceptance and deployment, maintenance, and evaluation. It also discusses problems with the traditional sequential SDLC model such as long development cycles and difficulty accommodating changes. The incremental waterfall model is presented as an alternative that develops the system in smaller incremental releases.
Similar to Online Consumer Panel simulator - demo 2: Exercise (20)
End-to-end pipeline agility - Berlin Buzzwords 2024Lars Albertsson
We describe how we achieve high change agility in data engineering by eliminating the fear of breaking downstream data pipelines through end-to-end pipeline testing, and by using schema metaprogramming to safely eliminate boilerplate involved in changes that affect whole pipelines.
A quick poll on agility in changing pipelines from end to end indicated a huge span in capabilities. For the question "How long time does it take for all downstream pipelines to be adapted to an upstream change," the median response was 6 months, but some respondents could do it in less than a day. When quantitative data engineering differences between the best and worst are measured, the span is often 100x-1000x, sometimes even more.
A long time ago, we suffered at Spotify from fear of changing pipelines due to not knowing what the impact might be downstream. We made plans for a technical solution to test pipelines end-to-end to mitigate that fear, but the effort failed for cultural reasons. We eventually solved this challenge, but in a different context. In this presentation we will describe how we test full pipelines effectively by manipulating workflow orchestration, which enables us to make changes in pipelines without fear of breaking downstream.
Making schema changes that affect many jobs also involves a lot of toil and boilerplate. Using schema-on-read mitigates some of it, but has drawbacks since it makes it more difficult to detect errors early. We will describe how we have rejected this tradeoff by applying schema metaprogramming, eliminating boilerplate but keeping the protection of static typing, thereby further improving agility to quickly modify data pipelines without fear.
ViewShift: Hassle-free Dynamic Policy Enforcement for Every Data LakeWalaa Eldin Moustafa
Dynamic policy enforcement is becoming an increasingly important topic in today’s world where data privacy and compliance is a top priority for companies, individuals, and regulators alike. In these slides, we discuss how LinkedIn implements a powerful dynamic policy enforcement engine, called ViewShift, and integrates it within its data lake. We show the query engine architecture and how catalog implementations can automatically route table resolutions to compliance-enforcing SQL views. Such views have a set of very interesting properties: (1) They are auto-generated from declarative data annotations. (2) They respect user-level consent and preferences (3) They are context-aware, encoding a different set of transformations for different use cases (4) They are portable; while the SQL logic is only implemented in one SQL dialect, it is accessible in all engines.
#SQL #Views #Privacy #Compliance #DataLake
Global Situational Awareness of A.I. and where its headedvikram sood
You can see the future first in San Francisco.
Over the past year, the talk of the town has shifted from $10 billion compute clusters to $100 billion clusters to trillion-dollar clusters. Every six months another zero is added to the boardroom plans. Behind the scenes, there’s a fierce scramble to secure every power contract still available for the rest of the decade, every voltage transformer that can possibly be procured. American big business is gearing up to pour trillions of dollars into a long-unseen mobilization of American industrial might. By the end of the decade, American electricity production will have grown tens of percent; from the shale fields of Pennsylvania to the solar farms of Nevada, hundreds of millions of GPUs will hum.
The AGI race has begun. We are building machines that can think and reason. By 2025/26, these machines will outpace college graduates. By the end of the decade, they will be smarter than you or I; we will have superintelligence, in the true sense of the word. Along the way, national security forces not seen in half a century will be un-leashed, and before long, The Project will be on. If we’re lucky, we’ll be in an all-out race with the CCP; if we’re unlucky, an all-out war.
Everyone is now talking about AI, but few have the faintest glimmer of what is about to hit them. Nvidia analysts still think 2024 might be close to the peak. Mainstream pundits are stuck on the wilful blindness of “it’s just predicting the next word”. They see only hype and business-as-usual; at most they entertain another internet-scale technological change.
Before long, the world will wake up. But right now, there are perhaps a few hundred people, most of them in San Francisco and the AI labs, that have situational awareness. Through whatever peculiar forces of fate, I have found myself amongst them. A few years ago, these people were derided as crazy—but they trusted the trendlines, which allowed them to correctly predict the AI advances of the past few years. Whether these people are also right about the next few years remains to be seen. But these are very smart people—the smartest people I have ever met—and they are the ones building this technology. Perhaps they will be an odd footnote in history, or perhaps they will go down in history like Szilard and Oppenheimer and Teller. If they are seeing the future even close to correctly, we are in for a wild ride.
Let me tell you what we see.
Codeless Generative AI Pipelines
(GenAI with Milvus)
https://ml.dssconf.pl/user.html#!/lecture/DSSML24-041a/rate
Discover the potential of real-time streaming in the context of GenAI as we delve into the intricacies of Apache NiFi and its capabilities. Learn how this tool can significantly simplify the data engineering workflow for GenAI applications, allowing you to focus on the creative aspects rather than the technical complexities. I will guide you through practical examples and use cases, showing the impact of automation on prompt building. From data ingestion to transformation and delivery, witness how Apache NiFi streamlines the entire pipeline, ensuring a smooth and hassle-free experience.
Timothy Spann
https://www.youtube.com/@FLaNK-Stack
https://medium.com/@tspann
https://www.datainmotion.dev/
milvus, unstructured data, vector database, zilliz, cloud, vectors, python, deep learning, generative ai, genai, nifi, kafka, flink, streaming, iot, edge
STATATHON: Unleashing the Power of Statistics in a 48-Hour Knowledge Extravag...sameer shah
"Join us for STATATHON, a dynamic 2-day event dedicated to exploring statistical knowledge and its real-world applications. From theory to practice, participants engage in intensive learning sessions, workshops, and challenges, fostering a deeper understanding of statistical methodologies and their significance in various fields."
The Ipsos - AI - Monitor 2024 Report.pdfSocial Samosa
According to Ipsos AI Monitor's 2024 report, 65% Indians said that products and services using AI have profoundly changed their daily life in the past 3-5 years.
Analysis insight about a Flyball dog competition team's performanceroli9797
Insight of my analysis about a Flyball dog competition team's last year performance. Find more: https://github.com/rolandnagy-ds/flyball_race_analysis/tree/main
Open Source Contributions to Postgres: The Basics POSETTE 2024ElizabethGarrettChri
Postgres is the most advanced open-source database in the world and it's supported by a community, not a single company. So how does this work? How does code actually get into Postgres? I recently had a patch submitted and committed and I want to share what I learned in that process. I’ll give you an overview of Postgres versions and how the underlying project codebase functions. I’ll also show you the process for submitting a patch and getting that tested and committed.
The Building Blocks of QuestDB, a Time Series Databasejavier ramirez
Talk Delivered at Valencia Codes Meetup 2024-06.
Traditionally, databases have treated timestamps just as another data type. However, when performing real-time analytics, timestamps should be first class citizens and we need rich time semantics to get the most out of our data. We also need to deal with ever growing datasets while keeping performant, which is as fun as it sounds.
It is no wonder time-series databases are now more popular than ever before. Join me in this session to learn about the internal architecture and building blocks of QuestDB, an open source time-series database designed for speed. We will also review a history of some of the changes we have gone over the past two years to deal with late and unordered data, non-blocking writes, read-replicas, or faster batch ingestion.
Beyond the Basics of A/B Tests: Highly Innovative Experimentation Tactics You...Aggregage
This webinar will explore cutting-edge, less familiar but powerful experimentation methodologies which address well-known limitations of standard A/B Testing. Designed for data and product leaders, this session aims to inspire the embrace of innovative approaches and provide insights into the frontiers of experimentation!
1. Alex’s life in SIMU LTD
Using simulations for
online consumer panels
operations analysis
advances of a model proposal
2. Project
● Previous work
■ slideshare.net/evaristoc
■ nl.linkedin.com/pub/evaristo-caraballo
● Objectives
○ to introduce concepts; sketch the model
■ Advances and modifications on version 2.0
○ to show potential and scope of a simulation
tool using a fictitious case
3. What is the company’s model?
Welcome to
SIMU LTD:
the Online
Panel!!
4. SIMU LTD asks people to subscribe at SIMU
LTD under company’s costs...
If they subscribe they will
receive emails inviting to
respond online
surveys in a portal...
… they could answer or not...
SIMU Ltd.
SURVEY Co .
5. SIMU LTD works hard in
keeping enough people “in
residence” willing to answer
the surveys of the client,
SURVEY Co., as well as
improving gross margin and
profits.
The set of subscribers
responding surveys is known
as THE PANEL.
SIMU LTD always strives to
understand how subscribers
turn into a better panel.
6. And this is Alex!
Alex has voluntarily joined SIMU
LTD. to answer surveys online.
Alex receives a “reward” for the
efforts.
As subscriber at SIMU LTD Alex is
usually referred as PANELLIST.
Alex shares certain gender and
age with other panellists
(“neighbours”) and has a demo-letter
(“a”, “b”, “c” or “d”) randomly
assigned.
7. Alex’s behavioural model (sketched)
Loyalty (desire to
stay)
- Comparison with other
panels
- Quality of service and other
perceived rewards
- Satisfaction; Joy; “What is
there for me?”
- Tenure (proxy)
Response (desire to initiate
and continue)
- Size and type of reward per survey
- Questionnaire length
- Quality of present and previous
questionnaire experience
- Topic, Frequency
Effort (“maximum per
week”; “yield”)
- `Planned´dedication time and
budget; Necessity, Availability
- Channel, Presentation (Exposure;
Accessibility)
- Season or other external events
All proposed variables
could interact to define
the response rate model
A simple probabilistic model based mainly on tenure and
effort was considered for this project
8. Alex’s response while active
u-Strategy m-Strategy
The more the expectations are
matched, the higher response
rate, tenure and max effort...
Unmatched expectations
Low effort-rr-tenure
(150 min/wk = 15 surv max)
● Alex is rational, knows what to expect and has already a
Matched expectations
High effort-rr-tenure
(300 min/wk = 30 surv max)
“strategy”
○ response is a function linked to a pre-assigned active tenure
● Assumptions and Constraints:
○ Alex is sensitive to number of surveys in portal up to a maximum
○ Alex is immune to number/quality of invites
○ the longer in the panel, the lower the initial* response
(*) Reflects empirical data
9. Nothing last forever and Alex will leave the
panel at some point, but when?
This is the period where Alex answers
some invitations
During inactive period Alex stops completely.
SIMU LTD automated system doesn’t know
exactly when this happens: it just “purges”
Alex after some consecutive weeks of non-responses
First Invite Last Response Purged
... ...
Active Tenure
(including seasonal
inaction)
(unsubscribed)
Inactive Tenure
(non-response no matter
number of invites)
Total Tenure
X X
NOTES: panellists reach a “Inactive Tenure” period
where they are most likely ignoring any invitation to
participate in the panel
10. Each member of any cohort* has specific
tenure, so panellists leave the panel at
different time
TIME wk
01
We begin
with a
relatively
large group
of recruits...
wk
02
wk
n-4
wk
n-3
wk
n-2
wk
n-1
wk
n
… many of
them leave
very soon…
others stay
longer...
… there are
these few
panelists
that last
longer ...
… reaching
a lower
churn rate
...
...
...
Alex Alex Alex
Alex Alex Alex
Alex
(*) Cohort: panellist that started same day
11. The mix of panellists who are active
or inactive at a given moment and
who could potentially receive
invitations to participate are all
together named...
THE (available) UNIVERSE
12. Invitation Model*
(*) For a sketch refer to a summary of version 2.0 at slideshare.net/evaristoc
1- In our model all panellists in the universe queue in a
random order to “respond” an also randomised list of
surveys for that “week”...
2- By turns, each panellist will be
exposed only to those surveys that
are still uncomplete, for which they
are selected or not excluded; we
say that they have a “portal”; they
can open the portal or not...
3- The process ends when all available
weekly surveys are completed or when the
instances in the universe are exhausted (i.e.
each panellist was presented once with the
respective portal)
13. What for questionnaires does
Alex find? SURVEY Co. projects
Ad-Hocs:
❏ One-time project that
last 1 week
❏ 70 completes
❏ Incidence: 20%
❏ Selection by demo
letter
❏ No Exclusion
Trackers:
❏ Same weekly project
for 1-3 years
❏ 40 completes
❏ Incidence: 10%
❏ Selection (fixed along
the project duration)
❏ Exclusion by previous
response status
Both last “10 min” and SIMU LTD gives 0,10 money-equivalent
per complete to the panellists
14. Now let’s talk about BUSINESS...
Issues being explored with the exercise
1. How does this particular model help to
determine a panel capacity* that could
affect business performance?
(*) Eg. max number of surveys that the panel would answer at a given time
2. Insights about business implications
when introducing changes that could
modify, e.g., panel response behaviour?
15. SIMU LTD. has been operating for
5 years to date delivering Survey
Co.’s surveys
Started with ad-hocs, adding
trackers from year 2
See weekly KPI developed for this
exercise in Annex A
Let’s see how SIMU
LTD has been doing...
OBS: SIMU LTD can correctly estimate
performance and invoice accordingly!
16. SIMU LTD. performance (5 years)
During the 5 years of operation
and with a fixed recruitment
effort SIMU LTD universe
trends to stabilise after the first
year.
OBS: in this project seasonality
was not evaluated
After stabilising, SIMU LTD
universe showed a weekly low
response rate (average num.
touches/num.surveys) for an
average of 11 surveys/week
per panellist’s portal.
17. SIMU LTD. performance (5 years)
With a stable revenue/cost ratio,
SIMU LTD keeps close to the
ideal gross margins*.
Higher portion of costs (75%)
due to survey costs.
Although only average 7% of
total costs, recruitment costs
threat earnings by growing at
10% per year!
(*) gross margins if not need of
buying PP services
Excluding first year, only 2% of
assigned completes were re-sold
to PP (average 2500
completes/year).
18. Increasing Gross Margins: How?
SIMU LTD has ideas for the next 2 years:
● Sales: doubling / tripling number of
weekly surveys after base price cuts
(see Annex B)
● Panellists: a campaign to extend the
permanence of u-strategists
(see Annex C)
Here the results of the analysis...
19. Sales increase by price cuts
Although gross margins
doubled, SIMU LTD wouldn’t
match the ideal gross margins
(only 60% during 7y).
Going from an average 75% to a
85% of total costs, survey costs
had a large impact...
… but now non-covered
completes become a source of
losses:
For every 30 additional
surveys/wk (every added 100k
completes/year) the fail-to-complete
quintuplicates.
20. And how did the respondents?
Simu ltd. typified respondents according to their
(active) tenure in bad, mid and good, took a
weekly sample of just-inactivated panelists at the 3
periods (1-5y, 6y and 7y). Then analysed the
performance of each respondent type per period.
Simu ltd. found that the
completion rate decreased,
no matter the respondent
type. The most affected was
the response of the good type.
The response rate also decreased, BUT actual
contribution of bad respondents lowered
substantially at each following period because their
lowered response rate AND their lower bound to accept
surveys per week, that forced a maximum.
Better respondents took the effort of weekly
checking the “leftover” of surveys other panelists “didn’
t have time to see”. Better respondent saw even more
surveys when the number of surveys tripled.
21. Sales + Targeting u-strategists
- The campaign affecting u-strategists will always increase median response rate and universe
size
- However the campaign has an insignificant and even negative effect over gross margins
whether combined with any kind of sales increase or not
- If no other changes are possible, doubling/tripling without campaign appears the best risk
to take when compared to the other proposals
22. What are SIMU LTD take-aways?
● Negotiating sample engagement (eg tenure,
“professionals”) composition based on
clients demands?
■ type of responsiveness and engagement is a
sample quality attribute for some researches!
● How OPERATIONS (eg. increase in surveys
per week) could affect the responsiveness?
● When thinking in a loyalty campaign, which
panellist(s) type should be targeted? why?
how? What is the best panellist model?
■ this point could be an STRATEGIC one…
23. Final Remarks
● Combines decision-making arguments
into single scenario visualizations
● When?
○ testing our assumptions; “what if…?”
○ taking existing tools a step further
○ prone to automation => good for reporting
○ explaining, even predicting
■ R&D... Training...
○ as support tool if actions/analyses are
costly, risky, complex
■ eg. “augmenting” vars./data; advanced stat.
○ data analysis, reliability, validity a pré
28. ● Cut is always over Base Price per complete at end
of previous year
● First cut (year 6) is for 90%; second cut (year 7) is
for 75%
● Doubling/Tripling means:
○ Doubling/Tripling number of fixed number of trackers at year
5 (7) as well as average ad-hocs at year 5 (24)
● The cut and the increase in sales take place
immediately at the beginning of each year
● Constraints
○ Only price is affected; the rest keeps unchanged
○ PP are also competitors; they offer similar prices per
complete as SIMU LTD
○ PP won’t match the price cut, keeping the same base price as
at year 5 during years 6 and 7
30. ● Based on fictitious few weeks A/B experiment
under average last 4y conditions
● Targets are instances with 1-2 wks active tenure
● The campaign:
○ initial investment (10k) to be paid at the beginning of year 6
○ “reward” (6,- units) for new recruits who respond at least
once per wk, 3wks in a row, given at the end of week 3
● After starting experiment SIMU LTD expects:
○ Increase of 1-3.0% more respondents at second week
○ Increase of 3-5.0% at the third
● Constraints
○ Because SIMU LTD doesn’t know who the u-strategists are the
campaign is generic
○ Those instances already programmed for a tenure longer than
2wks are believed to be immune
○ It is assumed that the campaign doesn’t affect response rate:
only tenure