Ask your data
  • Blog
  • About & Contact

Building your own fortune teller: An Introduction to Next Best Action Models

3/28/2023

0 Comments

 
Picture


​by Christina Köck
​
Data Scientist at joint systems

​What is Next Best Action?

Who does not sometimes wish for a mentor who always has an appropriate advice? In general, humans are very good at rating known situations and making the right decisions, but in case of new circumstances or a huge amount of aspects to consider, good advice is priceless. In marketing and similarly in fundraising, one-on-one support is very effective, but not possible in most cases. As companies and organizations were growing, mass marketing became the state of the art, trying to sell as many items to as many customers as possible. Customer received overwhelming amounts of advertisement and offers, which can be described as a “content shock”. Since the rise of machine learning and the possibility for sophisticated data analysis, companies try to stand out by putting the customer in the center. This is done by trying to predict what the customer wants and needs. New technologies should advise marketing specialists what to do next in order to satisfy every customer individually – supported by enormous amounts of data. This concept is called Next Best Action (NBA) or Next Best Offer (NBO) and can evolve into a useful fortuneteller in case of effective implementation. The benefits of the concept are reduced advertisement cost and increased customer loyalty.

How can NBA systems be developed?

The goal, structure and architecture of a NBA system strongly depends on the environment and purpose it is used for. Questions to answer before the development are for example:
  • Which parameters should be optimized?
  • Level of detail: What should be predicted? Only the next best action or the best time for this action or even the channel?
  • Which actions are available? How are these actions prioritized?
  • Which data are available? Consider ethics and privacy issues - which data can be used?
  • What technical capabilities and hardware/software options are available?
  • Where and how will the system be used? Directly during calls from/to customers? For planning newsletters and personal advertisements?

After goals and actions are defined, the general structure of a NBA system can be summarized as follows: 
Picture
Structure of a NBA system: Actions are presented as letters; colors illustrate different categories of actions. Source: https://databasedmarketing.medium.com/next-best-action-framework-47dca47873a3
​The process of step one - filter- is rather rule based, depending on what actions or offers make sense for a customer. For instance, there might be products, which cannot be offered more than one time and therefore their offer will be excluded from the action list after purchase. The same applies for the third step: Typically, the actions with the highest benefit should be chosen but it depends on the user of the system, whether more than one action is taken or the actions need to be filtered again. A reason for a second filtering step could be that only one single action of a certain category is chosen even though two actions from the same category show very high benefit values. As it turns out, most NBA systems are based on proven business rules and requires a well-defined framework before the most complex component – the rating of the actions - can be developed.
​
Scoring Systems for NBA​
​
​The centerpiece of the system is the scoring engine where the actual NBA is identified. The scoring can happen via different levels of data involvement and machine learning:
  • Rule-based (Level 0)
  • Use of (existing) scores (Level intermediate)
  • Full Deep Learning Model  (Level high)
 

Picture
A rule-based system requires a lot of experience in the domain and takes time to develop because it is all done by hand: Rule like “rate action xy high if the customer already bought from the category ab in the last two years” can be data driven if they are based on previous findings. Most companies already use such business rules for marketing. The scoring does not need to be metric (with numbers), as shown in the picture below. The system is not adaptive and needs to be adjusted, if the circumstances change. Rule-based systems are characterized by extreme simplification, since the real, very complex interrelationships cannot be completely cast into rules. The needs and the reactions of a customer will change depending on the previous action, which is very difficult to represent in a rule-based way.
​
Picture
Example for a rule based system. Experience and detailed development is needed to incorporate sufficient rules
​If a low or intermediate level of machine learning is considered, relatively simple scores and predictions can be incorporated into the system. The simplest approach would be to calculate the probability for all customers to react positively to an action (propensity). To just predict which action might provoke a positive reaction might not be suitable for predicting long term customer loyalty. Several scores and probabilities can be combined to achieve this goal. For example, the propensity can be combined with a prediction of long term customer engagement. The reason for this is that on one hand no action is useful if the customer does not like the action (propensity), but on the other hand the actions also need to have a benefit for the organization (prediction of customer engagement/success for the organization). The scores can be weighted according to their importance and any number of further adjustments can be made. Typically, such a system will still be embedded in several predefined rules. Similar like factors for weighing scores (by multiplication), a higher importance of a score can be expressed by adding meaningful numbers to the score.
Disadvantages are that still a lot of experience in the field of use is needed. Scores are required for all actions and they need to indicate the benefit or propensity to the same extent.
The system could automatically adjust to a certain extend if the scores themselves are linked to current data and regularly monitored.

​
Picture
Full Deep Learning Models require more abilities that are technical, experience with deep learning and an appropriate technical environment. The difference to the previous systems is that these models are black box-models without the necessity for the users to tweak and define the rules and scores themselves. While intermediate levels of machine learning require a programmer to intervene to make adjustments, in deep learning the algorithms themselves determine whether their decisions are right or wrong. Internally, the algorithm will calculate scores for each action but will determine the best calculation and weighing itself. Rating systems in this category can range from a deep neural network for predicting benefits of actions to reinforcement learning models, which do not need a lot of training data in advance but can learn implicit mechanisms by trial and error. As long as these systems are fed with current data, they are adaptable. Disadvantages of this approach are the complex development and reduced interpretability due to the fully automated process.
​
Picture
Fully automated deep learning model.
So What?

In our view, George Box´ good old quote “All models are wrong, but some are useful” is also worth considering in the context of NBA models. Whether and how these approaches suit the needs of respective (nonprofit) organizations should be evaluated holistically. If you are interested in getting to know more about the concept of NBA or even think about implementing it in your SOS association, please do not hesitate to get in touch with us. 


Sources
https://www.altexsoft.com/blog/next-best-action-marketing-machine-learning
https://databasedmarketing.medium.com/next-best-action-framework-47dca47873a3
https://medium.com/ft-product-technology/how-we-calculate-the-next-best-action-for-ft-readers-30e059d94aba
https://www.mycustomer.com/marketing/strategy/opinion-five-rules-for-the-next-best-offer-in-marketing


0 Comments

Attribution Modelling for Online Marketing and Fundraising

1/3/2023

0 Comments

 
Picture
by Carolina Pelegrin Cuartero
Data Scientist at joint systems

​
Non-profit organisations (NPOs) use offline-marketing strategies to both attract the attention of potential donors as well as collect donations. With mostly donors from “aging” generations reacting to direct campaigns and because those campaigns are slowly decreasing its impact, NPOs are investing into new ways of collecting donations, including online platforms. Not only online media has the potential to reach a wider profile of users; also, other advantages exist, including traceability, the option to gather user-level data and the fact that customers can potentially be better addressed over online channels. In addition, online media have proven to have a greater effectiveness in terms of customer conversion than traditional advertising.

With increasing inflation and life costs and shrinking marketing budgets, it is crucial for NPOs to understand their customer interests and behaviour to better and more effectively address them. If talking about online marketing platforms, this would translate into study what online sources (like email links, YouTube ads or paid search) are the ones bringing more potential donors to our websites.

Each person accessing a website to actually buy a product or donate something, may have visited the website before to research about the (donation) product, before finally deciding on converting or making a donation. Obviously, it is interesting to know the influence of each online source in a final donation through a website; this is known as the attribution problem and can be solved by using attribution-modelling techniques. Different attribution-modelling techniques exist, which try to predict the importance of the different marketing channels in the total conversion of customers (or donors). While advertisers tend to use very simplistic, heuristic ones, academia has focused on more complex and data-driven methods, which have been proven to obtain better results. 

Using data from the visits and donations done during the year 2021 on the website of a SOS organisation, we studied the effect that different marketing channel sources (i.e. "from where do users access a website") had on donations. Analysis were done on Jupyter Notebooks using Python and two different modelling approaches: 
​
Picture
  • Last-touchpoint attribution as a heuristic approach, which attributes all the merit of a donation to the last touchpoint of the user or the “channel used to access the website at the point of donation”. While several heuristic approaches exist, this is the one most widely used in practice.
  • Markov chains as a data-driven approach, which can attribute donations across all the different touchpoints on the user path, as well as take into account the sequence order of the donor path. 

Picture

​​​After doing some first data exploration and filtering data representing an “abnormal” visiting behaviour (visits to the career site and those done on #GivingTuesday), we applied both attribution-modelling methods. 

​From our analysis, it could be easily concluded that different online marketing channels do have different effects on donations and donation revenues. 

However, and to our surprise, minimal differences were found between LTA and MC results for both, estimated number of donations and revenue per channel, probably caused by a “special” behaviour of donors: as it seems, most of them will visit the website just once during the year and decide, on the fly, if they would like to donate something. This overrepresentation of paths with just one touchpoint causes both heuristic and data-driven methods to assign the same (donation) value to the channels. 
Picture
While results are similar, we find that working with data-driven methods as Markov Chains still has advantages, including the fact that:
  • They can calculate the so-called removal effects, which show the importance of each channel, i.e. the percentage of conversions that would get lost if removing that channel from the paths. This is very important whenever we are trying to better split our marketing budget between channels!
  • They can be represented as transition matrix, which show the expected donation probability for each channel under study into donation (“CONVERSION”). 
  • They show an overall better performance in terms of AUC values. 
Picture
Do you want to know more about our results and insights? Are you also interested in a similar study? Do you have online data but are not making use of it?
Get in touch with us! 💻📱📧
We at joint systems can help you get the most out of your data 🙂😉
0 Comments

The State of Nonprofit Data Science | Where is the sector in 2022?

10/11/2022

0 Comments

 
Some 3.5 years ago we discussed the state of data science in the nonprofit sector in this blog post. The world has significantly changed since then, however, being an insight-driven (nonprofit) organization is more imperative than ever. So, what is actually the status quo of data science and analytics in the nonprofit sector?
​
The most comprehensive survey on the state of data science and machine learning is the annual Machine Learning and Data Science Survey conducted by the platform Kaggle.com. In 2021, almost 26.000 people took part all across the globe. The participants were also asked about the industry they currently work in, Luckily, survey designers had added "Nonprofit & Services" as an option for the mentioned industry-related question. This enabled us to download the full survey response dataset from the Kaggle website. Using a global filter to focus on the responses from the nonprofit sector, we managed to put together this dashboard:


​Back in 2019, when we last blogged about the status of data science in the nonprofit sector, we had already started our joint journey with our customers and partners. Still, we are continuous learners. However, together with our clients, we managed to write numerous success stories on how data science and analytics can make fundraising more efficient and successful. If you want to learn more, please go ahead and browse through the free resources we offer on our platform analytical-fundraising4sos.com or watch the video below for some inspiration.


We wish you all the best in these turbulent and challenging times. Let´s keep in touch and jointly make the most of fundraising data!

​Johannes
0 Comments

The Science and Art of Data Storytelling

8/9/2022

0 Comments

 
The world has become - and maybe always was to an extent - a volatile, uncertain, complex and ambiguous place (VUCA). This has turned the communication of complex topics and interdependencies into an urgent need. Data is ubiquoutous but let alone it is useless unless converted into information and ultimately knowledge. This is where the concept of Data Visualization or more holistically ideas from the field Data Storytelling can make crucial contributions. According to Dr. Jennifer Aaker, an American behavioural scientist at the University of Stanford, stories are remembered up to 22 times more than facts alone.

Often, data storytelling is simply considered an effective data visualization. In fact, the practise of creating data stories is a structured approach with the goal to communicate data insights as an interplay of the three elements data, visuals and narrative. Creating convincing narrative visualizations not only requires the skills of data analysis experts but al so the knowledge from designers, artists and psychologists employing certain techniques, following specific structures, frameworks and using tools.

The creator of the visual may not be able to put his or her intention into explicit knowledge since successful data visuals are often “a matter of taste” and creating representations of data visually hence a subjective process . This suggests that communicating data driven insights to decision makers requires data storytellers, who are skilled in the “art of data storytelling" which can best be learned through best practises. Data Storytelling is essentially about an "art and science mindset".

Lisa Oberascher, data analyst and alumna of 
Management, Communication & IT at MCI Innsbruck recently finished her bachelor thesis about Data Storytelling. Lisa conducted interviews with data practicioners and put together a great Data Storytelling cheat sheet.

Bild
The respective PDF can be downloaded here:
Download Data Storytelling Cheat Sheet by Lisa Oberascher
File Size: 3042 kb
File Type: pdf
Datei herunterladen

We wish you a great summer and happy data-storytelling :-)
Best, Johannes
0 Comments

An Outlook for 2022: Analytics and Data Science in Fundraising

12/22/2021

0 Comments

 
Data is becoming more and more pervasive across industries. Analytics has come quite some way in recent years. A growing number of organizations have implemented analytics solutions and started exploring the potential of data science. With continuing technological advances and accelerating digitization, it is not always easy to overview the current developments in advanced analytics and data science. This end-of-year post tries to provide readers with information in a nutshell on contemporary issues in analytics and data science from a nonprofit and fundraising standpoint. 
The infographic below is a "one-pager" for decision makers, analysts, data scientists and anyone interested. We differentiate between the topics that seem to be here to stay and relevant trends that should definitely be considered. In addition, we drop some hyped buzzwords that might be topics for the future and are worth observing. Please feel free to download, share, comment etc.
​
Bild

It has been a challenging but also inspirational year for many of us. We wish you and your dear ones a happy and peaceful Christmas 2021 and a good start in a successful, healthy and happy 2022.
These are our Christmas wishes in dozens of languages ​🎄!
​
All the best and see you in 2022!

0 Comments

Why should Nonprofits care about the Ethical Use of Algorithms?

9/14/2021

0 Comments

 
Bild

Be it private or professional life, algorithms seem to be everywhere. There is definitely a ​growing pervasiveness of them in society. The increasing importance of algorithms calls for raised awareness reagarding the potential ethical implications of their use. In this month´s blogpost, we will take a closer lot at this topic.

Algorithms Everywhere!

The effects of the COVID-19 pandemic acted as an accelerator for digitalization in terms of processes, services, or whole business models. Digital technologies are transforming the economy and are becoming ubiquitous. An increasingly widespread application of algorithms is decision-making in businesses, governments, or society as a whole. Algorithms might, for instance,  determine who is recruited and promoted, who is provided a loan or housing, who is offered insurance, or even which patients are seen by doctors. Algorithms have become important actors in organizational decision making, i.e. a field that has traditionally been exclusive to humans. As these decisions often have an ethical dimension, the delegation of roles and responsibilities within these decisions deserves scrutiny. This is where Coporate Responsibility comes into play ...

Luckily, as social nonprofit organizations work in the interest of the common good in one way or the other way, Corporate Responsibilty tends to be rooted in the "DNA" of nonprofits. At the same time, ​algorithms have also made their way into the sector of fundraising nonprofit organization as we had already highlighted in specific a blogpost from 2019. Compared to other contexts such as human resource management or the labour market (see for example this critical discussion of the algorithm used at the Austrian Labour Market agency "AMS"), the consequences of algorothmic decision making in the context of fundraising nonprofits will tend to be rather harmless. However, in the light of technological advances and the need for nonprofits acting as as active members of modern society that have a voice, NPO decision makers should be aware of the big picture in terms of "Ethical AI".

Implications and Challenges
​

In the course of scrutinizing the ethics of algorithms, not only considering the algorithms themselves but also their actual implementation in software and platforms should be scrutinized. Two groups of concerns can be identified in terms of the ethical challenges implied by algorithms. there are epistemic concerns on the one hand when evidence provided by algorithms is inconclusive, inscrutable, or misguided. On the other hand, there are normative concerns related to unfair outcomes, transformative effects and traceability. These normative concerns have in common that they are related to the actions derived from algorithmic results. In a nutshell, the mentioned concerns can be summarized as follows:
Bild
leant against Mittelstadt, B. D., Allo, P., Taddeo, M., Wachter, S., & Floridi, L. (2016). The ethics of algorithms: Mapping the debate. Big Data & Society, page 4

So what? Three things nonprofit decision makers can do (at least)
  • Make efforts and invest in explainable Machine Learning, i.e. make models that tend to be "black boxes" but provide predictive power more transparent and intelligible for stakeholders. Earlier this year, my colleague Carolina Pelegrin published a great introduction on explainable ML that can be found here.
  • Be aware of general limitations of methods, particularly when it comes to "buzzy" topics like Artificial Intelligence. Have, for instance, a look at the "6 myhts regarding AI" in one or our blogposts from 2019 that can be accessed here.
  • Keep on dealing with and discussing ethical AI. There are great resources around such as an AI Ethics Toolkit for Nonprofits by Nethope, a technology initiative and exchange platform driven by the world´s largest nonprofit organizations.

Any questions or input? Let´s keep in touch!
​

We wish you a smooth start in a hopefully pleasant and successful fall of 2021.

All the best!
​Johannes
​
0 Comments
<<Previous

    Categories

    Alle
    Artificial Intelligence
    Because It´s Fun!
    Churn
    Clustering
    Data Sciene @NPO
    Data Visualization
    Facebook
    Machine Learning
    Maps
    Natural Language Processing
    Neural Nets
    Power BI
    Predictive Analytics
    Social Media
    Time Series
    Twitter

    Archive

    Januar 2021
    November 2020
    September 2020
    August 2020
    Mai 2020
    April 2020
    Februar 2020
    Dezember 2019
    November 2019
    September 2019
    Juni 2019
    April 2019
    März 2019
    Januar 2019
    Dezember 2018
    Oktober 2018
    August 2018
    Juni 2018
    Mai 2018
    März 2018
    Februar 2018
    Dezember 2017
    November 2017
    Oktober 2017
    September 2017
    August 2017
    Juli 2017
    Mai 2017
    April 2017
    März 2017
    Februar 2017
    Januar 2017

About

Copyright © 2018
  • Blog
  • About & Contact