Open Password – Wednesday, May 19, 2020
#924
Steep templates 2020 – Emmet Kilduff – Eagle Alpha – Use of External Data – Enhancing Decision Making – Case Studies – Chipotle – LinkUp – Peer Group Analysis – Hiring Strategy – Retailer – Sentiment Analaysis – Machine Learning – Consumer Acceptance – Consumer Insights – Shell – Waitrose – Geolocation Data – Competitive Intelligence – Lowe – Home Depot – Store Locations – Coupon Promotion – Quantum Technology – Disruptions – Solving Optimization Problems – Simulation of Complex Structures – Artificial Intelligence – Cryptography – Roland Berger – Corporate Strategies – Martin Streichfuß – Michael Alexander – Automotive – Pharma – Chemical Industry – Frederik Hammermeier – Financial sector – Transport industry – Public funding – Venture capital – Cloud of labor research – Connect & Collect – ZBW – Regional competence centers – Klaus Daughtermann – Isabella Peters – Educational mess in schools – Makerspace with teachers – Educational platforms at the state and local level – Joachim Paul – Educational media – Conference of Ministers of Education and Cultural Affairs – Digital educational content – Communication problems – Reporting – YouTube – Media libraries of the major broadcasters – BMBF – WirLernen Online – MUNDO – Digital Pact for Schools – Cooperations
1.
Cover story: How Innovative Corporate Use External Data
to Enhance Decision Making (2) – Case Studies
2.
Before new disruptions through quantum technology – solving optimization problems, simulating complex structures, artificial intelligence and cryptography
“Work Research Cloud”
4.
Outside the box: Against the educational muddle in schools
Steep assists 2020:
Keynote by Emmet Kilduff, Founder and CEO of Eagle Alpha
How Innovative Corporate Use External Data
to Enhance Decision Making (2)
Case studies
External data can be used for several purposes, eg:
– Digital media companies use alternative data to predict and improve employee retention rates.
– Logistics company uses alternative data to predict disruptions to customers’ supply chains.
– Agricultural giant launches new services to help predict and optimize crop yields.
– Grocer uses alternative data to improve forecasting and reduce stockouts.
– Credit union personalizes marketing offers for members based on customer profiles developed by help of alternative data.
__________________________________________________________________________________
Case Study 1: Honing Chipotle’s Hiring Strategy With Employment Data
__________________________________________________________________________________
Purpose: People Insights
Relevant Department: Procurement, Marketing, Product, Sales
Chipotle is an American chain of fast casual restaurants in the US, UK, Canada, Germany and France, specializing in Mexican food. The company was struggling to fill positions in restaurants at specific locations in the US. The company used LinkUp’s job postings data to understand the reasons for the hiring difficulties. LinkUp indexes millions of job postings every day from employer websites, to deliver job market information.
A peer group analysis was carried out on the fast-casual restaurant industry and Chipotle’s hiring strategy was compared to its peers. It was clear that job titles and job descriptions were a key factor in Chipotle’s hiring struggles.
Employment data has many value add use cases for corporates. Job postings data can be used to evaluate corporate strategy, growth rates, employee churn, employee location, part-time/full-time, gender diversity and demand for specific skills. Employee sentiment data can provide insights into internal opinion on management and sentiment by job type.
__________________________________________________________________________________
Case Study 2: Analyzing Consumer Trends And Sentiment With Online & Social Media Content
_________________________________________________________________________________
Purpose: Customer Insights, Competitive Intelligence, Product Development
Relevant Department: Marketing, Product, Sales
Another consumer sentiment use case involved analysis on a leading online retailer. The retailer was examined using deep sentiment analysis incorporating machine learning to identify positive versus negative sentiment. The analysis was able to directly identify building negative sentiment by consumers towards the online retailer. Analysis of the themes of the negative sentiment allowed the company to pinpoint the key driver, being poor service and follow-up. The customer success and marketing teams utilized these insights to implement key actions to improve the customer experience.
A social media listening tool was used to monitor unstructured text from 90 million web sources including blogs, forums, news sites and social media. It produced an output that was structured and used to analyze potential changes in consumer trends and sentiment. Eagle Alpha analyzed new product launches for Apple and Samsung. Data gathered online and from social media platforms showed poor consumer acceptance of the new phone launches, leading to weak Samsung Galaxy S9 and iPhone sales.
Most mentions show US and Europe Twitter mentions, and China blogs and forum mentions, over the 2016, 2017 and 2018 iPhone cycles. Lower mentions in 2018 compared to 2017 was an early indication that consumers were less than enthusiastic about the new iPhone offerings. This data can show fundamental issues that may influence consumer attitudes to a company or brand.
Analysis of the themes of the negative sentiment allowed the company to pinpoint the key driver, being poor service and follow-up. The customer success and marketing teams utilized these insights to implement key actions to improve the customer experience.
Key takeaway. Online and social media content can provide valuable customer insights
eg understand brand perception, consumer reaction to new product/product feature launches and customer behavior. Customer insights provide customer success , marketing and product teams valuable ammunition to derive the best value from customers.
__________________________________________________________________________________
Case Study 3: Shell & Waitrose – Using Geolocation To Understand The Impact of Co-Location
__________________________________________________________________________________
Purpose: Customer Insights, Product Dev. / Relevant Department Marketing
Fuel retailer Shell and groceries brand Waitrose in 2012 announced a partnership that would see Waitrose outlets co-located with Shell garages. Delays meant that it wasn’t until 2018 that store upgrades began at pace. Huq analyzed the geolocation data to understand the impact on Shell customers’ visitor frequency and in turn Shell’s exposure to fluctuating fuel prices.
During the period when little or no convenience facilities were present, the frequency with which Shell customers visited correlates very closely (0.736) to fuel price. It can be inferred that when filling up is the only reason to visit and fuel prices rise, consumers make more trips, while filling up less.
From June 2018 on, as Shell’s proposition diversifies through their partnership with Waitrose and others, the frequency with which customers visit their outlets becomes significantly more consistent through 2019 and is almost entirely disassociated with fuel price (0.736 -> 0.001). This study suggests that the Waitrose partnership reduces Shell’s retail exposure to fuel price.
Key Takeaway: Geolocation data allowed Shell to quantify the impact of diversifying away from fuel. The same analysis could be applied on competitors to provide key competitive intelligence .
__________________________________________________________________________________
Case Study 4: How Big Data Identified Where The Next Lowe’s Should Go
__________________________________________________________________________________
Purpose: Customer Insights, Product Development, Market/Competitive Intelligence, Acquisitions
Relevant Department: Marketing, Product, Sales, Board/M&A
Lowe’s is an American retail company specializing in home improvement and competes with the country’s largest home improvement retailer, Home Depot. In this use case we consider store location as a key consideration in retail wars.
A leading geolocation vendor, Orbital Insight, utilized location intelligence paired with market share, trade area and demographic analysis to identify Home Depot locations ripe for Lowe’s to cannibalize. The Lowe’s brand is 30 years older than Home Depot and despite this, its store sales have historically lagged those of Home Depot and it does not attract traffic on the scale of its rival.
Orbital Insights analyzed the issue as follows:
- Understand the areas of fastest population growth in the US utilizing core-based statistical areas (CBSA). A CBSA consists of one or more counties (or equivalents) anchored by an urban center of at least 10,000 people plus adjacent counties that are socioeconomically tied to the urban center by commuting.
- Identify the number of Home Depot and Lowe’s locations in each top CBSA. The analysis showed Lowe’s lagged Home Depot in 18 out of 20 fastest growing CBSAs.
- Analyze and compare the trading areas of Home Deport and Lowe’s. Quantitative geolocation data tracked store visits through mobile devices from each census block group, providing insights into trade area overlaps or uncontested
- Understand the trade area stats including population within trade areas, trade area overlap by company and demographics (wealth & density). The analysis showed that Lowe’s trade areas tend to be slightly more wealthy and lower in population density.
Geolocation analysis was also used to evaluate foot traffic at the targeted Home Depots to understand how the stores were actually performing.
- Apply the ideal demographics, market share and trade area overlap to discover new Lowe’s store locations. The analysis concluded that Lowe’s stores should be located in Home Depot trade areas that:
- Are within 30% of Lowe’s historical median wealth target
- Are within 30% of Lowe’s historical median population
- Have more than 10 minutes driving distance from an existing Lowe’s
- Are in a trading area where Lowe’s has less than 30% market
Key Takeaway: Geolocation data allowed Lowe’s to discover ideal real estate site selection to drive foot traffic and cannibalize Home Depot locations. Geospatial data can support corporate strategies through gaining an empirical understanding of a corporates trade area. The demographic insights can also be valuable information for marketing teams to leverage.
Last Part: Further Case Studies – Summary
Quantum technology
Before new disruptions caused by quantum technology
Solving optimization problems, simulating complex structures, artificial intelligence and cryptography
(Roland Berger) Quantum computers are likely to gain increasing influence on companies and their business models in the coming decade. This is shown by a study for which Roland Berger surveyed 110 managers from various industries across Europe: The majority expect significant changes from ultra-fast computing technology, either because it further accelerates digital change (42%) or because it triggers new disruptions (23%). 63 percent want to take the topic into greater consideration in strategic considerations in the future.
So far, quantum technology has played a subordinate role in the agenda of business leaders: only 8 percent of those surveyed are already specifically incorporating the expected change into their planning. Another 35 percent say they are at least keeping an eye on developments. “The reluctance is understandable, because the commercial benefits of quantum computers will take some time to come,” says Martin Streichfuß, partner of Roland Berger and one of the study authors. “Nevertheless, we believe it makes sense to look at the new technology and its potential applications at an early stage. The fact that the majority of our survey participants also plan to do so is a good signal.”
__________________________________________________________________________________
Quantum computing as an important topic for the future
__________________________________________________________________________________
Quantum computers could be used in a variety of ways. From the Roland Berger experts’ point of view, four areas of application for ultra-fast computers are particularly commercially relevant: firstly, solving optimization problems for which classic computers often take a very long time, secondly, simulations of complex structures, thirdly, artificial intelligence and machine learning and fourthly, the topic of cryptography. “There will definitely be more use cases in the future,” says Michael Alexander, partner and one of the heads of Roland Berger’s Advanced Technology Center. “As with any new technology, fields of application for quantum computers that are currently unimaginable will open up once they are tested by a larger group of users.”
__________________________________________________________________________________
The automotive, pharmaceutical and chemical sectors are likely to benefit .
__________________________________________________________________________________
Which industries are most likely to benefit from the new computing options depends primarily on the data intensity of the respective business model: “The decisive factors are the amount of data, its diversity and the required processing speed,” says Frederik Hammermeister, partner at Roland Berger. “We expect that the business models and value chains of the manufacturing sector in particular will benefit from quantum computing, such as the automotive, pharmaceutical and chemical industries, but also the finance and transport sectors.”
Quantum computers could help simulate the chemical processes inside batteries and thus enable the construction of more efficient batteries for electric cars. They could take the training of software, artificial intelligence and machine learning processes to a new level, improving applications such as autonomous driving or the automatic detection of diseases such as cancer. Furthermore, they could help the financial industry solve complex optimization problems that are currently causing problems for even the largest supercomputers.
The work with chemicals and active ingredients could also be revolutionized: “What today often requires years of laboratory work, for example the determination of molecular structures or the analysis of their properties, could take place in a fraction of the time with quantum simulations,” says study author Alexander. “This would greatly speed up and simplify the development of new drugs or materials.”
__________________________________________________________________________________
Massive public and private investments
__________________________________________________________________________________
Nobody can reliably predict when the technological breakthrough in the field of quantum computing will occur. Nevertheless, an interesting market environment for the technology is currently developing. This is due, among other things, to very high levels of public funding: countries around the world spend a total of 22 billion euros on research programs on quantum technologies, the most being China (10 billion), Germany (3.1 billion) and France (1.6 billion). Because quantum technology is considered critical infrastructure. This also inspires private investors: The number of venture capital transactions in the area of quantum applications has also reached an all-time high – an indication that more and more investors are expecting commercial use in the not too distant future.
The full study at: https://ots.de/tVMapS
Connect & Collect
“Work Research Cloud”
(ZBW) The ZBW – Leibniz Information Center for Economics ( www.zbw.eu ) is creating the conditions for sustainable labor research with partners from science and business in the newly launched CoCo project. In collaboration with regional competence centers, it should be possible to use AI technologies profitably in the world of work. The focus of the BMBF-funded project is a “work research cloud”. URL: www.coco-projekt.de
Digitalization and artificial intelligence are changing the world of work. They create opportunities for future work that have not yet been explored. The BMBF project “Connect & Collect” (CoCo) is now connecting the interdisciplinary actors in labor research as well as central infrastructure service providers for economic research in order to research precisely these opportunities across disciplines.
The central component of the project is an AI-supported “work research cloud”. This interdisciplinary data and knowledge storage supports networked research and innovation work between science and business and provides structures for sustainable knowledge transfer. In this way, the speed of research and ever shorter development cycles in the economy can be harmonized.
In this project, the ZBW – Leibniz Information Center for Economics ( www.zbw.eu ) will design, develop and implement the socio-technical infrastructure of this labor research cloud. In addition, the ZBW ensures the reach, sustainability and perpetuation of this data and knowledge storage. In order to make the cloud a lively and attractive place for networking and cooperation between science and business, the ZBW is developing incentive systems and concrete measures that are tested and optimized in regular cycles.
Together with the competence centers, the ZBW, led by Prof. Dr. Klaus Daughtermann and Prof. Dr. Isabella Peters requirements and success factors and develops a target image for the efficient research and innovation work of the future. This forms the basis for the conception of the technical infrastructure of the labor research cloud as an open web platform through which all actors can access research results and participate in the further development of the state of knowledge.
Outside the box (49):
Against the educational muddle in schools
Instead of empty federal search engines,
create virtual makerspaces with teachers
Find what
has long been available at the state and local level
Joachim Paul, School digital: Educational media for schools – nationwide mess, in: heise online, April 30th. The digital education infrastructure set up by the federal, state and local governments has so far not been able to bridge the loss of face-to-face teaching. “The fact that there was no complete catastrophe is once again thanks to the efforts of teachers.” The fact that they are called “digital grouches” only means that they have to pay for what politicians have neglected to do in recent years. On the other hand, the deficits of the state politicians, the Conference of Ministers of Education and Federal Politics are evident “not only in the late or inadequate expansion and operation of learning platforms, as various collapses and overload phenomena on several platforms demonstrate, they also lie in the procurement and provision of quality-tested digital educational content for the context of school.” At this point Paul argues the thesis: “Everything is already there and has been for years. The money should only be spent in the right place. The existing offers should be promoted, expanded and linked accordingly.” For this to happen, there is no need to change federalism, but rather to solve the existing communication problems between the authorities.
The author points out that there are “quite powerful platforms for the distribution of digital educational media specifically for schools” in all 16 federal states, and in some cases for more than 15 years. However, the municipalities are responsible for procuring material resources for schools, and they only buy digital content for their own area. “The usage licenses for the media are therefore limited to your own municipality or your own federal state, so the media are by no means nationwide or even freely available.” Paul regrets: “These platforms are also not in general awareness and often unfortunately not in the focus of federal politics . They rarely if ever appear in national reporting on education and digitalization. The media’s attention here regularly goes to the use of YouTube in schools or the media libraries of the major broadcasters. At the same time, the nationwide political activities – supported by the Conference of Ministers of Education of the States (KMK) and the Federal Ministry of Education and Research (BMBF) – concentrate in an uncoordinated manner on the provision of two search engines for so-called free educational materials (“WirLernen Online” and MUNDO).
What do these search engines do? After their review, Paul comes to this conclusion: “Both systems are essentially collections of Internet links. The extensive and convenient filter functions of both systems cannot conceal the poor content of the offering; rather, they remain a technical expression of political helplessness and lack of concept. Where there isn’t much, not much can be found.”
And what consequences for politics? “A school digital pact that really deserves the name must make the provision of digital educational content for schools and the promotion of its production an integral part of a joint strategy between the federal, state and local governments. These should not only address the equipment problem and limit themselves, as the BMBF and KMK do, to promoting two search engines for OER materials and hoping that the OER content will emerge virtually by itself.”
This could mean, for example: “Companies are asked – those who hold all rights to the internal image, audio, text and video components of their media – what it would cost to place one of their media under a CC license and thereby make it a nationwide, freely available OER educational medium. … On the basis of such sources and with these sources as media modeling clay, teachers could then – of course in compliance with the citation rules and the recognition of third-party authorship (CC-BY-SA) – produce further OER media and learning projects on the respective topic and make them freely available place. And by the way, this would also stimulate cross-border cooperation between teachers, a nationwide open scene, a virtual makerspace for educational media. The OER hit lists in MUNDO and WLO would then look completely different in a few years.”
Or so? “And where can you actually find these state portals that have been mentioned several times, these state infrastructures that are already filling their warehouses? Two demands of the KMK are: “1. General findability of educational media” and “4. public documentation. A solution to this would be, for example, to create a simple list of links to a centrally hosted website with good visibility. … So the KMK would have to meet its own demands and set something up here.”
OpenPassword
Forum and news
for the information industry
in German-speaking countries
New editions of Open Password appear four times a week.
If you would like to subscribe to the email service free of charge, please register at www.password-online.de.
The current edition of Open Password can be accessed immediately after it appears on the web. www.password-online.de/archiv. This also applies to all previously published editions.
International Cooperation Partner:
Outsell (London)
Business Industry Information Association/BIIA (Hong Kong)
__________________________________________________________________________________
Case Study 5: Utilizing Receipt Data To Understand The Promotion Effectiveness Of Coupons
__________________________________________________________________________________
Purpose: Customer Insights, Market/Competitive Intelligence
Relevant Department: Marketing, Sales
Marketing has been transformed by data as the new digital reality makes everything measurable. One such example of this is utilizing customer receipt data to understand promotion effectiveness.
The client ran a coupon promotion that saw the brand respond after experiencing low indices for over a year. The marketing team wanted to analyze whether the promotion was actually attracting new buyers or if returning buyers were just capitalizing on the lower price.
Numerator’s receipt data was used to analyze which users had used a coupon to purchase the product. The team then analyzed the shopper groups and performed a promotion analysis, seeing which were new to the brand.
The results of the analysis led the brand team to continue to replicate similar promotions for the brand. They demonstrated to retailers that coupon inserts were incremental and successful at driving new users to the brand.
Key takeaway. Promotion effectiveness analysis showed that coupon inserts were incremental and drove new users to the brand. Those insights led the brand team to duplicate similar promotions across multiple retailers. Marketers who build a technically savvy organization can analyze and optimize the performance of their marketing campaigns .
Last Part: Further Case Studies – Conclusions
Open Password Archive – Publications
OPEN PASSWORD ARCHIVE
DATA JOURNALISM
Handelsblatt’s Digital Reach