Recherche avancée

Sur d’autres sites (6880)

• A Guide to Bank Customer Segmentation

  18 juillet 2024, par Erin

  Banking customers are more diverse, complex, and demanding than ever. As a result, banks have to work harder to win their loyalty, with 75% saying they would switch to a bank that better fits their needs.

  The problem is banking customers’ demands are increasingly varied amid economic uncertainties, increased competition, and generational shifts.

  If banks want to retain their customers, they can’t treat them all the same. They need a bank customer segmentation strategy that allows them to reach specific customer groups and cater to their unique demands.

  What is customer segmentation ?

  Customer segmentation divides a customer base into distinct groups based on shared characteristics or behaviours.

  This allows companies to analyse the behaviours and needs of different customer groups. Banks can use these insights to target segments with relevant marketing throughout the customer cycle, e.g., new customers, inactive customers, loyal customers, etc.

  You combine data points from multiple segmentation categories to create a customer segment. The most common customer segmentation categories include :

  • Demographic segmentation
  • Website activity segmentation
  • Geographic segmentation
  • Purchase history segmentation
  • Product-based segmentation
  • Customer lifecycle segmentation
  • Technographic segmentation
  • Channel preference segmentation
  • Value-based segmentation

  

  By combining segmentation categories, you can create detailed customer segments. For example, high-value customers based in a particular market, using a specific product, and approaching the end of the lifecycle. This segment is ideal for customer retention campaigns, localised for their market and personalised to satisfy their needs.

  

  Matomo’s privacy-centric web analytics solution helps you capture data from the first visit. Unlike Google Analytics, Matomo doesn’t use data sampling (more on this later) or AI to fill in data gaps. You get 100% accurate data for reliable insights and customer segmentation.

  Try Matomo for Free

  Get the web insights you need, without compromising data accuracy.

  Start for free

  No credit card required

  Why is customer segmentation important for banks ?

  Customer segmentation allows you to address the needs of specific groups instead of treating all of your customers the same. This has never been more important amid a surge in bank switching, with three in four customers ready to switch to a provider that better suits their needs.

  Younger customers are the most likely to switch, with 19% of 18-24 year olds changing their primary bank in the past year (PDF).

  Customer expectations are changing, driven by economic uncertainties, declining trust in traditional banking, and the rise of fintech. Even as economic pressures lift, banks need to catch up with the demands of maturing millennials, Gen Z, and future generations of banking customers.

  Switching is the new normal, especially for tech-savvy customers encouraged by an expanding world of digital banking options.

  To retain customers, banks need to know them better and understand how their needs change over time. Customer retention provides the insights banks need to understand these needs at a granular level and the means to target specific customer groups with relevant messages.

  At its core, customer segmentation is essential to banks for two key reasons :

  • Customer retention : Holding on to customers for longer by satisfying their personal needs.
  • Customer lifetime value : Maximising ongoing customer revenue through retention, purchase frequency, cross-selling, and upselling.

  Here are some actionable bank customer segmentation strategies that can achieve these two objectives :

  Prevent switching with segment analysis

  Use customer segmentation to prevent them from switching to rivals by knowing what they want from you. Analyse customer needs and how they change throughout the lifecycle. Third-party data reveals general trends, but what do your customers want ?

  

  Use first-party customer data and segmentation to go beyond industry trends. Know exactly what your customers want from you and how to deliver targeted messages to each segment — e.g., first-time homebuyers vs. retirement planners.

  Keep customers active with segment targeting

  Target customer segments to keep customers engaged and motivated. Create ultra-relevant marketing messages and deliver them with precision to distinct customer segments. Nurture customer motivation by continuing to address their problems and aspirations.

  Improve the quality of services and products

  Knowing your customers’ needs in greater detail allows you to adapt your products and messages to cater to the most important segments. Customers switch banks because they feel their needs are better met elsewhere. Prevent this by implementing customer segmentation insights into product development and marketing.

  Personalise customer experiences by layering segments

  Layer segments to create ultra-specific target customer groups for personalised services and marketing campaigns. For example, top-spending customers are one of your most important segments, but there’s only so much you can do with this. However, you can divide this group into even narrower target audiences by layering multiple segments.

  For example, segmenting top-spending customers by product type can create more relevant messaging. You can also segment recent activity and pinpoint specific usage segments, such as those with a recent drop in transactions.

  Now, you have a three-layered segment of high-spending customers who use specific products less often and whom you can target with re-engagement campaigns.

  Maximise customer lifetime value

  Bringing all of this together, customer segmentation helps you maximise customer lifetime value in several ways :

  • Prevent switching
  • Enhance engagement and motivation
  • Re-engage customers
  • Cross-selling, upselling
  • Personalised customer loyalty incentives

  The longer you retain customers, the more you can learn about them, and the more effective your lifetime value campaigns will be.

  Balancing bank customer segmentation with privacy and marketing regulations

  Of course, customer segmentation uses a lot of data, which raises important legal and ethical questions. First, you need to comply with data and privacy regulations, such as GDPR and CCPA. Second, you also have to consider the privacy expectations of your customers, who are increasingly aware of privacy issues and rising security threats targeting financial service providers.

  If you aim to retain and maximise customer value, respecting their privacy and protecting their data are non-negotiables.

  Regulators are clamping down on finance

  Regulatory scrutiny towards the finance industry is intensifying, largely driven by the rise of fintech and the growing threat of cyber attacks. Not only was 2023 a record-breaking year for finance security breaches but several compromises of major US providers “exposed shortcomings in the current supervisory framework and have put considerable public pressure on banking authorities to reevaluate their supervisory and examination programs” (Deloitte).

  Banks face some of the strictest consumer protections and marketing regulations, but the digital age creates new threats.

  In 2022, the Consumer Financial Protection Bureau (CFPB) warned that digital marketers must comply with finance consumer protections when targeting audiences. CFPB Director Rohit Chopra said : “When Big Tech firms use sophisticated behavioural targeting techniques to market financial products, they must adhere to federal consumer financial protection laws.”

  This couldn’t be more relevant to customer segmentation and the tools banks use to conduct it.

  Customer data in the hands of agencies and big tech

  Banks should pay attention to the words of CFPB Director Rohit Chopra when partnering with marketing agencies and choosing analytics tools. Digital marketing agencies are rarely experts in financial regulations, and tech giants like Google don’t have the best track record for adhering to them.

  Google is constantly in the EU courts over its data use. In 2022, the EU ruled that the previous version of Google Analytics violated EU privacy regulations. Google Analytics 4 was promptly released but didn’t resolve all the issues.

  Meanwhile, any company that inadvertently misuses Google Analytics is legally responsible for its compliance with data regulations.

  Banks need a privacy-centric alternative to Google Analytics

  Google’s track record with data regulation compliance is a big issue, but it’s not the only one. Google Analytics uses data sampling, which Google defines as the “practice of analysing a subset of data to uncover meaningful information from a larger data set.”

  This means Google Analytics places thresholds on how much of your data it analyses — anything after that is calculated assumptions. We’ve explained why this is such a problem before, and GA4 relies on data sampling even more than the previous version.

  In short, banks should question whether they can trust Google with their customer data and whether they can trust Google Analytics to provide accurate data in the first place. And they do. 80% of financial marketers say they’re concerned about ad tech bias from major providers like Google and Meta.

  

  Matomo is the privacy-centric alternative to Google Analytics, giving you 100% data ownership and compliant web analytics. With no data sampling, Matomo provides 20-40% more data to help you make accurate, informed decisions. Get the data you need for customer segmentation without putting their data at risk.

  Try Matomo for Free

  Get the web insights you need, without compromising data accuracy.

  Start for free

  No credit card required

  Bank customer segmentation examples

  Now, let’s look at some customer segments you create and layer to target specific customer groups.

  Visit-based segmentation

  Visit segmentation filters audiences based on the pages they visit on your website and the behaviors they exhibit—for example, first-time visitors vs. returning visitors or landing page visitors vs. blog page visitors.

  If you look at HSBC’s website, you’ll see it is structured into several categories for key customer personas. One of its segments is international customers living in the US, so it has pages and resources expats, people working in the US, people studying in the US, etc. 

  

  By combining visit-based segmentation with ultra-relevant pages for specific target audiences, HSBC can track each group’s demand and interest and analyse their behaviours. It can determine which audiences are returning, which products they want, and which messages convert them.

  Demographic segmentation

  Demographic segmentation divides customers by attributes such as age, gender, and location. However, you can also combine these insights with other non-personal data to better understand specific audiences.

  For example, in Matomo, you can segment audiences based on the language of their browser, the country they’re visiting from, and other characteristics. So, in this case, HSBC could differentiate between visitors already residing in the US and those outside of the country looking for information on moving there.

  

  It could determine which countries they’re visiting, which languages to localise for, and which networks to run ultra-relevant social campaigns on.

  Interaction-based segmentation

  Interaction-based segmentation uses events and goals to segment users based on their actions on your website. For example, you can segment audiences who visit specific URLs, such as a loan application page, or those who don’t complete an action, such as failing to complete a form.

  

  With events and goals set up, you can track the actions visitors complete before making purchases. You can monitor topical interests, page visits, content interactions, and pathways toward conversions, which feed into their customer journey.

  From here, you can segment customers based on their path leading up to their first purchase, follow-up purchases, and other actions.

  Purchase-based segmentation

  Purchase-based segmentation allows you to analyse the customer behaviours related to their purchase history and spending habits. For example, you can track the journey of repeat customers or identify first-time buyers showing interest in other products/services.

  You can implement these insights into your cross-selling and upselling campaigns with relevant messages designed to increase retention and customer lifetime value.

  Get reliable website analytics for your bank customer segmentation needs

  With customers switching in greater numbers, banks need to prioritise customer retention and lifetime value. Customer segmentation allows you to target specific customer groups and address their unique needs — the perfect strategy to stop them from moving to another provider.

  Quality, accurate data is the key ingredient of an effective bank customer segmentation strategy. Don’t accept data sampling from Google Analytics or any other tool that limits the amount of your own data you can access. Choose a web analytics tool like Matamo that unlocks the full potential of your website analytics to get the most out of bank customer segmentation.

  Matomo is trusted by over 1 million websites globally, including many banks, for its accuracy, compliance, and reliability. Discover why financial institutions rely on Matomo to meet their web analytics needs.

  Start collecting the insights you need for granular, layered segmentation — without putting your bank customer data at risk. Request a demo of Matomo now.

  Try Matomo for Free

  21 day free trial. No credit card required.

  
  

  Your email address
  

  Your website address
  

  Your Analytics subdomain will be .matomo.cloud (change)

  Choose your analytics subdomain
  

  .matomo.cloud

  I accept the terms and conditions and data processing agreement (DPA)

  
  Your information will be used to create an account on our cloud service. For more information please consult our privacy policy.
  
  
  

  Start improving your websites now

  

• OCPA, FDBR and TDPSA – What you need to know about the US’s new privacy laws

  22 juillet 2024, par Daniel Crough

  On July 1, 2024, new privacy laws took effect in Florida, Oregon, and Texas. People in these states now have more control over their personal data, signaling a shift in privacy policy in the United States. Here’s what you need to know about these laws and how privacy-focused analytics can help your business stay compliant.

  Consumer rights are front and centre across all three laws

  The Florida Digital Bill of Rights (FDBR), Oregon Consumer Privacy Act (OCPA), and Texas Data Privacy and Security Act (TDPSA) grant consumers similar rights.

  Access : Consumers can access their personal data held by businesses.

  Correction : Consumers can correct inaccurate data.

  Deletion : Consumers may request data deletion.

  Opt-Out : Consumers can opt-out of the sale of their personal data and targeted advertising.

  Oregon Consumer Privacy Act (OCPA)

  The Oregon Consumer Privacy Act (OCPA), signed into law on June 23, 2023, and effective as of July 1, 2024, grants Oregonians new rights regarding their personal data and imposes obligations on businesses. Starting July 1, 2025, authorities will enforce provisions that require data protection assessments, and businesses must recognize universal opt-out mechanisms by January 1, 2026. In Oregon, the OCPA applies to business that :

  • Either conduct business in Oregon or offer products and services to Oregon residents

  • Control or process the personal data of 100,000 consumers or more, or

  • Control or process the data of 25,000 or more consumers while receiving over 25% of their gross revenues from selling personal data.

  Exemptions include public bodies like state and local governments, financial institutions, and insurers that operate under specific financial regulations. The law also excludes protected health information covered by HIPAA and other specific federal regulations.

  Business obligations

  Data Protection Assessments : Businesses must conduct data protection assessments for high-risk processing activities, such as those involving sensitive data or targeting children.

  Consent for Sensitive Data : Businesses must secure explicit consent before collecting, processing, or selling sensitive personal data, such as racial or ethnic origin, religious beliefs, health information, biometric data, and geolocation.

  Universal Opt-out : Starting January 1, 2025, businesses must acknowledge universal opt-out mechanisms, like the Global Privacy Control, that allow consumers to opt out of data collection and processing activities.

  Enforcement

  The Oregon Attorney General can issue fines up to $7,500 per violation. There is no private right of action.

  Unique characteristics of the OCPA

  The OCPA differs from other state privacy laws by requiring affirmative opt-in consent for processing sensitive and children’s data, and by including nonprofit organisations under its scope. It also requires global browser opt-out mechanisms starting in 2026.

  Florida Digital Bill of Rights (FDBR)

  The Florida Digital Bill of Rights (FDBR) became law on June 6, 2023, and it came into effect on July 1, 2024. This law targets businesses with substantial operations or revenues tied to digital activities and seeks to protect the personal data of Florida residents by granting them greater control over their information and imposing stricter obligations on businesses. It applies to entities that :

  • Conduct business in Florida or provide products or services targeting Florida residents,

  • Have annual global gross revenues exceeding $1 billion,

  • Receive 50% or more of their revenues from digital advertising or operate significant digital platforms such as app stores or smart speakers with virtual assistants.

  Exemptions include governmental entities, nonprofits, financial institutions covered by the Gramm-Leach-Bliley Act, and entities covered by HIPAA.

  Business obligations

  Data Security Measures : Companies are required to implement reasonable data security measures to protect personal data from unauthorised access and breaches.

  Handling Sensitive Data : Explicit consent is required for processing sensitive data, which includes information like racial or ethnic origin, religious beliefs, and biometric data.

  Non-Discrimination : Entities must ensure they do not discriminate against consumers who exercise their privacy rights.

  Data Minimisation : Businesses must collect only necessary data.

  Vendor Management : Businesses must ensure that their processors and vendors also comply with the FDBR, regarding the secure handling and processing of personal data.

  Enforcement

  The Florida Attorney General can impose fines of up to $50,000 per violation, with higher penalties for intentional breaches.

  Unique characteristics of the FDBR

  Unlike broader privacy laws such as the California Consumer Privacy Act (CCPA), which apply to a wider range of businesses based on lower revenue thresholds and the volume of data processed, the FDBR distinguishes itself by targeting large-scale businesses with substantial revenues from digital advertising. The FDBR also emphasises specific consumer rights related to modern digital interactions, reflecting the evolving landscape of online privacy concerns.

  Texas Data Privacy and Security Act (TDPSA)

  The Texas Data Privacy and Security Act (TDPSA), signed into law on June 16, 2023, and effective as of July 1, 2024, enhances data protection for Texas residents. The TDPSA applies to entities that :

  • Conduct business in Texas or offer products or services to Texas residents.

  • Engage in processing or selling personal data.

  • Do not fall under the classification of small businesses according to the U.S. Small Business Administration’s criteria, which usually involve employee numbers or average annual receipts. 

  The law excludes state agencies, political subdivisions, financial institutions compliant with the Gramm-Leach-Bliley Act, and entities compliant with HIPAA.

  Business obligations

  Data Protection Assessments : Businesses must conduct data protection assessments for processing activities that pose a heightened risk of harm to consumers, such as processing for targeted advertising, selling personal data, or profiling.

  Consent for Sensitive Data : Businesses must get explicit consent before collecting, processing, or selling sensitive personal data, such as racial or ethnic origin, religious beliefs, health information, biometric data, and geolocation.

  Companies must have adequate data security practices based on the personal information they handle.

  Data Subject Access Requests (DSARs) : Businesses must respond to consumer requests regarding their personal data (e.g., access, correction, deletion) without undue delay, but no later than 45 days after receipt of the request.

  Sale of Data : If businesses sell personal data, they must disclose these practices to consumers and provide them with an option to opt out.

  Universal Opt-Out Compliance : Starting January 1, 2025, businesses must recognise universal opt-out mechanisms like the Global Privacy Control, enabling consumers to opt out of data collection and processing activities.

  Enforcement

  The Texas Attorney General can impose fines up to $25,000 per violation. There is no private right of action.

  Unique characteristics of the TDPSA

  The TDPSA stands out for its small business carve-out, lack of specific thresholds based on revenue or data volume, and requirements for recognising universal opt-out mechanisms starting in 2025. It also mandates consent for processing sensitive data and includes specific measures for data protection assessments and privacy notices.

  Try Matomo for Free

  Get the web insights you need, without compromising data accuracy.

  Start for free

  No credit card required

  Privacy notices across Florida, Oregon, and Texas

  All three laws include a mandate for privacy notices, though there are subtle variations in their specific requirements. Here’s a breakdown of these differences :

  FDBR privacy notice requirements

  Clarity : Privacy notices must clearly explain the collection and use of personal data.

  Disclosure : Notices must inform consumers about their rights, including the right to access, correct, delete their data, and opt-out of data sales and targeted advertising.

  Specificity : Businesses must disclose if they sell personal data or use it for targeted advertising.

  Security Practices : The notice should describe the data security measures in place.

  OCPA privacy notice requirements

  Comprehensive Information : Notices must provide information about the personal data collected, the purposes for processing, and any third parties that can access it.

  Consumer Rights : Must plainly outline consumers’ rights to access, correct, delete their data, and opt-out of data sales, targeted advertising, and profiling.

  Sensitive Data : To process sensitive data, businesses or entities must get explicit consent and communicate it.

  Universal Opt-Out : Starting January 1, 2026, businesses must recognise and honour universal opt-out mechanisms.

  TDPSA privacy notice requirements

  Detailed Notices : Must provide clear and detailed information about data collection practices, including the data collected and the purposes for its use.

  Consumer Rights : Must inform consumers of their rights to access, correct, delete their data, and opt-out of data sales and targeted advertising.

  High-Risk Processing : Notices should include information about any high-risk processing activities and the safeguards in place.

  Sensitive Data : To process sensitive data, entities and businesses must get explicit consent.

  What these laws mean for your businesses

  Businesses operating in Florida, Oregon, and Texas must now comply with these new data privacy laws. Here’s what you can do to avoid fines :

  1. Understand the Laws : Familiarise yourself with the specific requirements of the FDBR, OCPA, and TDPSA, including consumer rights and business obligations.

  2. Implement Data Protection Measures : Ensure you have robust data security measures in place. This includes conducting regular data protection assessments, especially for high-risk processing activities.

  3. Update Privacy Policies : Provide clear and comprehensive privacy notices that inform consumers about their rights and how their data is processed.

  4. Obtain Explicit Consent : For sensitive data, make sure you get explicit consent from consumers. This includes information like health, race, sexual orientation, and more.

  5. Manage Requests Efficiently : Be prepared to handle requests from consumers to access, correct, delete their data, and opt-out of data sales and targeted advertising within the stipulated timeframes.

  6. Recognise Opt-Out Mechanisms : For Oregon, businesses must be ready to implement and recognise universal opt-out mechanisms by January 1, 2026. In Texas, opt-out enforcement begins in 2026. In Florida, the specific opt-out provisions began on July 1, 2024.

  7. Stay Updated : Keep abreast of any changes or updates to these laws to ensure ongoing compliance. Keep an eye on the Matomo blog or sign up for our newsletter to stay in the know.

  Are we headed towards a more privacy-focused future in the United States ?

  Florida, Oregon, and Texas are joining states like California, Virginia, Colorado, Connecticut, Utah, Iowa, Indiana, Tennessee, and Montana in strengthening consumer privacy protections. This trend could signify a shift in US policy towards a more privacy-focused internet, underlining the importance of consumer data rights and transparent business practices. Even if these laws do not apply to your business, considering updates to your data and privacy policies is wise. Fortunately, there are tools and solutions designed for privacy and compliance to help you navigate these changes.

  Avoid fines and get better data with Matomo

  Most analytics tools don’t prioritize safeguarding user data. At Matomo, we believe everyone has the right to data sovereignty, privacy and amazing analytics. Matomo offers a solution that meets privacy regulations while delivering incredible insights. With Matomo, you get :

  100% Data Ownership : Keep full control over your data, ensuring it is used according to your privacy policies.

  Privacy Protection : Built with privacy in mind, Matomo helps businesses comply with privacy laws.

  Powerful Features : Gain insights with tools like heatmaps, session recordings, and A/B testing.

  Open Source : Matomo’s is open-source and committed to transparency and customisation.

  Flexibility : Choose to host Matomo on your servers or in the cloud for added security.

  No Data Sampling : Ensure accurate and complete insights without data sampling.

  Privacy Compliance : Easily meet GDPR and other requirements, with data stored securely and never sold or shared.

  Try Matomo for Free

  21 day free trial. No credit card required.

  
  

  Your email address
  

  Your website address
  

  Your Analytics subdomain will be .matomo.cloud (change)

  Choose your analytics subdomain
  

  .matomo.cloud

  I accept the terms and conditions and data processing agreement (DPA)

  
  Your information will be used to create an account on our cloud service. For more information please consult our privacy policy.
  
  
  

  Start improving your websites now

  

  Disclaimer : This content is provided for informational purposes only and is not intended as legal advice. While we strive to ensure the accuracy and timeliness of the information provided, the laws and regulations surrounding privacy are complex and subject to change. We recommend consulting with a qualified legal professional to address specific legal issues related to your circumstances. 

• Could not read frame error when trying to decompress mp4 file with ffmpeg and Python's threading module

  23 janvier 2017, par mdornfe1

  I’m training constitutional neural networks with video data. So far the bottle neck of my application is decompressing the mp4 files before passing the images to the CNN for training. I had the idea to try to have multiple cpu threads decompress the images concurrently and having one thread pass images to the CNN for training. I made a class VideoStream which makes connection to the mp4 file using the ImageIO module which is built on top of ffmpeg. The structure of my program is a follows :

  1) Generate random ints which represent the frame numbers of the mp4 file that will be used in training. Store these ints in list frame_idxs.

  2) Pass this list of ints and an empty list called frame_queue to the worker function decompress_video_data.

  3) Each worker function makes a connection to the mp4 file using VideoStream.

  4) Each worker function then pops of elements of frame_idxs, decompresses that frame, and then stores that frame as numpy array in list frame_queue.

  Here is the code

  import numpy as np
  
  import os, threading, multiprocessing
  
  
  
  
  
  def decompress_video_data(frame_queue, frame_idxs, full_path):
  
      vs = VideoStream(full_path)
  
      while len(frame_idxs) >1 0:
  
          i = frame_idxs.pop()
  
          frame = vs[i]
  
          frame_queue.append(frame)
  
  
  
  video_folder = '/mnt/data_drive/frame_viewer_client'
  
  video_files = os.listdir(video_folder)
  
  video_file = video_files[0]
  
  full_path = os.path.join(video_folder, video_file)
  
  vs = VideoStream(full_path)
  
  
  
  num_samples = 1000
  
  batch_size = 1
  
  frame_queue = []
  
  decompress_threads = []
  
  frame_idxs = list(np.random.randint(0, len(vs), 
  
      size = batch_size * num_samples))
  
  num_cores = multiprocessing.cpu_count()
  
  
  
  for n in range(num_cores - 1):
  
      decompress_thread = threading.Thread(target=decompress_video_data, 
  
          args=(frame_queue, frame_idxs, full_path))
  
      decompress_threads.append(decompress_thread)
  
      decompress_thread.start()

  The program will sucessfuly decompress approximately 200 frames, and then ImageIO will throw an RuntimeError : Could not read frame. The full error is here. Does anyone know why this is happening ? Is it possible to do what I’m trying to do ? Does ffmpeg just not work with multi threading ?