Every moment, behind the scenes, the products you use are getting better at anticipating your needs and desires. Your Netflix homepage updates in real time, your food delivery app predicts what youre craving, and your fitness app fine-tunes recommendations based on your recent activity.This is hyper-personalizationan advanced approach to personalization that leverages real-time data, artificial intelligence (AI), and behavioral analytics to deliver highly individualized experiences for everyuser.In this article, well explore the different levels of personalization, the data that fuels it, and how to design interfaces that deliver truly individualized experiences at scale. Hyper-personalization is more than a marketing strategyits a fundamental shift in how we design interfaces.The Evolution of PersonalizationAs digital experiences evolve, businesses move from broad segmentation to real-time individualization:Traditional Segmentation (Millions of Users)Users are grouped by static attributes like geography or device type, receiving identical experiences.Cohort-based (500K to 10K Users)Users are dynamically grouped based on behavioral data (e.g., purchase intent), refining personalization.Individual AdaptationOne-to-one personalization, where experiences adjust in real time based on behavior, intent, andcontext.For years, UX designers have relied on personas to design intuitive products. However, modern personalization goes beyond static personas by leveraging real-time behavioral data to dynamically group users into cohorts that evolve overtime.Levels of PersonalizationPersonalization evolves across different levels, increasing in complexity and user engagement while introducing ethical concerns.No PersonalizationA generic experience with no adaptation to user behavior or preferences.Segmented PersonalizationUsers are grouped into broad categories (e.g., demographics, device type), with predefined content and recommendations.Behavioral PersonalizationUser actions, such as browsing history or past interactions, shape recommendations.Contextual PersonalizationReal-time factors like location, time, or device influence content and interface adjustments.Predictive PersonalizationAI anticipates user needs by analyzing past behavior, trends, and inferredintent.Hyper-PersonalizationA 1:1 adaptive experience, where AI continuously refines content, UI, and recommendations in realtime.Emotional/Sentient PersonalizationTheoretical next step, where AI interprets emotions and intent, creating deeply human-like interactions.As personalization advances, so do ethical concernsfrom data privacy to algorithmic transparencyrequiring a balance between user experience and responsible design.Essential Elements of PersonalizationHyper-personalization is built on multiple interconnected components, continuously refining experiences based on user behavior and real-time data:Data Collection & IntegrationAggregates user interactions, preferences, and contextual signals.Segmentation & ProfilingGroups users dynamically based on behavioral and demographic patterns.Predictive AnalyticsUses AI to anticipate user needs before theyact.Real-Time Contextual AdaptationAdjusts experiences based on factors like location, time, andintent.AI & Machine Learning ModelsContinuously optimize recommendations and interactions.Omni-Channel IntegrationEnsures consistency across web, mobile, and physical touchpoints.Dynamic UI PersonalizationInterfaces adapt layout, content, and visuals per user preferences.Feedback MechanismsCaptures explicit (user input) and implicit (behavioral) feedback to refine personalization.Continuous Learning fuels ongoing improvements, ensuring each interaction becomes more relevant overtime.The Critical Role ofDataPersonalization is only as good as the data that fuels it. But not all data carries the same weightsome data types drive meaningful personalization, while others offer only surface-level insights.Behavioral DataTracks user interactions (clicks, searches, purchases) and is the most valuable because it reflects real user intent and adapts to evolving preferences.Preferences & Explicit FeedbackCaptures user-stated interests, likes, and dislikes, allowing direct personalization.Contextual DataUses real-time signals like location, time, and device to tailor experiences dynamically.Demographic InformationIncludes age, gender, and location, forming a foundational layer of personalization.Intent SignalsDetects subtle indicators (e.g., search behavior, abandoned carts) to infer userneeds.Affinity & Relationship DataLooks at social connections and past engagement withbrands.Psychographic DataAnalyzes lifestyle, values, and interests for deeper personalization.Social & Network DataExamines peer influence and shared interests.Event-Based DataAdapts experiences based on key moments (holidays, birthdays, lifeevents).How to Request Data Without LosingTrustIf data is the foundation of hyper-personalization, then how we collect and manage it is just as critical as how we useit. Request Permissions inContextAsk for permissions only when the user engages with a feature that requires it, ensuring relevance and timing.Example: Google Maps requests location access when a user searches for nearby restaurants, rather than at the applaunch. Explain the BenefitClearlyCommunicate why permission is needed and how it enhances the user experience. When users understand the value, they are more likely to opt in.Example: Enable step tracking to get personalized fitness goals based on your daily movement. Offer Alternatives WhenPossibleAlways provide an alternative when users may be hesitant to share certain data. This allows them to engage with the product on their terms, building trust over time.Example: A food delivery app allows users to manually enter their address instead of forcing them to enable GPS location tracking. Dont Request All Permissions atOnceBombarding users with multiple permission requests right at onboarding can feel invasive and reduce trust. Its better to introduce permissions gradually, tied to relevant interactions.Example: A newly installed messaging app immediately asks for access to location, contacts, microphone, and camera before the user even sends a messagewithout explaining why. This creates suspicion and increases opt-outrates.Support explicit preference settingTo deliver hyper-personalized experiences without making assumptions, its crucial to let users define their own preferences. Giving them control from the start builds trust and ensures recommendations align with their actual interests. Ask for Preferences During OnboardingEncourage users to select their interests or preferences when they first sign up. This helps tailor content and recommendations immediately, setting the foundation for a more relevant experience.Example: Spotify prompts new users to select favorite artists, shaping their initial playlist recommendations. Allow Preferences to Evolve OverTimeUsers needs and interests change, so personalization should be adaptable. Provide easy ways for users to update or refine their preferences over time.Example: Flipboard allows users to follow or unfollow topics, ensuring their news feed remains relevant. Use Clear, Understandable LanguageAvoid technical jargon or vague phrasing when asking users to set preferences. Ensure instructions are simple, direct, and highlight the value of customization.Example: Instead of saying, Enable preference-based algorithmic adjustments, use: Select topics you love to see more of what interests you. Dont Overload Users with Too ManyChoicesWhile preference selection is helpful, overwhelming users with too many options can lead to decision fatigue and frustration. Keep the process simple and intuitive.Example: A streaming app asking users to select 30+ categories of content before they can proceed creates friction, making it less likely theyll complete theprocess.Design Modular UI for Scalability and FlexibilityA modular UI approach enables dynamic, personalized experiences while maintaining consistency and scalability. By breaking down interfaces into adaptable components, designers can create layouts that adjust seamlessly to different user needs and contexts. Build Self-Contained, Reusable UIBlocksDesign independent UI components that can be used across multiple sections without requiring significant changes. This keeps the interface flexible while maintaining a unified experience.Example: Amazons homepage uses modular product cards that can be rearranged or swapped based on user preferences and promotions. Implement Dynamic ContentAreasRather than static layouts, design sections that change based on user behavior, preferences, and engagement patterns.Example: Netflix dynamically updates its homepage, showing different content categories, thumbnails, and placements depending on viewinghabits. Use Context-Aware UIElementsAdjust the UI based on user location, device, browsing history, or engagement to provide a more seamless and relevant experience.Example: E-commerce apps display region-specific deals and shipping information based on the users location. Avoid Over-Flexibility Leading to RandomnessWhile adaptability is key, excessive flexibility without structure can lead to a disjointed and confusing user experience. Maintain consistency in navigation and UI hierarchy.Example: An e-commerce site frequently rearranges product categories and filters based on past searches, causing users to lose track of where they originally found specificitems.Leverage Contextual PersonalizationContext plays a crucial role in delivering relevant, timely, and meaningful user experiences. Location-Based PersonalizationTailor experiences based on a users location to provide relevant offerings without feeling intrusive.Example: Starbucks suggests nearby stores and updates available menu items based on regional availability. Time, Routine & SeasonalityAdapt content based on the time of day, seasonal trends, or user routines to maintain relevance.Example: Spotify curates Morning Motivation playlists for early hours and Chill Evenings playlists later in theday. User Role, Journey & ProficiencyPersonalize interfaces based on user experience level or where they are in their journey with a product.Example: Duolingo adjusts difficulty based on user progress, gradually introducing advanced Dont Personalize Based on Sensitive or Private InformationAvoid Using personal health, financial, or lifestyle data in recommendations. ex. Assuming pregnancy, medical conditions, or relationship status based on purchases.Example: Facebook faced backlash for using relationship status to target ads about pregnancy and engagement rings, making some users uncomfortable.Levels of PersonalizationProvide Effective Feedback MechanismsEffective personalization doesnt end at delivering recommendationsit requires continuous learning from user interactions to refine future suggestions. Feedback mechanisms help algorithms assess whether personalized experiences are resonating withusers. Use Explicit & ImplicitFeedbackCombine direct user input (explicit) with passive behavioral signals (implicit) to evaluate personalization accuracy.Example: Instagram lets users hide posts they dont like (explicit), while also tracking time spent on content to adjust future recommendations (implicit). Provide Clear & Accessible FeedbackOptionsMake it easy for users to indicate whether recommendations were relevant or not.Example: YouTube Music's thumbs-up/thumbs-down system refines future content suggestions based on userratings. Show Users That Their FeedbackMattersReinforce that user interactions shape their personalized experience by quickly adapting recommendations.Example: On Instagram, when users report or hide a post, it immediately disappears from their feed, and the platform adjusts future recommendations to show less similarcontent. Dont Make Feedback Feel Like aChoreAvoid interrupting the experience with lengthy surveys or forcing users to take extra steps to refine personalization.Example: A shopping app that requires users to fill out a long form adds friction, making engagement feel like work rather than a seamless experience.Designing for Emotional ConnectionEffective personalization doesnt end at delivering recommendationsit requires continuous learning from user interactions to refine future suggestions. Feedback mechanisms help algorithms assess whether personalized experiences are resonating withusers. Use Emotionally Aware Microcopy & Tone ofVoiceCraft copy that acknowledges user emotions and provides warmth and empathy. A conversational, supportive tone enhances trust and engagement.Example: Duolingos owl encourages users with playful nudges like Youre on fire! Keep up the streak! to make learning feel more personal and rewarding. Implement Emotionally Intelligent FeedbackLoopsCreate systems that respond to user emotions and actions in real-time, making interactions feel reciprocal.Example: AI chatbot Replika adapts its tone based on user sentiment, offering supportive or cheerful responses depending on the context of the conversation. Celebrate Achievements & MilestonesRecognizing progress reinforces positive engagement and keeps users motivated. Small wins create a sense of accomplishment.Example: Fitness apps like Nike Training Club celebrate milestones, such as Youve completed 10 workouts this monthamazing dedication! to keep users motivated. Dont Use Emotion as ManipulationLeveraging emotions to pressure users into decisions erodes trust and creates negative experiences.Example: Duolingo faced criticism for its push notifications that made users feel guilty for missing lessons, with messages like Your streak is in danger! Dont disappoint Duo! While intended to encourage learning, such tactics can create stress rather than motivation, leading some users to disable notifications altogether.AI-driven emotional personalization apps like Replika can be both helpful and risky. They offer companionship, emotional support, and personalized interactions but also raise concerns about over-reliance, data privacy, and potential manipulation. Without ethical safeguards, these apps risk exploiting user emotions rather than supporting them.Why Hyper-Personalization MattersPersonalization is no longer a competitive advantageits an expectation. Research shows that 71% of U.S. consumers now anticipate personalized interactions, while 78% are more likely to recommend brands that deliver them. Companies leveraging personalization effectively see up to 40% more revenue from tailored marketing and product experiences.Beyond consumer expectations, hyper-personalization directly impacts key performance metrics:Higher EngagementUsers interact more with personalized recommendations, leading to increased sessiontimes.Improved ConversionsTargeted content and offers drive higher conversion rates.Stronger Retention & Reduced ChurnPersonalization fosters long-term loyalty.Revenue GrowthCompanies using data-driven personalization report increased revenue per user(ARPU).Despite its clear benefits, the precise ROI of hyper-personalization is difficult to quantify.Limited Transparency: While companies report success, they rarely disclose granular attribution data.Industry Insights: Firms like McKinsey, Accenture, and Forrester highlight major revenue lifts but often rely on broad case studies rather than rawnumbers.Survey Bias: Self-reported studies may overstate success due to sponsor influence or optimism in responses.While exact attribution is complex, one thing is clearbusinesses that invest in hyper-personalization consistently see gains in engagement, conversions, andrevenue.Hyper-personalization: a practical UX guide was originally published in UX Collective on Medium, where people are continuing the conversation by highlighting and responding to this story.

We may earn a commission from links on this page. Deal pricing and availability subject to change after time of publication.The JBL Flip 6 is currently $89.95 on Woot (and will remain so for the next four days or until stocks last)$40 off its usual $129.95 price tag, and if you're a Prime member, you also get free standard shipping (otherwise, it's $6). Its available in black, white, teal, gray, and green, giving you a decent range of options to match your vibe. The design is classic JBLcylindrical, durable, and built to be tossed into a bag without worry. Its passive radiators on both ends let you feel the bass in your hands (earning itself a place in PCMag's "Best Outdoor Speakers for 2025" roundup). Just dont expect a true subwoofer experience; its still a compact speaker, after all. JBL Flip 6 Portable Bluetooth Speaker $89.95 $129.95 Save $40.00 Get Deal Get Deal $89.95 $129.95 Save $40.00 Inside, the Flip 6 packs a 20-watt woofer and a 10-watt tweeter, covering a frequency range of 63Hz to 20kHz. It connects via Bluetooth 5.1 and supports AAC and SBC Bluetooth codecs, but not AptX, so audiophiles looking for higher-quality streaming may want to look elsewhere. The top panel houses volume and playback controls, though the multi-function button weirdly only lets you skip forwardnot back. JBLs PartyBoost feature is also here, letting you link multiple compatible JBL speakers for a bigger sound. Battery life is estimated at around 12 hours, but that depends on how loud you like your music. Crank it up, and youll burn through that charge much faster. Theres no speakerphone function or aux input, which might be a dealbreaker if you like having wired options.Durability is where the Flip 6 shineswith an IP67 rating, its fully dust-proof and waterproof, meaning it can survive a pool dunk and even float (though dont expect Bluetooth to work well underwater). It also comes with a 90-day Woot Limited Warranty for added peace of mind. JBLs companion app gives you access to customizable three-band EQ and PartyBoost controls, though the app itself is pretty barebones. Sound-wise, the Flip 6 delivers impressive low-end depth, and at moderate volumes, the bass is strong without distortion. Push it too high, though, and its digital signal processing (DSP) kicks in, thinning out the lows, according to this PCMag review. If you considering alternatives, the Sony SRS-XB23 might be worth checking out at around $95.Note: This item can't be delivered to Alaska, Hawaii, and PO Box addresses.

Panasonics mirrorless cameras are best known for their video powers, but with its latest model, the company is finally getting serious about photography as well. The $3,300 full-frame Lumix S1R II is not only the companys first camera capable of 8K video but can fire off 40 fps bursts of high-resolution (45 megapixel) RAW photos. It also offers a more advanced autofocus system, putting it squarely into the conversation with Sony, Canon and Nikon.Ive had a pre-production unit for several days now and so far, Im impressed. The original S1R was a chunky 2.24 pound monster, but the S1R II is much lighter at 1.75 pounds and considerably smaller in size. That makes it less burdensome than before though still a bit bigger than the Nikon Z8 and Canon R5 II. It also has one of the fattest grips Ive ever seen that gave me a firm, secure hold on the camera.Panasonic made other substantial changes to the body, removing the top display and shifting the mode dial from top left to top right. In the latters place a dedicated burst mode dial and photo/video/S&Q switch was introduced, while a new autofocus dial was placed at the back. Handling is now near the top among high-end full-frame cameras in my book, matching or even beating my favorite, the Sony A1 II.PanasonicThe rear 1.8-million-dot display was also overhauled and not only flips out for content creators but also tilts up and down for photographers putting it on par with the excellent screen on Sonys A1 II. The electronic viewfinder has the same 5.76-million-dot resolution and 120Hz refresh rate as before, but now offers a more natural 100 percent magnification.The S1R II supports both SD UHS II and fast CFexpress Type B cards (rather than XQD as before) with a slot for each, while also allowing SSD recording via the USB-C port like the S5 IIX and GH7. Other inputs include mic, headphone and a full-size HDMI slot, along with a 10Gbps USB-C port. The battery is the same as the one on the GH7 and G9 III but delivers just 350 shots max on a charge, unless you buy the optional DMW-BLK22 battery grip. The S1R II will also offer full-float 32-bit audio recording, but only with the optional DMW-XLR2 audio accessory.Steve Dent for EngadgetShooting performance is dramatically better with up to 40 fps (RAW 12-bit) burst speeds with continuous autofocus enabled, compared to a rather pitiful 6 fps on the previous model. It can also hit 10 fps speeds in mechanical mode or 9 fps with 14-bit RAW output. That matches the lower-resolution Canon R1 (though that model does it in 14-bit RAW mode) and is only below Sonys A9 III that can hit a mind-boggling 120 fps burst speeds in RAW mode.Panasonic also overhauled the phase-detect autofocus system to add more speed and AI smarts. It can now lock onto a subjects face and eyes quicker and follow their movements more smoothly, but also detect and automatically switch between humans, animals, cars, motorcycles, bikes, trains and airplanes. From my observations so far, its not quite up to the speed and fluidity of Sony and Canons latest models, but Panasonic is nearly there.PanasonicWhen it comes to image quality, JPEG photos look natural with realistic colors, though I wasn't able to open RAW files on this pre-production camera. One big improvement is at high ISOs in low light thanks to the dual ISO sensor that keeps noise under control all the way up to about ISO 12800.On the video side, the S1R II now supports internal ProRes RAW and 8K video capture, but not both at the same time. ProRes RAW captured to CFexpress or USB-C is limited to 5.8K but uses the full width of the sensor, so its a good solution for 4K productions. 8K, meanwhile, can only be captured in 4:2:0 LongGOP MP4 formats at 30 fps (4K tops out at 120 fps in Slow & Quick mode).That compares to 60 fps max on the Canon EOS R5 II and Nikon Z8, and both of those cameras can do that format in RAW. Sonys A1 II, by contrast, can also do 8K 30 fps video but doesnt support RAW recording at all. And of course, the S1R II supports V-Log recording and promises up to 14 stops of dynamic range, slightly less than the S1R. It'll also allow for external ProRes RAW recording at up to 8K to Atomos recorders via a firmware update coming after the initial launch.PanasonicPanasonics updated stabilization system promises up to 8 stops of shake reduction, but there are a few significant updates. It reduces edge distortion for video without cropping, though theres a small amount of vignetting if you do that. As before, it supports e-stabilization at regular and high strengths, and now offers the high-strength mode for anamorphic lenses.That raises the issue of rolling shutter, since the S1R II doesnt use a stacked sensor like its main competitors, the Nikon Z8 and Canon R5 II. Distortion is definitely more prominent than on those models, but readout speeds are relatively quick so its not offensive except when doing whip pans or filming very fast moving subjects.At $3,300, the S1R II is priced well below the $4,300 Canon R5 II but nearly on par with the Nikon Z8, which can currently be found at $3,400. Its set to start shipping at the end of March 2025.This article originally appeared on Engadget at https://www.engadget.com/cameras/panasonic-takes-on-canon-with-the-full-frame-45-megapixel-s1r-ii-mirrorless-camera-140048286.html?src=rss

A free version of Microsoft Office for Windows might be on its way. Microsoft has been silently testing an ad-supported version of Office reported Beebom. Free versions are currently only available on the web.The update would allow users to access PowerPoint, Word, Excel and more for free. There appears to be a few conditions for the fee-free option. It appears to include an ever-present banner on the righthand side and 15-second videos that play every few hours. Any documents must also be stored in One Drive rather than in local files.Each system also comes with limitations. For example, Word wouldn't have drawing and design tools or dictation. Excel foregoes conditional formatting, recommended charts and more. PowerPoint loses all draw, animation and record tools, among other features."Microsoft has been conducting some limited testing," a representative for the company told PC Mag in an email. "Currently, there are no plans to launch a free, ad-supported version of Microsoft Office desktop apps."This article originally appeared on Engadget at https://www.engadget.com/big-tech/microsoft-is-testing-a-free-with-ads-version-of-office-140030345.html?src=rss

Panasonic unveils the Lumix S1R II, its best full-frame mirrorless camera yet, with open gate 8K video, a new 44MP sensor, 32-bit float audio and more, plus a new Lumix Flow app.

Two ways to exploit a flaw in Parallels Desktop have recently been published.

Facing competitors like Cash App, Zelle, and Apple Pay, PayPal is focused on expanding Venmos role in digital commerce.

Nvidia reports fourth-quarter financial results on Wednesday after the bell.

Nearly 40% of the federal contracts that the Trump administration claims to have canceled as part of its signature cost-cutting program arent expected to save the government any money, the administrations own data shows.The Department of Government Efficiency run by Elon Musk last week published an initial list of 1,125 contracts that it terminated in recent weeks across the federal government. Data published on DOGEs Wall of Receipts shows that more than one-third of the contract cancellations, 417 in all, are expected to yield no savings.Thats usually because the total value of the contracts has already been fully obligated, which means the government has a legal requirement to spend the funds for the goods or services it purchased and in many cases has already done so.Its like confiscating used ammunition after its been shot when theres nothing left in it. It doesnt accomplish any policy objective, said Charles Tiefer, a retired University of Baltimore law professor and expert on government contracting law. Their terminating so many contracts pointlessly obviously doesnt accomplish anything for saving money.Dozens of them were for already-paid subscriptions to The Associated Press, Politico and other media services that the administration said it would discontinue. Others were for research studies that have been awarded, training that has taken place, software that has been purchased and interns that have come and gone.An administration official said it made sense to cancel contracts that are seen as potential dead weight, even if the moves do not yield any savings. The official was not authorized to discuss the matter publicly and spoke on condition of anonymity.In all, DOGE data says the 417 contracts in question had a total value of $478 million. Dozens of other canceled contracts are expected to yield little if any savings.Its too late for the government to change its mind on many of these contracts and walk away from its payment obligation, said Tiefer, who served on the Commission on Wartime Contracting in Iraq and Afghanistan.Tiefer said DOGE appeared to be taking a slash and burn approach to cutting contracts, which he said could damage the performance of government agencies. He said savings could be made instead by working with agency contracting officers and inspectors general to find efficiencies, an approach the administration has not taken.DOGE says the overall contract cancellations are expected to save more than $7 billion so far, an amount that has been questioned as inflated by independent experts.The canceled contracts were to purchase a wide range of goods and services.The Department of Housing and Urban Development awarded a contract in September to purchase and install office furniture at various branches. While the contract does not expire until later this year, federal records show the agency had already agreed to spend the maximum $567,809 with a furniture company.The U.S. Agency for International Development negotiated a $145,549 contract last year to clean the carpet at its headquarters in Washington. But the full amount had already been obligated to a firm that is owned by a Native American tribe based in Michigan.Another already-spent $249,600 contract went to a Washington, D.C., firm to help prepare the Department of Transportation for the recent transition from the Biden to the Trump administration.Some of the canceled contracts were intended to modernize and improve the way government works, which would seem to be at odds with DOGEs cost-cutting mission.One of the largest, for instance, went to a consulting firm to help carry out a reorganization at the Centers for Disease Control and Preventions National Center for Immunization and Respiratory Diseases, which led the agencys response to the COVID-19 pandemic. The maximum $13.6 million had already been obligated to Deloitte Consulting LLP for help with the restructuring, which included closing several research offices.Foley reported from Iowa City, Iowa.Ryan J. Foley, Associated Press

Investors in cryptocurrencies are seeing red today. In the past 24 hours, the prices of most major cryptocurrencies and meme coins have plummeted. This includes crypto heavyweights like Bitcoin, Ethereum, and XRP as well as popular meme coins like Dogecoin (DOGE) and TRUMP. Heres what to know about the biggest losers and possible reasons behind the crash.Bitcoin drops below $90K for the first time since NovemberUnsurprisingly, the crypto that is getting the most headlines today is Bitcoin, which as of the time of this writing currently sits at around $89,000 per coin. Thats a 7% drop in the last 24 hours alone. It also represents of few other ignominious milestones for the cryptocurrency king as of late.First, todays decline marks the first time that Bitcoin has fallen below $90,000 since November 2024, after Donald Trumps election victory sparked a crypto rally. Second, Bitcoin is now also down 20% since another important Trump marker: his inauguration day on January 20.On Trumps inauguration day, Bitcoin had hit an all-time high of over $106,000. The post-election victory and post-inauguration gains were largely fueled by the belief that a Trump presidency would be good for the crypto markets in general.But its not just Bitcoin that is plummeting today. Other cryptocurrencies are as well. Those include (as of the time of this writing):Ethereum, which is down almost 10% in the past 24 hours (and down over 27% in the past month).XRP, which is down over 12% in the past 24 hours (and down over 31% in the past month)Solana, which is down almost 12% in the past 24 hours (and down over 46% in the past month)Dogecoin, which is down almost 11% in the past 24 hours (and down over 42% in the past month)Official Trump, which is down almost 14% in the past 24 hours (and down over 56% in the past month)But what exactly is causing todays crypto crash? While the digital assets do tend to be highly volatile anyway, the general consensus among crypto industry watchers is that two main events could be contributing to the plunge.Trumps tariffs lead to macroeconomic uncertaintyWhile Trumps election victory was hailed as the best possible outcome for the crypto industry, it now appears that Trump could be doing more harm than good for digital currency markets.Thats because since Trump was sworn in, he quickly set to initiating tariffsor at least threatening toagainst Americas major trading partners, including China, Mexico, Canada, and nearly every other country in the world.Many fear these that tariff threats may lead to an all-out trade war between America and other major economies. Indeed, as BeinCrypto points out, after Trump seemed to confirm yesterday that the tariffs on Mexico and Canada would be moving ahead, Bitcoin sank and crypto markets saw almost $1 billion in liquidations.When there is macroeconomic uncertaintylike the kind generated by potential trade warsinvestors usually seek to lock in gains where they can by selling assets that have had good returns as of late. Larger anxieties about where the economy may be headed may be driving many crypto investors to take their profits now to buffer any losses in the future.The Bybit hack reminds crypto investors they are vulnerableIts not just Trumps actions that are rattling crypto investors. Last week, the cryptocurrency exchange Bybit was hacked, leading to $1.5 billion dollars of cryptocurrencies being stolen. It is reportedly the largest crypto heist on record and is believed to have been carried out by actors in North Korea, reports CNN.Major crypto heists in the past have rattled investor confidence, and this one is no different. Heists such as this remind investors that cryptocurrencies are more readily at risk of theft than other assets, such as stocks or properties.To put the Bybit hack into greater perspective, Reuters reports that in all of 2024, there was $2.2 billion worth of crypto stolen. The ByBit hack signals that 2025 may be an even bigger year for crypto heists.