eWeekTweetChat

So true!

Automation is essential. From data ingest to insights. It's really the only way to stay competitive

(edited)

A4: Every pipeline has #data producers & data consumers– understand the requirements for each. Start w/ a customer-back approach to #dataanalytics by understanding end users & their needs for business solutions – then work back leveraging data to solve the problem.

A4: Bring together your data and analytics in one seamless experience. The more disparate systems you need to manage, the bigger the silos, the greater the challenges and margin for error — things your business can't afford

A4. Hire the right leaders that elevate the team. Simple but applicable to anything. And start small, but think big.

A4 - Intentionally creating a culture of being fact driven. This is hard and needs to be a strategic initiative at the CXO level. #eweekchat

A4:A key practice is implementing scalable data analytics architecture—cloud native architecture—because it allows you to scale compute and storage separately.

A4. Automate, automate, automate. It's the only way to free up smart people to deliver meaningful data products.

A4 To optimize #data analytics, we must first catalog and centralize all #metadata. This helps us understand the data and where it sits. It allows us to contextualize data across separate silos, truly unlocking #value while increasing #DataLiteracy and #DataCulture

A4: Let complex data speak for itself. Embed AI/ML into your #analytics practice. This improves decision-making and ensures you’re identifying the right problems, and implementing the most effective data-driven solutions. We call this
#IntelligentExploration

(edited)

A4 - @rqrivera - now 92% -- attribute the “principal challenge to becoming data-driven” to people, business processes, and culture, with only 8% identifying technology. #eweekchat

"Business insight driven/oriented data discovery and analysis" is the best practice mantra that can ensure organizations make the best use of data analysis investments. #DataAnalytics #eWEEKchat @CIOStraightTalk

A4: Focus on building a data fabric on an Open Data Lakehouse that enables you to work with any dats exactly as required by the business — no work arounds or hopping between point solutions. Silos and lock-in are your enemy. #DataAnalytics #eWEEKchat @CIOStraightTalk

(edited)

A4: Be picky with your data. Trying to employ #ML or #DataScience on bad data yields bad results. Choose tools that help you focus on finding quality data easily and reliably (less effort on security, etc.) so your hard work on top of it is not skewed or faulty.

(edited)

@RKs2cents 100% agree, metadata capture, normalization and management is crucial

Clean and consolidate data into a single, actionable view. Siloed data, retained in different systems and platforms, has little value. #DataAnalytics #eWEEKchat @CIOStraightTalk

A4. Understand the limitations of your analytics. Especially the limits of your dataset and algorithms. Don't follow blindly, verify the outcomes. Treat data as augmentation for intelligent humans, not replacement.

A4: Organizing data. Without sorting, labelling, and cleaning your data in a way that everyone can agree on, you're not going to get results you can replicate. Having results that can be replicated creates trust and buy-in in your insight.

@rqrivera Agree, but it's also about having a #DataCulture within your organization.

@RKs2cents metadata and semantic layers will be extremely important in the coming years, if not now.

@namessanti I love that concept. I wish there were better tools and better integrations - especially for seamlessly analyzing datasets across multiple silos.

@RKs2cents 100%.

A4: Executive buy-in to see analytics as distinct within data and dedicate the budget to build out and focus the function, structuring it as a critical strategic engine to compete and win

A1: The truth here is that getting it right for your business can be challenging. Every business challenge is different, every data set unique. Do everything possible to have your cake and eat it too — enable fast collaboration across hybrid and multi clouds

@rqrivera "In god we trust; all others bring data!"

A4. One for the data teams Benchmark your data maturity to know where you are, what’s the state of your data, how bad is it? Get metrics around it, so you can measure it but also understand how you can improve the data. That itself will identify where your weaknesses may lie.

A3: Being understaffed, under-resourced, and performing data management functions

A3:The scale of platform as data is exploding. The ability to effectively derive insights depends on how you scale data analytics platform for big data processing and scaling of ML deployments. As data sets become massive, scalability is a challenge

(edited)

A3: Datasets are rich and complex: a challenging initial step is understanding data quality (e.g., what information is missing and how that may impact decisions)

Other challenges are identifying key relationships within data that impact business decisions and AI

A3 1/2 There are many challenges with #data analytics. The vastness of data is one huge hurdle. It often hampers our ability to determine the right data and drives us to constantly question #DataQuality. There are also issues around low data literacy and siloed user

(edited)

A3: By far, the biggest challenges are in operationalizing #data at scale w/ rapidly rising stakeholder expectations on speed, flexibility, timeliness & customization of new capabilities.

A3 Not all businesses are equally ripe for data transformation. A wide variety of factors, from budgets to industry-specific challenges, serve as either a challenge or opportunity to transformation.

Top 3 challenges revolve around People, Data and Technology. It's difficult to hire and retain the right skills, difficult to have right data and finally, difficult to have the right platform for specific needs. #DataAnalytics #eWEEKchat @CIOStraightTalk

A3: Data unification and evangelization. Companies will link tools, move data around or find ways to connect to a data warehouse — but none of these solutions address the fundamental problem of getting their data to work in the first place. (1/1)

A3: it ultimately has to do with two key pieces of the puzzle: tech and people. You need people that can manage the entire data lifecycle, but you also need the technology to fast-track collaboration and results securely

@kklive (Or, you know - top 3 challenges - people, people, and people. 😉 )

A3: Getting value from data analytics does involve sharing it between tools and generating insights, but you need the data to be in a condition where you can make sense of it. This means unifying and stitching data from all the different systems it touches. (2/2)

Growing platform complexity over time. "New data type? Add a new tool!" IMO, this is why there's a clear industry trend towards comprehensive data platforms, to reduce fragmentation. Second is choosing the right data. Less better quality data > overwhelming amounts.

A3 Avoiding vendor lock-in will remain a challenge as more organizations migrate to the cloud and strive to ensure better financial governance. The shift towards open data ecosystems that create options to manage data in a disparate, distributed and in a hybrid capacity will help

@namessanti Take the Cloudera Data Platform as an example. it enables everyone from streaming developers to data engineers, to analysts, and scientists to work together in a transparent and interconnected way. But these technologies only work well with the right people using them

A3. But more seriously - it is still lack of trust. Black box algorithms, incomplete datasets, unexpected results, unverifiable predictions. Hard to know whether you’re getting it right or not.

Also, from a data perspective, the 3 biggest challenges include data quality, data diversity, and data latency. Being able to deliver high-quality data, which connects data assets across the org’s data sources is essential. #DataAnalytics #eWEEKchat @CIOStraightTalk

A3 2/2 Metadata-driven #datafabric is a key trend. To rein in our #data, we must have a holistic view. The idea is to create a single pane of glass across all the data in our environment. A fabric stitched across all data is necessary because the silos aren’t going away.

(edited)

A3. Defo agree w/ others talking about #data quality though. GIGO is amplified exponentially when we talk data scale!

@AndiMann Completely agree. Trust in data is the biggest thing we hear from CEO in our conversations.

For sure - lack of trust (in data, analytics, algorithms) it has the potential to take us back to 'if it's all just opinion, then let's go with mine'

A2. We also have a real push into AI/ML use cases that isn't just 'let's do AI' - 'boring' tasks can now be automated with LLMs. I'd expect to see AI Explainability, Transparency, and Ethics to be a part of any org adopting it at a strategic level.

100% — This is clearly where having turn-key options is necessary. At Cloudera we built these interoperable pieces from SaaS with CDP One, to true interoperable hybrid deployments for complex businesses

A2: There are a few mega-trends; #DataOps, democratization of #data & the convergence of physical data (e.g. IoT) with other sources near the edge. DataOps allows you to leverage #automation technologies to free up data engineers for higher value projects. (1/2)

A2: Trust in #data is the biggest trend. We need to know the data is clean and understand why and how it’ll be used. That’s why we must add meaning to data and apply data governance and best business policies to ensure company and regulatory rules are followed.

(edited)

Q2: I would put at number one XAI as a way to increase trust and adoption. A way to easily deploy at a larger scale inside an organization is another must have, same for augmenting people and decisions.

A2: Also, data is being leveraged outside of IT – driving more value in the lines of business. + new tech such as edge platforms &advanced connectivity are revolutionizing the type &frequency of #data collected on devices, enabling a whole new set of edge & #IoT application (2/2)

A2: The fact that most data is being generated outside the cloud, but most analytics are designed for cloud - creates a big gap that needs to be filled.

A2:One key trend comes to mind. Data analytics are increasingly being asked to drive greater multicloud data freedom. Analytics can drive data freedom (to move, to deliver value, etc.) say, via open data lake architectures that leverage open source frameworks.

(edited)

A2:We are trying to accomplish it with Lyve Cloud Analytics platform at Seagate

(edited)

Contextualized analytics centered around business aware insights, which are driven by scale AI, are what businesses are looking for. Insights driven by AI devoid of issues including biases, diversity, etc., are in-demand. #DataAnalytics #eWEEKchat @CIOStraightTalk

@BCorr_BMC #DataOps is a trend @HitachiVantara has been following for a number of years. It is all about agile data management.

(edited)

A2 Deep learning, leveraging massive and varied datasets to get to unique outcomes; stream/edge processing to reduce cost, increase speed. But the holy grail of predictive analytics still seems far off!

At Amperity, we are working on a new predictive model that will help brands identify audiences of customers that are most likely to complete important events, like loyalty or credit card sign-ups. This insight drives efficient resource allocation for marketers.

A2: A big one is real time data flows and streaming — getting data from anywhere to anywhere where it can drive analytics as fast as possible. Every second counts and can be the difference between leading the pack or millions lost.

@brucekornfeld We're seeing the same, with Gartner predicting that 75% of data will be at generated at the edge

A2: Aiming for visualization, considering data democratization, data fabric technologies, data mesh approaches, and accounting for Internet of Things (IoT) data

A2: Let's say you’re a manufacturer. How do you collect massive data from hundreds of sensors, customer data, supply line information and get down-to-the-second insights? This is where universal data distribution and using data at the edge shine.

A demand for real-time insights driven by sensitivity to FX & interest rates, increase in fraud and cyber activity, as well as supply chain challenges. People need instant insight to act immediately. #DataAnalytics #eWEEKchat @CIOStraightTalk

@BCorr_BMC Hey Bill, good to see you here. Agree 100% with that, we're doing pretty good as geeks with AIOps, O11Y, etc. - but he real returns are when you get to novel decisions for business units.

A2 - Semantic Layers, Metadata, just look at #dbtcoalesce. Best-o- breed solutions will need solutions that integrate with one another easily. #eweekchat

A2 One trend I see is that expectations for speed to value continue to increase. This means a tighter focus on improving data management automation that fast tracks devops cycles, & improves operational efficiencies around building data pipelines, ML models and data accessibility

A2: Simplification and unification are big trends, especially as customers add more tools to get at increasing amounts of data; it's one of our focuses with #DataCloud - we also see strong interests in open formats especially @apacheiceberg and @apacheparquet.

@RKs2cents 100% agree - with automation being a key driver

@chimerasaurus Simplification is a big call - we are far from making analytics simple. Successive approximation perhaps?

@BCorr_BMC When you say on the edge do you mean my non-technical or business users? or using NLP queries?

@chimerasaurus Iceberg vs. Delta will be the next big war (I know you're already in it) :)

A2 Data mgrs, CIOs & CDOs will increasingly focus on improving the fluidity of data access and usage – data democratization. One consequence, the debate between centralized and decentralized data teams, data as a product & supporting architectural paradigms will reach fever pitch

@namessanti Santiago, we agree with you. IT/OT data management is another area @HitachiVantara sees growing in importance in the future. A lot of that streaming data comes in as OT data.

(edited)

@rqrivera Anything that is running outside the data center or the cloud (eg. IoT devices)

@hmschwenk . Totally agree! Cycle times are decreasing and the demand for insights is increasing. This is where Live Analytics from tools like ThoughtSpot are extremely beneficial. #eweekchat