Enterprise hits and misses - Retailers hit Black Friday numbers, DevOps gets quantified, and AI projects get real
- Summary:
- This week - retailers get a pretty decent Black Friday, but questions linger. AI projects get a reality check, as data on ROI and employee buy-in rolls in. DevOps isn't easy, and neither are open source language models (just ask Meta). As always, your weekly whiffs.
Lead story - Wrap up warm for AI Summer, or you might catch cold - AI lessons and cautionary tales
As AI adoption continues to surge, burning enterprise questions remain, not the least of which is ROI. A Chris notes in his AI summer piece:
A new report from IT services and consulting firm Infosys reveals that most companies are still failing to convert data science into what they really need: measurable business value.
So how do organizations avoid this? A more comprehensive/strategic approach is in order. Chris:
Among the leaders in AI adoption, the most satisfied are those that have focused on improved data sharing, trust and ethical usage, and integrating data science with their line-of-business functions. This suggests that strategic initiatives in support of clear, top-down aims are more likely to succeed than tactical, hype-led programs.
Chris cites a recent session from the Information Commision's Office:
At the ICO, we see a bright future for AI in the UK, underpinned by terrific trust and confidence in how AI is developed and used. But the actions of an irresponsible few risk undermining this.
"Irresponsible few?" Problematic examples include controversial facial recognition systems, unproven use cases like emotion recognition, and overreaches like Meta's latest open source language fiasco, Galactica (I'll get to that shortly). Leading Chris to conclude:
The lesson is we may be expecting too much too quickly from some AI ventures, while losing sight of obvious risks and challenges. As decision-makers, we need to do better than that, especially when handling the personal data of millions of citizens and customers, many of whom may be vulnerable.
So, we press on. Chris examines a path forward in AI demands a contract of trust, says KPMG. Given the US Thanksgiving holiday, Neil served up a timely argument for why AI can - and should - be applied to food insecurity in Can AI help us address food insecurity and food waste?
Diginomica picks -
- Retail transformation updates - the shifting balance in the omni-channel. Just when retailers had enough to contend with in the Vaccine Economy, inflation and consumer price sensitivity adds an unwanted twist. That said, retailers have to be happy with the largely upbeat Black Friday results. (Black Friday online retail rises - Adobe and Salesforce crunch the numbers). Stuart also reports on three retailers contending with these shifts, starting with Macy's: Digital reverts back to pre-pandemic levels at Macy's even as bricks-and-mortar sales continue to decline. Also see: Stuart's updates on Nordstrom (supply chain a bright spot), Wayfair and Williams Sonona.
- Web scraping bad, information harvesting good? Why there's a need for new legal frameworks in a data economy - Martin gets to the heart of the data economy:
Vendor analysis, diginomica style. Here's my three top choices from our vendor coverage:
- Zoom’s enterprise growth steady, but share price falls on online segment outlook - Derek on Zoom's latest challenges: the segment blues. But are investors too busy bemoaning Zoom's consumer demand crunch to note its enterprise progress? Derek: "In my mind, the metrics that matter (in the enterprise business) are solid and suggest a positive outlook."
- IFS customers shed light on the core role of data in servitization and XaaS - Phil with a meaty post that traces the dual evolution of servitization across Saas - and manufacturing. Also see Phil's use case, Customer success from the viewpoint of IFS customer Hexagon Agility.
- Ceridian Insights - drilling down on the latest product announcements - Brian finally went to an enterprise show with enough
Dr. Peppersubstantial product news to keep him satiated.
Customer use case selections:
- The underrated practice of ERP systems automation - Scotts Miracle-Gro shares its SAP example - Avantra use case (Jon)
- INGKA Group (IKEA) adopts Celonis to create ‘the perfect order’ - Derek
- Managing 400 TUI Resort Hotels with a central cloud data warehouse - Gary
A few more vendor picks, without the quotables:
- ClickUp's viral strategy to connect and manage work across the enterprise - Phil
- Toyota Connected improves emergency response experience with Twilio Flex - Derek
- Blue Yonder’s Corey Tollefson - ‘We have the capital to invest and we will compete on service excellence’ - Derek
Jon's grab bag - Madeline profiled a company that plenty could learn from: ServiceMax shows how gender parity in tech can be done. She also took a hard look at Black women in tech - a UK perspective. Finally, Neil issued an important post on why semantic layers need a rethink: We need a real semantic layer - but something is missing:
Semantic abstraction benefits analytics by moving information integration to a new level, where intelligence can more easily and swiftly proliferate throughout an organization. Productivity will increase because machines will make everyone less reliant on a small number of go-to super-users and help us get beyond today's rigid, often brittle schemas and thin metadata.
Sounds like pretty good marching orders for a BI team near you...
Best of the enterprise web
My top seven
- DevOps is hot but most IT pros say practices need improvement - Joe McKendrick throws more cold water/sobering data on DevOps: "Security is the number one driver behind most DevOps and DevSecOps implementations. Yet only 30% feel confident in the level of collaboration between security and development."
- Is there a simple answer to developer burnout? - As it turns out, no: "The problem of developer burnout cannot simply be solved by automating their processes and increasing salaries."
- Outsourcing will shine or fail as we combat this global assault on our stability - Phil Fersht of HfS filed a vintage purposeful rant: "If you can’t keep hold of your talent now, you’re in real trouble as a services firm. Enterprise customers are walking away from providers whose delivery quality is falling apart from staff turnover."
- Report finds employees embrace AI when they see its value - Continuing our make-AI-projects-count theme is this new study from MIT Sloan Management: "AI often helps workers to need less management oversight, which creates a feeling of autonomy and can improve job satisfaction." A glass-half-full view to be sure, but we could probably use that right about now.
- How to create winning industry solutions - Thomas Wieberneit riffs on a recent CRMKonvos session, where yours truly was given the hot seat as we sorted through industry cloud hype.
- 5.4 million Twitter users' stolen data leaked online — more shared privately - This breach happened before Musk took over, but it highlights the concerns many have over how Twitter will fare from a security standpoint amidst cutbacks and business model shifts.
- ‘Full-on robot writing’: the artificial intelligence challenge facing universities - I've mocked the ability of AI to create compelling art, but helping stressed out university students file mediocre assignments is another matter entirely, and universities need to get out in front of how AI is changing how students study, test, and create.
Whiffs
Another week, another retracted AI open source language model. This time, Meta did the (dis)honors:
Why Meta’s latest large language model survived only three days online https://t.co/jgfRa06XDA
"Galactica is that it is not able to distinguish truth from falsehood."
-> other than that, it was a really nice model for scientific text :)
oh well, it was a nice three days
— Jon Reed (@jonerp) November 28, 2022
Bonus: the I-hope-you're-happy petulance from the lead Meta scientist on the project... Meanwhile, the breathless search for the next click, err, the next-big-thing continues:
A bot that watched 70,000 hours of Minecraft could unlock AI’s next big thing https://t.co/t74T9TQDuu
-> it's very unclear how the editors made a leap from another source of training data to the "next big thing" - except perhaps the temptation to exaggerate for clicks
— Jon Reed (@jonerp) November 26, 2022
Picked this one up for Neil Raden... Something tells me that a reality TV deal for the next chapter won't be forthcoming...
Todd and Julie Chrisley sentenced for bank fraud, tax evasion https://t.co/Wc5TXGw2Fr
-> Crisley knows best is really being put to the test here.....
— Jon Reed (@jonerp) November 23, 2022
Oh, and the details on how Alexa loses gobs of money are coming in dribs and drabs:
Amazon Alexa is a “colossal failure,” on pace to lose $10 billion this year https://t.co/wbl6EF58FV
"conversations were trivial commands to play music or weather. Those questions aren't monetizable"
-> servitization only works if you can monetize the services :)
cc: @philww
— Jon Reed (@jonerp) November 28, 2022
It used to be fun to mock Alexa; now I worry one day, Alexa won't wake up, and I'll be back to setting manual alarms, and figuring out how to turn music on. All is forgiven... If you find an #ensw piece that qualifies for hits and misses - in a good or bad way - let me know in the comments as Clive (almost) always does. Most Enterprise hits and misses articles are selected from my curated @jonerpnewsfeed.