The purpose of this CXTech Week 23 2023 newsletter is to highlight, with commentary, some of the news stories in CXTech this week. What is CXTech? The C stands for Connectivity, Communications, Collaboration, Conversation, Customer; X for Experience because that’s what matters; and Tech because the focus is enablers.
You can sign up here to receive the CXTech News and Analysis by email or by my Substack. Please forward this on if you think someone should join the list. And please let me know any CXTech news I should include.
Covered this week:
- Is Twilio the Canary in the CPaaS Coal Mine?
- Commio: Telecom Wise Guys, SMS Compliance
- Why Do Industry Analysts still not understand Programmable Communications?
- Webio Launches DebtChat
- Stand up for Open Source Software Patent Defense
- The Hows and Whys of Twilio’s Demise
- Guide on Optimizing prompts for GPT models
- And finally, My last post asking you to complete the Open Source Telecom Survey
- People, Gossip, and Frivolous Stuff
Twilio’s got Activists
The canary in the coal mine title is from a panel session at TADSummit 2023, created over one month before the activist shareholders got going with Twilio. TADSummit leads, it does not follow. We’re not a “book-club” that talks about a fashionable acronym or term. TADS has the tough conversations that can impact industries. There’s an existential threat to the programmable communications market, and because of Twilio’s size they’re getting hit first.
We learned over last weekend activist investor Legion Partners (previously got Ericsson to overpay for Vonage, that now owns at least a stake of $40M in Twilio) has met several times with the company’s board of directors and management, urging them to make changes to the board and consider divestitures, among other moves. Wall street is piling in as they scent blood, given the recent stock price rise, $40-$70 within the month of May.
The conversations come at a critical moment for the $12 billion programmable communications company. Its use of supervoting shares, which gives CEO and co-founder Jeff Lawson a voting stake of 21.8% even though he owns only 3.7% of the stock, is due to expire in less than a month under a longstanding arrangement. At that point, all the shares will combine to form one class, stripping jeff of his big voting stake and potentially opening up the company to more pressure from stockholders. Jeff has been buying Twilio’s stock, about $10M (158k at $62) in Feb, increasing his holding by 25%.
An Existential Threat to the Programmable Communications Industry
From my perspective the move to 10DLC and the use of TCR (The Campaign Registry) is resulting in significantly increased margin pressure on Twilio and all programmable communications companies. It also creates the odd situation of Kaleyra selling A2P services, while also running TCR. I know there will Chinese walls, but couldn’t an independent entity run it?
The carriers and their proxy Syniverse were jealous of Twilio’s valuation and are taking advantage of their state granted oligopolies to now charge between 0.4c to 1c for what was once sold at a wholesale rate of one thousandth that cost. 0.4 to 1c per SMS bears no relation to the full absorbed network cost, its punitive. This mode of operations is not new, the likely result will push the brands to IP messaging faster. In my recent personal experience, brands like Wells Fargo are now using email for 2FA, where once SMS was an option.
I strongly recommend you review the Commio video in this week’s CXTech newsletter, see below, in particular the section on using the campaign registry. I also recommend you consider Commio if you need to use TCR, they are exceptional at working with TCR and are really nice people. However, back to the focus of this article, the video is a positive review of a complex process that is confusing and frustrating the brands.
Twilio’s $750M ‘investment’
Twilio has the broadest programmable communications offer, dominant in SIP trunking in North America, FLEX, workflows, security, customer data, etc. BUT they can not escape the simple ubiquity of SMS based services. Which make up about half its revenues.
With the $750M investment Twilio made in Syniverse, I thought Twilio would have become part of the closed group of TMO, AT&T, Verizon and Syniverse who I believe interconnect outside the TCR at zero cost to themselves. Its a closed group, that enables the TCR pricing and processes to significantly disadvantage anyone not part of that group.
But it appears $750M did not but them into that closed club, they are still an outsider. I’m not too sure what they got for the $750M they gave to the Carlyle Group, apart from some Syniverse stock. The programmable communications industry is being choked, while the telcos are essentially cutting off their nose to spite their face as they can not deliver what the programmable communications industry does.
The Call to Action
Brands are moving away from SMS for their business processes because of price and trust. Price because IP messaging is now far cheaper, by a factor of up to 10,000. Trust is lost as the TCR processes are something from a British comedy show depicting civil service processes.
There is also the issue of SMS SPAM, it’s a nice money earner, crooks pay the carriers to SPAM elderly people and make money. The TCR was meant to stop that, but it has not, as described here from Giovanni Tarone; the more I’ve learned, the sadder I’ve become on what is happening in the US.
An amazing industry, programmable communications, which has delivered value to everyone, is being choked by oligopolists. The whole US programmable communications industry needs to act together to protect American consumers and enterprises. I’ll just remind everyone, if you have an Android phone and use Google Messages, SMS SPAM goes to SPAM and Blocked folder. It’s just like using gmail.
As you can imagine TADSummit is going to have a field day discussing what is happening and what we need to do as our policy is no BS.
I know the modus operandi at times like these is to keep your head down while the rich, powerful, and greedy kill innovation. This is an inflection point for programmable communications, we must act openly and together.
This episode of the Telecom Wise Guys is a great review of the practicalities of using the campaign registry. I also recommend you consider Commio if you need to use TCR, they are exceptional at working with TCR and really nice people. The video is a positive review of a complex TCR processes.
These marketspaces, universes, landscapes are heavily influenced by who is spending money with the analyst firm. The analyst firms will loudly protest their innocence, but it’s the truth.
This post from Twilio surprised me. They’ve kept a distance from the term CPaaS. Generally referring to themselves as programmable communications, as they are so much more than a CPaaS. However, Wall Street listens to analysts like IDC, and the CPaaS label has stuck, despite the category error it creates, see my article here. The world is complex, you should always beware a technology label being used for a market category.
I’ve brought the analyst problem up before, with this article, “CPaaS Report: A Problem Large Analyst Firms need to Solve“. If you see any CPaaS market size reports, know that its a report built on a category error, and given how TCR is impacting the industry, its future revenues have a large error bar at the moment.
Conversations about credit and debt can be handled much more effectively if they are handled conversationally. Webio has just launched a new product, DebtChat.
DebtChat is part of constellation of inter-related and reinforcing technologies at Webio to bring you the next generation of AI powered digital solutions. Paul will also be presenting at TADSummit in October.
Congrats to team Webio Ltd.
The United States Patent and Trademark Office (USPTO) is considering rules that will help patent trolls, raise patent litigation expenses, and increase the volume of patent litigation overall.
We have seen open source targeting by patent trolls many times, including a direct lawsuit against the GNOME Foundation in 2019. You can help support open source ecosystems, reduce wasteful litigation by non-practicing entities (NPEs), and stop the assertion of bad patents.
All it takes is a few minutes of your time. The volume of comments is important – please share this with your friends and colleagues who want to protect open source software from patent attacks.
This is an insightful article from Thomas Howe, written in 2019.
The key point Thomas makes is, “The next generation of CPaaS is conversational, because the future of CPaaS is to aid in the collection and dissemination of real time data from people.”
He makes a critical point that backs up why my focus in programmable communications, not CPaaS. “The building blocks of the CPaaS are text messages, voice sessions and API calls. Because of the history of CPaaS, and the history of visionaries that built them, these were logical choices, but they are inefficient from a development standpoint, and will be quickly commoditized from a business standpoint.”
Twilio is further up the conversation stack than most, yet it can not escape where the money is today. I would posit, Thomas is highlighting CPaaS’s demise, rather than Twilio’s. So those focused principally on CPaaS, will likely struggle in the coming 2-3 years.
Paul Golding did a great post highlighting a much needed resource.
Exciting news for GPT enthusiasts! OpenAI has a comprehensive guide on optimizing prompts for GPT models, especially GPT-4: https://lnkd.in/gy9Arfap
If you’re deeply engaged in research with ChatGPT, you may have already stumbled upon most of these insights. Rather than reiterating them, I’d like to emphasize two points, especially for those unfamiliar with these new tools.
Firstly, let’s put to rest the misconception that “prompt engineering” will become obsolete. OpenAI’s guide highlights a fundamental reality: “GPTs can’t read your mind.”
Language operates in a way that necessitates ongoing prompt manipulation and creativity as we leverage GPT’s programming-like capabilities.
Secondly, the guide emphasizes the significance of incorporating a persona to elicit specific types of responses. Keep in mind that a language model draws from a vast array of training sources related to your topic. Without guiding the model’s focus, it tends to generate results in a general sense. By introducing a persona, such as “You are a tax expert,” the model will search its knowledge base for tax-related insights from the perspective of professionals like accountants.
However, you could equally begin: “You are a comedian…” and get a list of flippant and jovial remarks
To take it a step further, consider starting your prompt with something like: “You are a tax expert who specializes in advising clients on saving money…” This nuanced approach will yield even more targeted responses tailored to your specific needs.
Wow! Thank you to everyone who has completed the survey. Some of the bigwigs of our industry have completed it. If you haven’t, please include your opinions, they matter.
People, Gossip, and Frivolous Stuff
I just missed including Tobias’s post in last week’s newsletter. Tobias Goebel is now Product Marketing Director at Kore. He moved over as part of the Twilio IoT acquisition. At TADSummit 2022 he gave a couple of great IoT presentations, and we’re hoping for more at TADSummit 2023.
Steven Goodwin’s book has a new cover. Buy it!
And yes, that’s James Body doing his dangerous demos shtick 🙂
Reading the news and seeing the Hubble Network announcement. I checked the calendar, nope, not April 1st. This has a whiff of BS to me, or should I say Blue$h!t.
Yes the top end of BT and bottom end of LEO overlap at a power level of 100mW, but most BT operate at far lower powers, like 2mW (3dBm). Hence their claim of 1000X sensitivity at the space segment.
The BT protocol assumes low power limited reach operation, so hops between 79 channels as it assumes there aren’t that many transmitters locally. So that’s going to be a lot of interferers at the satellite, the noise will be quite something given WiFi, microwaves, and everything else operating at 2.4GHz
I’m not an expert, but the limitations of link budget and protocol seem hold-points. Of course innovation requires we do crazy stuff, perhaps a hack with a high altitude balloon over a city to show the principle would be a cheaper experiment?