Chart - cumulative relative share of social media discussion by volume - DataSwarm Analytics We have used the DataSwarm Analytic Engine we have been developing to look at the odds for the various possible contenders at this point as the starting gun ...
‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ 

Click here to read this mailing online.

Your email updates, powered by FeedBlitz

Here is a sample subscription for you. Click here to start your FREE subscription

broadstuff"broadstuff" - 5 new articles

  1. Predicting the Tory Leadership race using social media analysis
  2. Talkin 'bout a Revolut (ion) - DataSwarm Research
  3. Inconsiderata
  4. All you ever need to know about Blockchain in one short post
  5. Retail banking in the Fintech Age
  6. More Recent Articles

Predicting the Tory Leadership race using social media analysis

Chart - cumulative relative share of social media discussion by volume - DataSwarm Analytics

We have used the DataSwarm Analytic Engine we have been developing to look at the odds for the various possible contenders at this point as the starting gun goes off. It goes without saying that Boris Johnson ("Boris") has the bulk of the memeshare at the moment (see the chart above) - but it also goes without saying that the Tory favourite at the beginning tends not to win the race.

Anyway, below is the candidate odds running chart at the moment, and the odds expressed as a % probability - more details on the how it is all done is on the DataSwarm website:

Boris 42.9%
Hunt 11.4%
Gove 8.8%
Stewart 6.0%
Rudd 4.0%
Raab 4.0%
Lidington 2.9%
Davis 2.9%
Hancock 2.5%
Mordaunt 2.0%
Javid 1.8%
McVey 1.3%
Gauke 0.7%
Truss 0.3%
Brokenshire 0.1%

Talkin 'bout a Revolut (ion) - DataSwarm Research

Diagram - ongoing down and upspikes in perception of the topic of Fraud for various banks

No sooner do we do some research on the problems of retail banks managing perception (using our own DataSwarm analytics suite, natch) and find that the Challenger Banks/Fintechs will not be immune, than the proof comes down thick and fast on one Challenger's head (Revolut).

In a few short days they have suffered bad publicity over parts of their system's issues re money laundering tracking, their uber-culture, and their App not working.

Our research had shown that overall, the "Challenger banks" like Revolut had better customer perception overall, for a number of reasons (seehere for the relevant blog post). At that point it was only the Traditional retail banks that were being hit by negative stories, due to ongoing complex operations and old bad news continually resurfacing - a factor we call "perception drag". You can see this on the diagram at the top of the page, its those big downspikes. However, we also wrote (to quote the report):

These types of issues do not yet impact the Challenger banks, but we would suggest that this is currently because they haven’t been around long enough. It is yet to be proven that they will not fall into these sorts of policy errors and PR gaffes over time.

No sooner had we finished the report and turned the trackers off than Revolut fell into one mess a few weeks ago, and now all this.

Fortunately, not all is bad news. If you look back up the diagram, there is an interesting effect if a bank chooses to try and adress these issues - again from the report, look at two of the spikes:

October 27th – Lloyds faces probe alongside KPMG over HBOS possible fraud a decade ago

November 14th - Lloyds shortlisted for an award for anti fraud detection system

Being seen to try and fix he things that go wrong is a good antidote to the downspikes. What we also know' from our work experience with cients is that not only the devil, but salvation, is in the details - the report again:

Digging down to the next levels [in the perception data] reveals which specific factors are causing positive and negative impressions, when, and why, for each issue. At that point it becomes possible to attempt to optimise perception from an informed base.

Our experience is that the impact of perception on sales and service can change market structure over time, but is not well understood, especially the effect of service on future sales. It is an opportunity for most companies to improve on significantly.

And one last crumb of comfort - in matters of publicity, to quote Wilde, "the only thing worse than being talked about is not being talked about...."

The DataSwarm report "Retail Banking in the Fintech Age" is available here



An ode to the digital comminications technologies of our times, and the wannabe influencers who infest them:

Go with rage into the noise and haste, and remember there is no value in silence. As far as possible never surrender the mike, get your opinions out to all other persons.

Speak your truth loud and often, to all the others, they are mostly dull and ignorant and they need to know they are wrong.

Override quiet and thoughtful people, they are impediments. Compare yourself with others frequently, ignore those lesser than you as they are a waste of effort, and suck up to those still better than yourself.

Trumpet your achievements, and broadcast your plans. Focus on your own career, don’t be humble, this is your way to make your fortune.

Exercise duplicity in all your business affairs, there is one born every minute. Do not be blind to virtue, idealism or quiet heroism, there is always a way to use those suckers.

Be You, but feign whatever you must to get what you want. Especially feign affection, the losers always fall for it as they are continually seeking love, and anyway they are used to falling on their arse.

Resist the counsel of the years and try and look as young as you can. Nurture a feeling of self-worth to get you over misfortunes, and ignore any dark imaginings, fear is for wimps.

Have a good workout discipline, and always forgive yourself anything. You are a master of the universe, much more than the trees and the stars, it needs you to be here.

Labour and aspire to create noise and bustle in life, peace is for the meek souls. It’s a beautiful world despite all the drudges and dreamers, so don’t worry about anyone else, be happy.

Note to self. Take 2 frog pills, eschew social media, and see if someone will sell you one of those devices that splats mobile phones on trains....

(With grovelling apologies to Desiderata)

All you ever need to know about Blockchain in one short post

To say that blockchain is overhyped is to put it mildly - not a day goes by without some pundit proclaiming that industry X will be transformed by it. Most of this is speculation, based - as far as I can see - on little understanding of the technology or the viable alternatives. For most proposals, somewhere between the hype and heuristics is a major disconnect.

This post is not going to go into the ins and outs of blockchains, but will just point out the basics of what defines whether it will transform Industry X (this post summarises a more detailed article I wrote last year - go there for the more tech stuff - in our preparation for the blockchain work last year for helping to run the London part of Global Legal Hackathon.

In short, its all about ROI on the transaction costs..... in essence blockchain technology (as designed today) is great if you want transactions that are:

secure (ish - the overall flowchain of a blockhain system has shown itself to be prone to all the standard hacking problems),
no trusted 3rd party (TTP).

But, owing to the high calculation load to be secure/resilient/not require a TTP, for blockchain to work it also needs transactions to be:

relatively low volume (c 6 a second for Bitcoin for eg)
not particularly time sensitive (hours, or even days to complete the agreement between nodes)
relatively high transaction value - as these calculation loads massively increase the blockchain transaction cost.

This is captured in the 2 x 2 matrix above - there is only one "sweet spot" square for blockchain, that where the value and rate of transaction flow makes blockchain a useful solution. So for applications that are:

High volume, Low value - blockchain is a lousy solution in every single way unless the need for the secure/resilient/TTP overrides everything.

High Volume, High Value - the value means blockchain transactions cost in, but the high volume and resulting time lags will probably break the system or render it useless for most cases

Low volume, Low value - can cope with the volume, but unless the need for the secure/resilient/TTP overrides every consideration, or one is prepared to fund the transaction in another way (VC stump money, subsidies, indirect gains) there are far better approaches around today.

Low Volume, High Value
- this is the sweet spot - transaction values make it economic to use, volumes don't break it.

Worryingly, for a lot of mooted applications there is not a hope in hell that Blockchains (in their current forms), will be fast enough or cheap enough to work. To be fair, there are now more and more blockchain approaches entering the market, but there is always a trade-off of speed and/or secure processing. Here is what is more likely to happen in most cases:

1. For most non "sweet spot" blockchain applications there are existing, cheaper, faster “Good Enoughs” today. There will nearly always be an alternative case to employ more scalable/lower cost/faster performance blockchain system designs with:

Less distributed architecture to scale it for speed and cost
Less complex security in the blockchain to reduce the processing load time and cost

2. This will probably be done by implementing them as “private and much less distributed” systems behind IT datacentre security walls, ie “Walled Garden” blockchains will very probably put themselves in the position of the TTP supplier. (This is what is in fact happening today, I first saw this predicted in 2015 by Dinis Garda looking at IoT outcomes).

Which of course negates a lot of the theoretical benefit of the Blockchain idea itself.....


Retail banking in the Fintech Age

For the last day or so Revolut Bank (or more accurately, non-bank) has been in trouble with data privacy advocates, singletons, vegans, users and now possible even the regulators

For the new "Challenger banks" this is a very bad idea.

We have been researching perceptions of Challenger Banks vs Traditional Banks from October 2018 to January 2019, but in a nutshell, perception matters for capturing and retaining customers, especially in an age when 90%+ of people go online for recommendations.It also matters more for the Challenger banks, who don't have large service portfolios - so they need to be seen to be really, really excellent at what they do. Implying they spy on their users to the extent the regulator is interested (never mind the rest of the fallout) is a bad idea.

Revolut have not done as well in the perception stakes as their main competotor, Monzo - another Challenger bank - as the chart above shows (the faster the rise in the line, the better), their perception is lagging and is not much better than the Traditional banks, with all their history dogging their steps. This latest episode is not going to help (and may well point to deeper issues)....

More analysis on the DataSwarm site over here

More Recent Articles

You Might Like

Safely Unsubscribe ArchivesPreferencesContactSubscribePrivacy