The emperor criticized Mozart's composition with, "Too many notes."
Twitter allows only 140 characters to get a point across.
Automation in Banking - 2014 needs 628 pages to tell its state of technology story.
I like one-liners.
So here are some short messages that cover a broad spectrum of banking technology for almost everyone who deals with the space.
1. Transaction processing is in excellent shape. Usable information to better operate the bank is only 30% complete. The right people (5 companies in AiB-14) are working hard to grow the 30%.
2. Functionality is what bankers look for when acquiring technology, not system architecture.
3. IT technicians care about architecture but they don't decide on bank software procurements.
4. Finding best architecture and best functionality is like finding the best luxury car and gets 45 mpg.
5. There are no bargains in technology. Be skeptical when any vendor offers the lowest price.
6. Bank tech vendors do not sell commodities. Every solution has at least 30% uniqueness. That 30% should correlate with the bank's wish list to achieve a best-fit result.
7. Uniqueness is in the heart of the buyer. Don't mess with a banker's heart if you can find it.
8. Scalable software is an individual determination. Pity the $200 million bank that thinks it should have Chase's tech system.
9. There's a huge gap between the developer of software and the user, like PhD and 6th grader.
10. The best loan system without a user manual is like a mother without a set of rules for the kids.
11. The simplest test to determine system effectiveness: When bank employees say "their" system, it stinks. When they say "our" system, it's fine.
12. Wrong reason to change a core system: It's time for a change. Right reason: A consultant documented 200 debilitating deficiencies in the present system.
13. Three ways to post transactions: real-time, memo post, day-delay. Time it takes to alter a custom system: We'll get back to you on that; It'll be in the next release; They're working on it; It's in the pipeline; We've got a three-year backlog; Haven't
you heard from our development guys yet?
14. Custom system is a euphemism for, You're driving a Lamborghini cross country and you broke down in Wapiti, Wyoming. Who ya gonna call?
15. Running on a cookie-cutter core system is like driving a Lexus in Miami. Any mechanic can fix it.
16. The best CFO won't compute the true costs of a new tech correctly, and the CIO won't teach him. And the CEO will believe the lowest costs. When the 12th vendor invoice arrives, they all learn the truth.
17. When a mid-tier bank converts to a new system, it will acquire core plus 18 ancillary solutions. The solution that will make or break the selection is the commercial loan system.
18. The best commercial loan system is 44 years old (legacy system), is used by the toughest banks (thus the best functionality), and only 80% of the top tier banks (125) can afford it (no bargains in technology). See p. 279, Automation in Banking - 2013.
19. Don't be fooled into thinking technology is a profit center. You have to have a top line in order to have a bottom line. Technology spends money. If it made money tech companies would own a few banks themselves.
20. As mobile gives bank customers more DIY, and as Chairwoman Yellen continues with 0% interest rates, and Congress imposes more burdens, banks will need to find ways to charge customers for what they do. Puzzled? Ask gmail, yahoo, hotmail and other free
internet providers how they do it. Banks might have to start selling advertising to their customers. "Need a good deal on a car? Go to Happy Harry's and we''ll finance your car."
21. When was the last time you read about a Gartner or Pew survey that was conducted at Hamburger Heaven in Palm Beach? Never? It's because the customers there know a lot but they are few in numbers. They have the bucks and the Bentleys but banks are now led
by how twenty-something customers use their bank, very few of which have billions and the rest have hundred-dollar balances. Good luck with that demographic if you're a bank.
22. I confess to a huge bias when it comes to bank employees. I begin my assignments at the bottom, with women, in crowded offices, lots of interruptions, and grounded honesty regarding the system's hurdles that keep these loyal people from performing their
tasks more efficiently.
23. Every bank should have an internal, experienced IT person to perform computer system-related audits, even if it's part-time. The bank auditor isn't doing it, and the external auditor suffers from too much political correctness to disclose the truth. Even
regulators aren't doing the job completely.
24. There's a direct relationship between technology's ubiquitousness, as well as volume, and the frequency of criminal data assaults. Common one-liners voiced by bankers: "We'll get to it." hackers: "Been there, done that, we'll be back."
25. We live in a society where close enough is good enough. Look at the "list of 55" from the 2008 financial crisis. They were the people we trusted.
26. There are 14 good core system companies serving 13,700 financial institutions in the U.S. The Top Five have 93% and the other nine have 7%. The 14 have one thing in common - they know their business and prove it every day.
27. Every year, 200 financial institutions in the U.S. figure out their core system isn't right for them, and they acquire a new one. Most of the switches are by small banks; 60 are by mid-tier; none or one is by a top tier bank.
28. Lending could use a strong dose of IT intelligence. Bankers went from "a mortgage on every citizen's back" to "foreclosures du jour." Technology can do a better job (85% of the grunt work) but common sense (the 15%) is mandatory before the rubber stamp
29. Every bank has a tech cynic. Don't bury him, just ask him to write a confidential report about his objections for the bosses. React based on the content. Read and burn all but one copy.
30. The best tech vendors don't say yes to every request. Pay attention to why some say no.
31. Don't expect too much people support from your vendor. Show you can do your own support work, and you'll get the vendor's respect, and ironically, more help.
32. Keep a tidy notebook of all events with vendors. Bank examiners love real-world communications. They'll spend hours reading them and the bank will get a "1."
33. Keep the CEO and CFO in tune with new IT issues. Tell it like it is and life will be easier in the long run. Side-stepping IT issues is like a tumor that keeps growing.
34. Bank tech vendors have plenty to sell, however, much of what is in their bag has already been sold. These days, there is no absolutely brand new technology to hit the market. That means sales people are not racking up lots of sales. In the past, "absolutely
brand new" were solutions such as ATMs, Remote Job Entry to Host, Internet Banking, Digital Imaging, Remote Deposit, CRM, PC-based apps, In-House Processing, Business Intelligence, Platform Automation, Electronic Payments, Smartphone and Tablet apps, and a
plethora of Data Security capabilities, Compliance Solutions and Analytics .
35. In the past five years, bank tech vendor revenue has been growing at an average rate of 4% per annum. In previous years, it was 17%. Exhibit 109 in "Automation in Banking - 2013" lists 35 solutions that Fiserv, Jack Henry and D+H sold under the heading
of "popular products purchased by banks and credit unions." Good ancillary apps, but not big ticket items, thus the paltry revenue gains.
36. Bankers love technology. They just don't want to pay for it. In past decades, it was, "Gotta have it." Now it takes a thorough ROI process to spend any dollars on tech. If done correctly, ROI is not a license to buy. It may be a "Come back with a
better analysis." In 1970, it took three return visits before I could spend $55k to do a prototype ATM project. But the seed grew into a nice shade tree. In 2004, Fleet ranked #9 in number of ATMs.
37. I am very picky about the word, "integration" as it applies to bank systems. When it first appeared on the scene, I knew that Don Dillon and Dale Jensen designed a system that was like a hard-wired circuit board. No changes permitted. That attitude
changed in time where customers could run on IBM computers instead of the standard Burroughs, and they could even interface a foreign vendor ancillary solution. That destroyed the integration, in my opinion. So I don't believe integration is a 100% thing
anywhere. But it should be at least 85% among key applications.
38. Following are my reasons for integration:
• Uniformity of application rules. Learn them once, apply them everywhere.
• Ease of software maintenance by the vendor
• Eliminate interface maintenance
• Eliminate redundancy of maintaining foreign ancillaries
• Eliminate proprietary data silos
• Getting to know the customer
• Common updates made one time, feed all relationships automatically
• Ad hoc retrieval is easier and accurate
• Financial accounts tell the same status everywhere
39. Integration is like invisible weave. Interface is like a patch. Which would you prefer in your Sunday best?
40. IT, as an operating expense, is different by size of bank. Small banks spend 6% of operating expense. Mid-tier banks spend 12%, and top-tier banks spend 20%. Those are industry averages. The best way to achieve precision is for each bank CFO to create
a spread sheet of the 128 expense items by capturing each actual expense from original sources. The General Ledger is not the best original source because it has been contaminated by people who misidentify vouchers to hide some IT costs.
41. Vendors would be wise to invest in what they call a partnership with customers. Train the customers and it will return the investment in fewer calls for customer support. Good customer support is very costly because the highest paid people provide it.
42. The first step for each bank executive in assessing the power of technology is to know what it costs and why. They don't. The reason I love Campbell's vegetable soup is a nerdy cost analyst, along with a market-savvy product specialist, changed the frequency
distribution from potato pieces to carrot pieces. Carrots cost more than potatoes, but Campbell's knew that their customers would pay more.
43. Every division manager of a bank should know how much of the bank's total IT cost is charged to his budget. Either they don't know, or they don't know how to challenge it. There are hidden gremlins in every IT budget because the CIO likes them, but they
serve no useful purpose to the guys who pay for them. So the CIO should pay for his share of the total.
44. We hear about all the benefits (and cost savings to the banks) from mobile banking but there will be a growing increase in security costs to the banks. Any time a bank expands and facilitates access to customer accounts, the interloper activity will grow
exponentially. Banks should be paying more attention to data security than to data access right now. Down the road, when all the "doors" are secure, then it will be OK to make access friendlier.
45. There are all kinds of simple measures to determine value. For example, "on a scale of 1 to 10." One that is clearly missing is a performance scale for IT. I have one but I can't reduce it to a number either because I'm not tricky or smart enough.
I need lots of words to tell my clients what's wrong with their system. And that's why 321 hired me. Some things about IT just aren't easy.
46. Bank examiners are a very valuable resource. Listen to them. Answer them. Negotiate with them. Work out a conclusion with them. But don't ever ignore them.
47. There isn't a system in the world that can calculate a numeric value for any bank customer, retail or commercial. In my opinion, there should be. Where would lending be today if there wasn't a credit score? Just make sure it doesn't become a profiling
thing. After 2008, Congress might be for it.
48. Bankers rely on several ratios for peer comparisons. One is called "efficiency ratio." How much a bank spends on overhead to produce one dollar of revenue. The lower the percent, the greater the efficiency. IT efficiency appears to live on sacred turf.
I've never seen a ratio. Who will be the first to build it?
49. Processing transactions is by far the single biggest task of an outsource company. But it's peripheral tasks that they perform well and can save money for the bank. It's the same algorithm as a slice of the pie is a lot more cost effective than paying
for the whole pie, especially if you can't eat the whole pie.
50. Annual audits with the bank's primary tech provider should not include mojitos. Make it a duel where no one gets killed.
I'm done, what's on your list?