Last week I enjoyed my freedom from political correctness by ridiculing New Jersey for the state of its roads (D+ grade from the American Society of Civil Engineers), its correlatively low gas tax, and the consensus among its governor, legislature and citizens that tax increases are off the table.
Who to ridicule this week? I know … most of the private sector, because if you think this sort of behavior is limited to public governance, you aren’t paying attention to your own backyard.
For example …
I know of an insurance company that’s grown through acquisition to the point that it now has eleven functionally equivalent underwriting/policy administration systems. And no, it isn’t run as a holding company.
Whenever there’s a change to business logic, it has to make that change eleven different times in eleven different ways.
From all reports, the company’s IT department has become quite good at coordinating and implementing business logic changes. As it should, because when it comes to deciding what capabilities your organization needs it’s good to concentrate on what the business is likely to need from you.
The only choice that would be better would be to retire ten of the eleven systems.
Except that by just about every reasonable measure, the company in question is extraordinarily successful.
This is the sort of thing that keeps management consultants awake at night. If you’re metaphorically inclined, it’s as if, by policy, a state limited road maintenance to dumping asphalt into potholes and still became one of the nation’s primary transportation hubs.
But that isn’t what this week’s column is about. This week it’s about planning for the obvious and how organizational dynamics so easily prevents it.
Planning for the obvious first: If you add something to your fund of stuff, whether it’s a house or car for your household, or a machine, information system or facility if you’re a business, you’re going to have to spend money in the future to maintain it.
Otherwise your fund of stuff (from here on in, FoS) will deteriorate, losing its value or performance over time.
It’s a simple, inarguable equation: FoS increases, maintenance costs increase too, or else FoS value steadily decreases.
As a nation, during the Eisenhower administration we built our interstate highway system, and, in 1956, created the Highway Trust Fund, supported by a penny per gallon federal gas tax increase (to 3 cents, which, in case you care, is equivalent to 24 cents today, compared to the current federal gas tax rate of 18.4 cents).
Meanwhile, I’d bet good money (enough to pay five gallons worth of New Jersey gas tax) most KJR readers work in companies that, when they implement new information systems, don’t increase the IT budget by enough to cover the easily predicted need for ongoing maintenance.
Why not, given that any meteorologist would kill to be able to make predictions with this level of confidence?
Based on my exposure to and experience in quite a few businesses over the course of my career, it’s because of:
- ROI computation: Proposed projects are usually evaluated on their financials. Include the cost of maintenance and the financials look worse, making the business case less attractive. Better to conveniently forget about them.
- Tradition: This isn’t just the opening number for Fiddler on the Roof. If nobody else had to include maintenance costs in the past, why should my pet project be burdened with them now?
- No good deed going unpunished: We haven’t increased the IT budget to support maintenance yet, and yet IT has managed to maintain everything so far. What’s changed?
- Wishful thinking: Maintenance is a separate spending bucket. If IT needs to maintain a system, the maintenance will have to be cost-justified on its own merits.
- Baumal’s Cost Disease: Everyone else is expected to continuously improve. Instead of increasing IT’s budget, IT should continuously improve enough to cover the difference.
- Reality Distortion Fields: We can’t increase IT’s maintenance budget because we’re going to need this money to invest in new strategic initiatives.
- Distrust: If we increase IT’s budget by enough to cover maintenance of this system, how do we know IT will actually spend the money on this system? (Several correspondents from New Jersey explained their anti-gas-tax-increase position on this basis.)
- Siloes: If we increase IT’s budget by x to cover the cost of maintenance for someone else’s system, that will leave less on the table to cover the cost of the new systems and system enhancements I’m going to want.
Against these forces, the CIO is armed with nothing beyond logic and maybe a Gartner study or two.
It’s time to buy more asphalt.
Planned Obsolescence: the system will be replaced before maintenance is actually needed. Also how do we budget for adaptive maintenance to implement requirements we can’t currently anticipate and still call it maintenance? (We don’t.)
As for why the insurance company is so successful despite having redundant systems (past belt-and-suspenders and into OCD-level stockpiles of toilet paper kept at 24 rolls at a minimum): It’s likely because of what I believe you have drilled into our heads many many times: What matters most in an organization is people, and what matters most to people are relationships.
Insurance companies tend to pride themselves, or at least advertise themselves, on high-quality relationships with their customers. Let’s assume for this highly-successful instance, that that’s true. Perhaps then they also highly value relationships within their organization, including acquisitions. So when a new acquisition states that it is perfectly happy with its policy admin system, the new management doesn’t make it change. Perhaps they even retain the IT staff that know how to efficiently maintain and make updates to it.
I think you might have recently written about why so many mergers and acquisitions that look good on paper end up failing. I think that reason was the attempted merging of cultures (created by people and their relationships) that proved incompatible cost too much – “culture” isn’t highly visible on financial analysis, so wasn’t accounted for in the original M&A. But cultural clashes can lead to large impacts on finances.
So perhaps the insurance company is so successful because it values all the people and cultures and relationships it acquires much more highly than it values efficiencies of scale.
I think that’s the likely reason. I know you have talked about how optimizing for the whole means sub-optimizing for parts. Perhaps optimizing for the whole of an acquistion-based insurance company means sub-optimizing for the IT admin systems used.
That’s my people-brain analysis. My IT-brain analysis gives this guess: Maybe the company is reducing the number of policy admin systems, just veeerrryyy sloooowwwwlllyyy. Through attrition – the software provider has gone out of business/been acquired/stops maintaining it/costs become prohibitive (in-house built systems are subject to many of the same forces). Or the users get fed up with it, see that their counterparts elsewhere in the company have a better one and demand to switch.
Why would this attrition-based method work better than converting during the inevitable upheaval of being acquired? Because you don’t know in advance which of these systems are going to go down some negative road, or prove to be the most comfortable for users (hey, the competitive market at work! 🙂
I like this second reason for selling your CIO on why you’re asking IT to maintain so many redundant systems. But I suspect it’s the first reason – it’s the People & Relationships! – that is the driver.
A successful insurance company has to have its finances understood and well-managed. But probably even more importantly, it needs to understand and manage people and relationships – externally and internally.