Lessons from the Ice Storm

Share via Twitter Share via Facebook Share via Linkedin Share via Reddit


So we had this ice storm you may have heard about. It was kind of a big deal, with hundreds of thousands of households without power and so on. But while not having power, heat or running water is what we in the business might term suboptimal, there’s always a silver lining to be found.

Quite literally, in some cases.

Besides that Big Picture rip off, however, there’s the opportunity to turn this storm into a learning opportunity. As was repeatedly stressed by Central Maine Power’s President Sarah Burns during her appearance on Maine’s NPR affiliate yesterday.

As a current CMP customer with some small expertise in the collaboration/social media space, and – as of Saturday – an expert in gasoline generators, I thought I’d post some public ideas on where and how new mechanisms could be employed alongside of or as a complement to the old to improve the process of information dissemination.

Because that was, to my mind, the real problem. The folks in Alfred that are still without power might beg to differ, of course, but the actual loss of power is like a sunk cost: it is what it is. Unless the entire power grid could be put underground, and lord knows this state could not afford that, you’re going to have outages. That’s a fact of life. It’s unfortunate, and the state and CMP need to do what they can to proactively minimize the damage – cutting back tree limbs and so forth – but there’s only so much you can do about an inch of ice.

While that’s outside of CMP’s control, however, how they communicate with customers is not. And while I give full credit to the CMP and other out of state crews who did magnificient work under trying circumstances, the communication process around their efforts was far less consistent.

Burns, in yesterday’s appearance, claimed that CMP’s position since Saturday had been consistent: the company was targeting a full restoration by midnight on Wednesday. Which is interesting, because in spite of the fact that I visited CMP’s website over the weekend, read the papers, listened to the radio and called CMP’s toll-free emergency service number, I never heard anything about Wednesday’s deadline. Nor had two of the other callers into the NPR program. Which illustrates the problem: communication.

CMP had information, many (all, in my experience) of those affected by the outage did not. This asymmetry is not, as I’ve heard it argued, due to the inevitable difficulties of predicting precisely when the power would be restored. No one expects that, particularly in the aftermath of a storm such as this one that left downed lines strewn all over the state like worms after a rain. None of the folks I spoke with were looking for restoration predictions accurate to the minute, the hour, or even – really – the day. But we were all curious as to the approximate timeframe: were we talking hours? days? weeks? what?

This information, apparently, was known as of Saturday, but no one I knew of had it. Nor, as nearly as I can determine, did the Portland Press Herald. Tracking the storm stories, I see no mention of a Wednesday date. So if we didn’t know, and the Press Herald didn’t know, who did?

Rather than beat this dead horse further, let’s simply agree that the communication process could stand for some improvement.

What would I do, were I working for CMP, to better communicate the status of restoration efforts going forward? Several things. Call it the shotgun approach: between downed power lines and uneven broadband and cellphone penetration, the key is reaching out through as many channels as possible. This ensures the latest information and data can make its way to the widest possible audience. You’ll miss people, of course, but that’s unavoidable: the trick is getting the word to as many as possible. Door to door, as we saw in this storm, is an adequate last resort, but it scales exceptionally poorly.

Particularly when the ground is icy.

Instead, each update – and if I were CMP, I would try to update hourly, or at least on a predictable, regularly scheduled basis every few hours, the following channels:

  • 211:
    The informational equivalent of emergency’s 911 needs better visibility to go with more up-to-date information. This should be an alternative direct line to the latest and greatest information with respect to the outage status.
  • IVR (AKA CMP’s Toll-Free):
    This was, sadly, basically unhelpful during the outage. Aside from the ability to report outages, the automated IVR system was as unable to connect callers to actual humans as it was to provide basic information on the status of the recovery efforts. The best you could get – at least as of Saturday – was an automated recording that would look up your account (by account or phone number) and provide you with a long list of the other towns experiencing outages. Which didn’t help much. Like 211, this should have included a predictably updated status report with a description of ongoing efforts as well as – as soon as it became available – expectations for outage lengths.
  • General Stores:
    One suggestion I haven’t heard yet was to assemble a roster of the general stores, community centers or even coffee shops and bars that serve as the de facto town hubs in rural towns like Georgetown. Obviously it’s not going to be possible to call each and every resident affected with the news, but even the most rural of communities has some sort of establishment that links residents to each other. Providing these, at least, with up to date information on progress, estimates and so on would go far to addressing the problem of information dissemination. Assuming, of course, that you’ve created a roster of such facilities in advance of the storm.
  • Radio/TV:
    I’m not quite sure how CMP worked with – or did not – the mainstream media outlets of both radio and TV during the storm, but the Channel 6 radio feed (88.7 FM, I think) was as uninformed of CMP’s progress as I was during the peak of the storm. They covered the total numbers of affected households, had comments from the Governor and so on, but nothing that I heard in the way of specific information for affected regions in terms of when they could expect their power to be returned. This seems like a major lost opportunity.
  • Twitter:
    What use would Twitter have for a state that is heavily rural and non-technical, you ask? Well, perhaps little. But there is nonetheless a sizable contingent of Twitter users here in Maine – all the major papers including the Press Herald – are on the service, as is – shockingly – the Maine state government. The latter’s storm related Twitter output? Notices about the state offices being closed; nothing about services available to residents, nothing about recovery information, and nothing about alternative information sources. I’m not going to argue that Twitter is the avenue to reach the bulk of the Maine population, but as it’s so low effort, why not leverage it?
  • Website:
    Much was made of the CMP website during the initial hours of the outage, but frankly I found it terribly uninformative. While it might not have been a high priority, given that most Mainers probably were not in a position to visit it without power, even had they been able to, it wouldn’t have told them much. As pictured, the primary information returned – just as with the IVR – was which towns were reporting outages. That’s not useless, but it doesn’t tell me much about what I can expect in my own town. In past years, when I’ve actually been able to speak with CMP customer service representatives during outages, they haven’t been able to promise exactly when my power would be returned, of course, but they could at least provide a timeframe. And during a storm like this, when it would be impossible for CMP to field all of the inbound calls, it would seem imperative that the organization utilize a scalable approach like the website to impart at least some of that kind of information.

Do the above channels carry risks and introduce new problems? Certainly, the “telephone” process of garbling messages being exhibit a.

But it’s hard to see how some information could be more damaging that the complete absence of information, which spawns rumors like the predictions of “two weeks” that I was hearing Friday night. No one is asking that utlities overpromise on restoration schedules that they may or may not be able to deliver to, but when they do have solid information to communicate to the public, it would be nice if they actually used all of the channels at their disposal for doing so.

Bad news being, in my view, always preferable to no news.

One comment

  1. this gets back to the whole push/pull discussion for getting information out.

    my 0.02€ on this; update the website with a link to a very light page (or change the home page for the duration) that details current status (locations without that resource and latest expected restoration) and the next update time.
    This “next update time” should at least initally have *big* time gaps (4-6 hours), along with the explanation that the time gap is big as everyone is busy mobilising and the situation is very fluid and they don’t want to overload it with status reports, as time passes, this time gap i would expect to naturally come down as more information is natuarally flowing.

    My preference for updateing a web-page is that:
    – stores/bars/cafes can, if they so desire, as a service publish this information
    – media services can re-broadcast it, without putting extra load on the utility company. The reduction of “load on the utility that this provides should not be under-estimated. Also those media outlets (that are clever) can advertise the fact that they are *not* currently trying to talk to the utility company so that more effort can be put into the fix
    – people can contact relatives/friends outside the problem area and ask them to check the website for them.

Leave a Reply

Your email address will not be published. Required fields are marked *