After two weeks away from everything, I'd planned on writing this post Sunday night. Somehow, I got a bit distracted. The 2011 news cycle has been incredibly intense so far, and shows no signs of slowing down anytime soon.

In any case, there's been a lot of news over the past few years about the smart grid and how utilities and startups and VCs are working to make the grid more data-driven, flexible, automated, etc. Very often, the discussions about the timing and pace of smart-grid rollout are focused predominantly on utilities and major vendors of transmission and  distribution equipment, including smart meters.

But while that's a large and important aspect of the smart grid market, I think it's not going to be the pacesetter for this industry. Instead, smart grid features and applications will likely be driven from inside the meter, out into the grid.

Some years back, in my consulting days, I had the opportunity to work inside a large utility. I got to see how these kinds of decisions are made, and the limits such entities face in trying to quickly roll out anything new. In fact, I helped pull together a business case for a "smart meter" investment by the utility (really, just automated meter reading: this was a while back). The business case included a lot of "soft" cost savings -- those that didn't have a specifically identified cost savings level, but were still likely to drive down costs in areas like improved customer service, more accurate billing, better blackout detection, etc. The problem is, lacking a concrete projected savings level, these savings got heavily discounted by the utility decision-makers because they knew they would have a hard time defending those projected savings to their real customers -- namely, the Public Utility Commission, with which they needed to negotiate for any significant capex program. And the "hard" cost savings weren't any easier -- they basically boiled down to fewer meter-reading jobs required. Plus, the local communities and PUC weren't thrilled about the idea of downsizing of any kind, so that was not an easy "win" for the utility either. In the end, the savings were sufficient that the project would potentially have been warranted, all things being equal; all things not being equal, the utility ultimately passed on the opportunity.

Things have changed in the decade since then.  Many of the "soft" cost savings have more proof behind them by now and PUCs are more open to accepting them in rate case negotiations. Additional functionality from smart meters and the smart grid also bolster the argument. And the aging workforce at utilities means that they're increasingly worried about how they're possibly going to be able to manage their assets going forward, even more than they're worried about having to explain a downsizing due to more automation. As a result, smart grid technologies are doing better now, and will continue to do better over time.

But these same obstacles remain. Talking with one utility's managers recently, they told me quietly that they'd like to do more smart grid investments, but their PUCs won't let them. They estimated it would take two rate cases (at five years for each one) for the PUC to agree to any significant investment. They also said that the PUC was actually allergic to smart grid investments as a potentially risky investment, even going so far as to reject investments in "smart grid" equipment that the utility had been purchasing for 20 years, before the term even existed! Other examples of this include all the negative press that has surrounded Xcel's "Smart Grid City" and the cost and time overruns there, and the recent pushback in California around smart meters and data communications. These don't make me pessimistic for the long-term future of smart grid technologies; progress continues to march on.  But these examples underscore how slowly it will likely happen at the meter and in the T&D grid.

Meanwhile, inside the meter, things are happening quickly. The secret weapon of the smart grid is the "demand charge" -- the fee large utility customers pay the utility for their single biggest spike in demand each billing period.  It is billed separately from overall consumption for the month.  You can find a useful explanation here (PDF).  Demand charges have been in place for decades -- and for large, variable electricity users like manufacturers and even office buildings, they can be upwards of one-third of the monthly electricity bill (the other two-thirds being consumption).

This isn't really something that can be addressed by simple efficiency improvements, or by putting solar on the roof.  Efficiency improvements may not avoid the monthly spike.  And with intermittent power production from solar on the roof, the monthly spike might also be just as high as before, even while net consumption goes down. So demand charges are a major pain point for large commercial and industrial electricity consumers.  And to really address them requires automation -- inside the meter.  Thus, even in the absence of any utility smart grid program, or demand response or dynamic pricing program, such customers can see compelling economic reasons to invest in energy-automation systems for this purpose.  Of course, with such utility programs coming into effect in many regions, the economics get even better.  So we're seeing a rapid uptake of such inside-the-meter systems.

Furthermore, as electricity customers large and small see energy prices going up in the future, there's been a wave of interest by early-adopting building owners in better energy information and automation in general, to reduce the consumption portion of the electricity bill as well. Thus, there's been a rapid rise in the number of existing building services companies looking to get into building energy services (I talked about this in regards to EnerNOC in particular a while back), but they are very far from alone in getting into these services.  But whether it's through a service provider, or just self-managed, such energy automation and information is only enabled by a smart building energy management system, with a lot of "smart grid"-like attributes. So for instance, with our investment Powerit Solutions (as one example among others, I'm sure), we're seeing a lot of inbound interest from many of these service providers who are looking for tools that can simultaneously address demand charges, enable participation in demand response and other utility "smart grid" programs, and also enable the service provider to remotely manage efficiency programs at the customer site.

So while "Utility Time" means a pretty slow pace of roll-out of the smart grid on the utility side of the meter, inside the meter, it feels like it's "off to the races" right now. Lots of systems are being introduced to the market, early adopters are jumping on board, and major equipment vendors are getting involved. What will be critical is for a common set of standards to enable such systems and equipment to easily connect with each other, and also to the eventually automated grid.  But there too, with efforts like OpenADR underway, the groundwork is rapidly being laid.  

And you know what? Having "smart-grid-enabled" customers already in place makes it even easier for the utility to talk to their PUC about making investments in smart grid technology in the T&D network. This is why, when we start seeing massive smart grid rollouts for real, my guess is it will have been really driven from the inside out, regardless of where the smart grid pundits and writers may be focusing their attention today.