Custom vs Off-The-Shelf Solar Management Software: How to Choose

It's Usually The Tariff

For a site on a common tariff, packaged solar energy management software will probably do the job. Once tariffs, dispatch rules or financing get unusual, custom work starts paying off.

Say two sites run the same panels and the same battery model. Their results can still drift apart, and the usual suspect is the utility tariff (or, to be fair, the software's grip on it). One tool applied to every bill will nail a few sites and miss the rest.

So when does a boxed product run out of road? And at what point does a custom solar software project stop being a luxury and start being arithmetic? It all comes down to the tariff. The sales deck matters much less.

Where Custom Builds Earn Their Keep

A demand charge works a bit like a speeding ticket for power. The utility looks at the single biggest draw (in kW) during a billing period and charges for that peak, even if the rest of the month was quiet. Time-of-use (TOU) pricing, meanwhile, ties energy prices to the clock, so evening kilowatt-hours often cost more than overnight ones. Add export rates and incentives, and a neat savings estimate gets tangled quickly.

None of this is new to researchers. The U.S. National Renewable Energy Laboratory has been publishing about its REopt platform since at least 2014. Give it a site's tariff, incentives, net metering rules and grid interconnection limits, and its mixed integer linear program returns a system size, a net present value and a lowest-cost dispatch plan. The practical lesson for buyers: good storage dispatch optimization needs the real rate, not an average of rates.

Commercial projects run into the same thing. Custom builds in this space tend to model each site's tariff, decide when storage should charge, discharge or export, use machine learning to refine those decisions over time, and compare realized savings with the original projections. 

Tariffs don't sit still either. A utility may revise its rate schedule mid-contract, and a model that was built once and left alone slowly stops matching the bill.

Who feels this most? Three illustrative cases:

• A developer quoting projects across several utilities, each with its own tariff rules.
• An asset owner who needs a documented record of realized versus projected savings before a refinancing or sale.
• A commercial energy manager coordinating several storage assets at one site.

What Does Solar Energy Management Software Actually Do?

At its simplest, solar energy management software watches the system and acts on what it sees. Monitoring comes first: production, consumption, battery state of charge and faults. Forecasting comes next, with estimates of tomorrow's solar output and site load, the kind of work that predictive analytics software is built for. The third job is dispatch, which decides when stored energy should charge, discharge or be exported. Reporting closes the loop by comparing what happened with what was promised.

Data reaches the software from inverters, meters and battery controllers. Teams that run many sites often track those devices through IoT device management software, so a silent sensor does not corrupt a month of readings. Packaged platforms usually cover monitoring and reporting well. The middle two jobs, forecasting and dispatch, are where tariff detail starts to matter.

When Does Custom Solar Software Earn Its Keep?

A demand charge works a bit like a speeding ticket for power. The utility looks at the single biggest draw, measured in kW, during a billing period and charges for that peak, even if the rest of the month was quiet. Time-of-use (TOU) pricing, meanwhile, ties the price of energy to the clock, so evening kilowatt-hours often cost more than overnight ones. Add export rates, net metering rules, and incentives, and a neat savings estimate gets tangled quickly. 

None of this is new to researchers. The U.S. National Laboratory of the Rockies, known as NREL until late 2025, has published about its REopt platform since at least 2014. Give it a site's tariff, incentives, net metering rules, and grid interconnection limits, and its mixed-integer linear program returns a system size, a net present value, and a lowest-cost dispatch plan. The practical lesson for buyers is that good storage dispatch optimization needs the real rate, not an average of rates.

Commercial projects that buy solar energy management software run into the same wall. Custom builds in this space tend to model each site's tariff, decide when storage should charge, discharge or export, use machine learning to refine those decisions over time, and compare realized savings with the original projections. One example of that scope is worth a look. Tariffs do not sit still either. A utility may revise its rate schedule mid-contract, and a model that was built once and left alone slowly stops matching the bill.

Who feels this most? Three cases come up again and again. A developer quoting projects across several utilities lives with a different rulebook for each. An asset owner needs a documented record of realized versus projected savings before a refinancing or a sale, and that record often sits beside the rest of the portfolio in asset management software. A commercial energy manager coordinating several storage assets at one site cannot afford dispatch rules that fight each other.

Where Off-The-Shelf Tools Hold Up

Custom isn't automatically better. A packaged platform has one big advantage: it exists today. If a site runs on a single, simple tariff, uses common monitoring hardware and belongs to a small portfolio, a ready-made product can often cover it, usually with lower upfront spend and a quicker rollout. There's no prize for building something that's already for sale.

Money matters here, too. One vendor's published estimate puts the average custom solar management build between $400,000 and $1,000,000, depending on complexity. Treat that as a rough marker rather than a benchmark, since it comes from a single company. Next to a figure like that, a monthly subscription looks very comfortable, at least until the limits appear.

Those limits usually show up in three places. The tool simplifies the tariff. Its dispatch logic can't be adjusted. Or it won't connect to the monitoring, battery management, financing or billing systems already in use. Any one of them can change the math.

Questions That Settle The Choice

Before committing, buyers can put each candidate through a short checklist:

1. Does it model the actual tariffs, including demand charges and TOU windows, rather than a simplified version?
2. Can dispatch rules be tuned to the balance between bill savings and backup reserve?
3. Will it connect to existing monitoring, battery management and financing systems?
4. Can it reconcile billed savings against projections?
5. Will the portfolio span more utilities or regions over the next few years?

Mostly "yes" on the first four, and a packaged tool likely does the job. A cluster of "no" answers points toward pricing a custom build, or a hybrid in which a packaged core handles the basics and custom modules cover the gaps.

How Should You Test a Vendor Before Signing?

The most useful test is also the cheapest: run any solar energy management software candidate on real billing data before anything is signed. A projection that falls apart against an actual utility bill costs far more than a license ever will. Ask vendors to model one of your live sites instead of a polished demo, using twelve months of bills so every season and rate change shows up. Look at three things in the results: how the tool treated the monthly peak, how it handled exports, and whether its numbers reconcile month by month.

Then compare the tool's estimate with what the utility charged. If the gap is large, ask why, and note whether the vendor can explain it in plain terms. Growth deserves a thought as well, because software that handles five sites comfortably can struggle with fifty.

Pro-tip

Give every vendor the same site, the same twelve months of bills, and the same question: what would this tool have saved, and how does that compare with what the bills actually show? Identical test data makes the answers easy to compare.

Conclusion

Neither option wins by default. Packaged platforms make sense as a starting point for simple sites, and custom builds earn their budget when tariffs, dispatch rules, and reporting needs grow past the defaults. The most useful test is also the cheapest: run the candidate on real billing data before anything is signed. A projection that falls apart against an actual utility bill costs far more than a license ever will. Buyers can ask vendors to model one of their live sites instead of a polished demo, and growth deserves a thought as well, because software that handles five sites comfortably can struggle with fifty. Whichever route fits, may the savings on paper match the savings on the bill.