Every provider ranking on LightCompanies comes from one weighted formula applied the same way to every company: five factors, scored 0 to 100, combined into a single number you can audit yourself. No editorial thumb on the scale. No pay-to-rank. The same lens that produces a 41 for one retail electricity provider (REP) produces an 88 for another, and the inputs behind both numbers are public records you can pull on your own.
This page is the hub for how that works. If you want the short version: track record first, opinion second, and the opinion has to show its math. The longer version is below.
The five-factor lens
LightCompanies scores every provider on the same five factors, weighted by how much each one actually predicts whether you will regret your choice in month seven.
- Rate transparency (25%). Does the advertised rate match the rate you pay at your actual usage? Texas providers quote a single cents-per-kWh number tied to a specific monthly usage (usually 1,000 kWh). Pull the Electricity Facts Label and the price at 500 kWh and 2,000 kWh often tells a different story. Bill credits that only trigger at one usage band are the most common trick. Transparency scores high when the per-kWh cost stays flat across usage tiers and low when the rate doubles the moment you fall outside the sweet spot.
- Billing reliability (25%). Estimated reads, surprise charges, deposit handling, refund speed. This is where most PUCT complaints land, which is why it carries the same weight as price honesty.
- Customer service responsiveness (20%). Hold times, resolution rates, whether a billing dispute gets fixed or gets you transferred four times.
- Plan flexibility (15%). Contract lengths, early termination fees, and whether leaving costs you a flat $150 or a punitive per-month-remaining penalty.
- Renewable mix (15%). The percentage of the plan sourced from wind and solar, verified against the Facts Label rather than the marketing page.
Transparency and billing carry half the total weight between them because they are the two failures that cost real money and are hardest to spot before you sign. Renewable mix matters, but it sits at 15% because it rarely produces a bill you did not expect.
Where the data comes from
Three public sources do most of the work.
PUCT complaint data. The Public Utility Commission of Texas logs formal complaints against every REP operating in the state. This is the single most useful signal because it is filed by customers who were angry enough to escalate past the provider’s own support line. PUCT publishes these as quarterly snapshots, not a live feed, so any complaint figure on this site is the most recent quarter available and is labeled as such. When a number is two quarters old because the Commission has not posted newer data, the page says so.
BBB ratings. The Better Business Bureau grades providers A+ through F and publishes the complaint volume and resolution rate behind each grade. The grade itself is less useful than the raw numbers under it (more on that below).
Years in market. A provider operating in Texas since 2002 has survived multiple price spikes, a near-grid-collapse in February 2021, and several rounds of consolidation. A provider that launched eighteen months ago has not been tested. Tenure does not earn points directly, but it changes how much weight thin complaint data deserves. A six-month-old company with two complaints is not safer than a twenty-year-old company with two hundred. It is just younger.
How PUCT complaints get normalized
Raw complaint counts lie. A provider with 800,000 customers will always generate more complaints than one with 100,000, and ranking by raw count would punish scale and reward obscurity.
So every complaint figure gets converted to complaints per 100,000 customers before it touches a score. The arithmetic:
Provider A: 240 complaints, 800,000 customers. That is 240 / 800,000 = 0.0003, or 30 complaints per 100,000.
Provider B: 60 complaints, 100,000 customers. That is 60 / 100,000 = 0.0006, or 60 per 100,000.
By raw count, Provider B looks four times cleaner. Normalized, Provider B generates twice the complaint rate of Provider A. The ranking uses the normalized figure every time. This is why a large, well-known provider sometimes outscores a small one you have never heard of: the small one just has fewer customers to file against, not fewer reasons to.
When customer counts are not published (PUCT does not require every provider to disclose them at the same cadence), the page flags the complaint figure as a raw count and declines to compare it directly against normalized scores. A number you cannot normalize is a number you cannot rank fairly, and pretending otherwise would defeat the point.
BBB ratings, used carefully
The BBB letter grade is the weakest of the three sources, and it gets used as a tiebreaker rather than a pillar. Two reasons.
First, the grade is partly a function of whether the business pays for accreditation and how fast it responds to BBB-mediated complaints, which measures responsiveness more than it measures whether customers were treated well. A provider can hold an A+ while resolving complaints by simply closing them.
Second, the grade compresses information. An A- could sit on top of 12 complaints or 1,200. So LightCompanies reads the numbers under the grade (total complaints filed, percentage resolved, how many recurred on the same issue) and scores those, not the letter. A provider’s BBB grade gets mentioned for context. It does not move the composite score much on its own.
How the five scores become one number
Each factor is scored 0 to 100. The composite is a straight weighted sum:
Composite = (Transparency × 0.25) + (Billing × 0.25) + (Service × 0.20) + (Flexibility × 0.15) + (Renewable × 0.15)
Work an example. A provider scores 70 on transparency, 60 on billing, 75 on service, 80 on flexibility, and 90 on renewable mix:
- 70 × 0.25 = 17.5
- 60 × 0.25 = 15.0
- 75 × 0.20 = 15.0
- 80 × 0.15 = 12.0
- 90 × 0.15 = 13.5
Total: 73.0.
That 73 is what shows up at the top of a provider profile. It is not a star rating dropped from the sky. It is five auditable inputs and five fixed weights. If two providers tie at 73, the breakdown shows where they differ: one might earn it on renewable mix and lose on billing, the other the reverse. LightCompanies rates a provider with strong billing and weak renewable mix above a provider with the inverse profile at the same composite, because a clean bill is the thing most shoppers actually need and the green premium is optional. The weights encode that judgment openly so you can disagree with it.
This is also why the site never leads with the number alone. A 73 means nothing until you see whether it was built on the factor you care about. A shopper who runs the AC at 2,000 kWh every summer should care most about transparency across usage tiers. A renter on a six-month lease should care most about flexibility and early termination terms. Same composite, different right answer.
What the score does not tell you
The formula has limits, and naming them is part of the methodology.
It does not price your specific plan. Two customers on the same provider at different usage levels pay different effective rates, and the actual cents-per-kWh comparison happens on ComparePower, not here. This site ranks the company. The plan math is a separate step.
It does not capture last week. PUCT data is quarterly, BBB updates on its own schedule, and a provider that fixed its billing system in March will not show improvement in a snapshot taken in January. When the data is stale, the page says how stale.
It does not predict a grid emergency. No score anticipated how providers would handle February 2021. Tenure is the only proxy, and it is a weak one.
And it does not weight your priorities. The 15% on renewable mix is LightCompanies’ call, not yours. The factor breakdown exists precisely so you can reweight in your head.
How to cross-check a ranking yourself
You do not have to trust the composite. Every input is public.
- Pull the PUCT complaint record. Search the provider on the Commission site and note the complaint count and the quarter. Divide by the provider’s reported customer base to get the per-100,000 rate, then compare it against the named comparison set on the profile.
- Read the BBB page, not the grade. Look at total complaints, resolution percentage, and how many cluster on billing versus service.
- Open the Electricity Facts Label. Check the per-kWh price at 500, 1,000, and 2,000 kWh. If the rate swings hard between bands, transparency is low regardless of the headline number.
- Check the contract terms. Find the early termination fee and confirm whether it is flat or scales by months remaining.
- Note the launch year. A short track record means thin data, and thin data means the score deserves less of your confidence, not more.
If your math lands somewhere different from the ranking, the breakdown will show you which factor drove the gap. That is the point of publishing the formula instead of a verdict. LightCompanies is not a referral shop, and when the numbers say avoid a provider, the page says avoid it. The cluster pages linked from here apply this same lens to individual providers and head-to-head comparisons, so the 73 you see on one profile means exactly what the 73 on another means.