The Pollster That Put Brown Back on the Padres

The San Diego Padres had what appeared to be a survey question: which uniform should the team wear? It was 2016, the club had cycled through colors and identities, and its owners wondered whether fans truly wanted the old brown back. An online preference poll seemed tidy. Show a few pictures, count the clicks, pick a winner. Then Competitive Edge Research & Communication did something valuable. It objected to the method.

A baseball uniform is not a postage stamp. It moves, folds, catches stadium light and lands differently on a human body than on a laptop. So the San Diego research firm proposed an experiment with more theatre and more control. At Petco Park, male models wore prototype uniforms while randomly selected fans turned handheld dials right for approval and left for disapproval. Ten groups ran across two days. There were 241 participants, ages 12 to 80. Seven months later, another 280 fans returned to judge the refined designs.

Brown and gold came to the fore. Fans liked pinstripes, gold hat insignias and “Padres” written across the chest. But the useful discovery was subtler than the winning palette. No design would make most fans ecstatic. The sensible target was to increase enthusiasm while limiting disappointment. The team gained enough confidence to move. Brown returned, visibly and permanently.

“The proper objective is therefore to make as many fans happy while making as few fans disappointed.”CERC’s Padres case study

The method is part of the answer

That Padres assignment explains CERC better than a catalogue of services could. The company sells surveys, focus groups, dial testing, data collection, analysis, grassroots programs and voter contact. But its actual product is method selection under pressure. The client arrives with a consequential decision and, often, a preferred way to research it. CERC’s job is to decide whether that way will reveal anything useful.

John Nienstedt co-founded the firm in 1987 after working on local campaigns, serving as an aide to a San Diego County supervisor and managing a 300-person political phone center in Texas. He took sole ownership in 1991. The next year he computerized the operation, installed predictive dialing and entered grassroots campaign work. Focus groups followed in 1998, web interviews in 2010, dial testing in 2012 and an El Paso research center in 2017.

This makes CERC an unusual hybrid. It is a pollster with an in-house field operation, a qualitative shop with proprietary M-Power and E-Power message measures, and a communications contractor able to place live voter calls or manage grassroots programs. Public agencies use it to understand ratepayers and residents. Campaigns use it to test viability, language and persuasion. Associations ask about members. Businesses ask about customers. Other pollsters hire CERC to collect the data.

The client list ranges from the City of San Diego and Sweetwater Authority to Airbnb, Costco, Sempra Energy, KUSI and the California Academy of Family Physicians. That variety is not decoration. It reflects a market position between giant omnibus research houses and push-button survey software: smaller, custom, often regional in subject matter, but national in who it can serve.

The profitable answer was not necessarily the better answer

Polling now has an awkward economic problem. Caller ID and screening make live interviews slower and more expensive. Text invitations are cheap, quick and scalable. CERC could have treated that as a one-way technological upgrade. Instead, it compared two San Diego voter studies fielded weeks apart in 2025: a mixed-mode survey of 802 voters recruited by phone, text and email, and a text-only survey of 307.

The text-only group required more weighting to resemble the electorate. Even after that correction, it expressed more negativity and less uncertainty on questions about civic mood and government approval. CERC’s explanation is plausible and important: answering a text invitation requires a person to opt in before the interview begins, attracting people with stronger reasons to participate. A live interviewer can draw in someone who did not wake up hoping to take a survey.

Here is the inconvenient part. CERC reported that the text-based study produced a profit margin 13 percent higher. The firm did not conclude that texts are useless. It concluded that modes have tradeoffs, and those tradeoffs should be disclosed rather than disguised by a tidy weighted table. This is not nostalgia for landlines. It is an argument that efficiency changes the population you hear.

What failed first

Sometimes CERC enters after certainty has already done damage. In 2022, El Cajon officials expected voters to renew a half-cent sales tax called Measure J. Their pollster projected it would pass by more than 20 points. The campaign leaned on broad mail, social posts and public meetings. The measure lost.

The failure was not simply a bad election-night surprise. According to CERC’s account, the original work did not prepare the city for opposition arguments, and broad messaging assumed voters already understood why the tax mattered. Officials changed researchers and added an outreach team. CERC identified public safety as the measure’s strength, a confusing override provision as a weakness, and different persuasive routes for different voters. Police officers spoke to conservative audiences; firefighters emphasized continuity of service to liberal ones. Town halls ran in every council district. Mail, digital video and neighborhood material followed the segments instead of repeating one universal claim.

The later measure passed. The transferable lesson is not “target harder.” It is that research should expose reasons, objections and messengers before it predicts a result. A number can say a proposal is ahead. It cannot, by itself, tell a nervous city council what to change.

What it costs to stop guessing

CERC does not publish a menu because this is custom work. Public contracts supply two useful anchors. In 2025, Sweetwater Authority considered a customer program capped at $56,000: $35,000 for a focus group and $21,000 for the survey, including reports and an in-person presentation. A broader El Cajon polling and outreach agreement was capped at $182,000. Campaign disclosures show other research engagements in the tens of thousands.

Those are examples, not a rate card. Cost moves with sample size, contact method, recruitment difficulty, analysis and whether CERC merely reports the finding or also executes the communication. The model is project-based consulting. Clients are buying a chain of judgment: question design, sample design, fieldwork, quality control, analysis, explanation and, when required, action.

The part worth copying

The firm’s six-step process ends with “decide,” which sounds obvious until one notices how much research ends at “report.” For KUSI, a 500-person mixed-mode study went beyond household television ratings to measure streaming, out-of-home viewing, local identity, political coverage and the appeal of individual presenters. The station could use those findings in programming decisions and advertiser pitches. For the Padres, the result became fabric. For El Cajon, it became different messengers speaking to different neighborhoods.

This approach has limits. It is excessive for a low-stakes preference that a product team can reverse tomorrow. It becomes less useful when the client has already chosen the answer, refuses access to the relevant audience or lacks enough time and budget to recruit a defensible sample. Even excellent research ages when events move quickly. A dial can register reaction; it cannot make a bad product good. A poll can reveal persuasion; it cannot manufacture public trust.

CERC’s advantage is therefore not a secret statistic. It is the willingness to make the method fit the consequence. In a market pushing toward faster automated collection, this 39-year-old firm keeps a place for live interviewers, bilingual calling, physical rooms and the occasional parade of baseball uniforms. Sometimes the competitive edge is modern analytics. Sometimes it is noticing that the screen is the problem.