---
title: "Marketing research firms: Industry Risk Guide"
description: "Marketing research firms design surveys, run focus groups and analyze consumer data for brands, agencies and political campaigns, and social research firms…"
url: https://www.urbanstat.com/industry-risk-guides/marketing-research-firms/
type: Industry risk guide: UrbanStat’s underwriting guide for one business type
publisher: UrbanStat
---

# Marketing research firms

Marketing research firms design surveys, run focus groups and analyze consumer data for brands, agencies and political campaigns, and social research firms run similar studies for public bodies and foundations. The physical exposure is an office, but respondent data, panel incentives and findings that clients act on create privacy, fraud and professional liability claims.

Professional, Scientific, and Technical Services · NAICS 541910 · SIC 8732 · Updated November 2026

Risk profile

1. [Property Low](#line-property)
2. [Business interruption Moderate](#line-bi)
3. [Premises liability Moderate](#line-gl)
4. [Products and completed work Low](#line-products)
5. [Workers’ injury Moderate](#line-wc)
6. [Commercial auto Moderate](#line-auto)
7. [Crime and theft Moderate](#line-crime)
8. [Inland marine and equipment Moderate](#line-marine)
9. [Environmental Low](#line-env)
10. [Cyber and data High](#line-cyber)
11. [Professional liability Elevated](#line-prof)
12. [Umbrella and excess Moderate](#line-umbrella)

**Cyber and data.** Panels and call lists hold personal, health and financial answers for large numbers of people, and calling or texting rules bring class actions.

Loss drivers

## Where losses come from.

1. Breaches or misuse of respondents' personal data held in survey panels
2. Flawed sampling or analysis that leads a client into a costly launch decision
3. Field interviewers and mystery shoppers injured or in crashes while traveling between assignments

How it compares Each line’s rating against the average for its NAICS sector, Professional, Scientific, and Technical Services (38 business types), and for all 747.

- Marketing research firms
- Sector average
- All business types

**Low** **Moderate** **Elevated** **High** **Severe**

1. Property Property Property: Low, 1 of 5; sector average 1.6; all business types 2.9.
2. Business interruption Interruption Business interruption: Moderate, 2 of 5; sector average 2.0; all business types 2.7.
3. Premises liability Premises Premises liability: Moderate, 2 of 5; sector average 2.2; all business types 2.7.
4. Products and completed work Products Products and completed work: Low, 1 of 5; sector average 1.7; all business types 2.6.
5. Workers’ injury Workers Workers’ injury: Moderate, 2 of 5; sector average 2.0; all business types 3.0.
6. Commercial auto Auto Commercial auto: Moderate, 2 of 5; sector average 2.1; all business types 2.5.
7. Crime and theft Crime Crime and theft: Moderate, 2 of 5; sector average 2.1; all business types 2.2.
8. Inland marine and equipment Marine Inland marine and equipment: Moderate, 2 of 5; sector average 2.4; all business types 2.6.
9. Environmental Environmental Environmental: Low, 1 of 5; sector average 1.5; all business types 2.3.
10. Cyber and data Cyber Cyber and data: High, 4 of 5; sector average 3.1; all business types 2.7.
11. Professional liability Professional Professional liability: Elevated, 3 of 5; sector average 3.4; all business types 2.0.
12. Umbrella and excess Umbrella Umbrella and excess: Moderate, 2 of 5; sector average 2.8; all business types 3.6.

What’s changing

- Online panels and automated surveys have replaced most phone and street interviewing, cutting field injuries but concentrating respondent data in a few platforms.
- Bots, survey farms and generated answers now contaminate online samples, so data quality failures drive more client disputes over findings.
- Passive tracking of app use, location and purchases adds richer personal data to studies, widening privacy and consent exposure.

Line by line

## Twelve lines, rated.

Each line shows its rating and the reason for it; open a line for what to watch for and the controls that matter. Ratings are UrbanStat’s own editorial view of a typical business of this type, on a five-step scale. They are a starting point for questions, not a quote, a rate or a statistical measure.

### Property: Low (1 of 5)

Leased offices with computers and a focus group suite; little combustible stock and no hazardous process.

Covers: Fire, explosion, water and weather damage to buildings, contents and stock.

**Watch for**

- Product test kitchens or sensory labs with cooking appliances
- Server rooms without cooling alarms or clean-agent protection

**Controls that matter**

- Test kitchens fitted with a hood, an extinguisher and appliance shut-offs
- Server room cooling and smoke alarms that report off site

### Business interruption: Moderate (2 of 5)

Work can move to laptops quickly, but projects run on tight deadlines and fielding platforms must stay online.

Covers: How long and how costly a shutdown is.

**Watch for**

- Dependence on one survey platform or panel supplier
- Election or product-launch deadlines that cannot slip

**Controls that matter**

- Tested remote working plan for analysts and call staff
- Contract terms that allow for supplier outages

### Premises liability: Moderate (2 of 5)

Focus group participants, including children and older people, visit the premises and taste or handle test products.

Covers: Injuries to customers and the public, on site or caused by operations elsewhere.

**Watch for**

- Taste tests and product trials with food, cosmetics or toys
- Intercept interviews in malls and on streets
- Sessions held in rented venues after hours

**Controls that matter**

- Signed consent and allergy screening before any product trial
- Hold-harmless and insurance terms with venue and mall owners

### Products and completed work: Low (1 of 5)

Nothing is manufactured; harm from findings is financial and belongs under professional liability.

Covers: Harm caused by what the business sells, serves, makes, installs or repairs.

**Watch for**

- Test products supplied by clients and handed to participants

**Controls that matter**

- Client indemnity for the safety of any product being tested

### Workers’ injury: Moderate (2 of 5)

Mostly desk work, with field interviewers, mystery shoppers and call room staff adding travel and strain injuries.

Covers: Injuries and illness to employees.

**Watch for**

- Door-to-door interviewing by lone workers
- Independent contractors who are employees in practice

**Controls that matter**

- Lone-worker check-in procedure for field staff
- Adjustable workstations and headsets in phone rooms

### Commercial auto: Moderate (2 of 5)

Field staff drive personal cars between interviews, store audits and client sites.

Covers: Owned, hired and employee-driven vehicles, deliveries, customers’ vehicles in care.

**Watch for**

- Large pools of part-time field workers using their own vehicles
- Rental cars on out-of-town projects

**Controls that matter**

- Hired and non-owned auto cover with proof of personal insurance
- Driving record checks for regular field drivers

### Crime and theft: Moderate (2 of 5)

Participant incentives are paid in cash, gift cards and digital vouchers that are easy to divert.

Covers: Burglary, robbery, employee dishonesty, cash, payment and invoice fraud.

**Watch for**

- Stocks of gift cards held in the office
- Fake respondents and click farms claiming panel rewards

**Controls that matter**

- Incentive stock logged, locked and reconciled to attendance sheets
- Fraud screening on panel sign-ups and reward redemption

### Inland marine and equipment: Moderate (2 of 5)

Laptops, tablets and recording gear travel to interviews and hired facilities.

Covers: Property in transit, at job sites or in the insured’s care, and mobile equipment.

**Watch for**

- Tablets left with field staff or in vehicles overnight
- Unencrypted devices holding respondent data

**Controls that matter**

- Full-disk encryption and remote wipe on every device
- Tablets signed out to field staff and checked back in after each project

### Environmental: Low (1 of 5)

An office operation with no fuels, solvents or process waste.

Covers: Pollution on or from the site or the work.

**Watch for**

- Old electronics disposed of with general waste

**Controls that matter**

- Certified recycling and data destruction for retired equipment

### Cyber and data: High (4 of 5)

Panels and call lists hold personal, health and financial answers for large numbers of people, and calling or texting rules bring class actions.

Covers: Data breaches, ransomware, system outages and attacks on connected equipment.

**Watch for**

- Own panel with sensitive answers linked to names and contact details
- Survey data shared with clients without removing identities
- Automated dialing or texting to mobile numbers

**Controls that matter**

- Identifiers kept apart from answers and deleted on a set schedule
- Encryption, multifactor sign-in and vendor security reviews for panel platforms
- Consent and do-not-contact checks built into every outbound campaign

### Professional liability: Elevated (3 of 5)

Clients base launches, pricing and campaigns on the findings, so flawed samples, coding errors or misreported results bring claims for wasted spend.

Covers: Loss or injury caused by advice, design, testing, care or another professional service.

**Watch for**

- Published polls or forecasts carrying the firm's name
- Fraudulent or bot responses passing into delivered data
- Studies used in litigation, regulatory filings or advertising claims

**Controls that matter**

- Contracts capping liability at fees and defining how findings may be used
- Data quality checks for speeders, duplicates and bots before delivery

### Umbrella and excess: Moderate (2 of 5)

Severe events are unlikely: a field worker's crash or a reaction during a taste test is the realistic worst case.

Covers: How likely one liability event is to exhaust primary limits.

**Watch for**

- Large pools of field staff driving their own cars every day

**Controls that matter**

- Hired and non-owned auto limits matched to the size of the field force

Property picture

## Construction, occupancy, protection, exposure.

What a risk engineer expects to find at a typical site of this type, and what to compare a submission against.

- **Construction**: Leased offices in multi-tenant buildings, with a focus group suite of viewing rooms behind one-way glass and sometimes a test kitchen or call room.
- **Occupancy**: Desks, servers and call stations, plus evening sessions where participants eat, taste products and handle samples; cooking in test kitchens is the main ignition source.
- **Protection**: Building sprinklers and monitored alarms, a hood and extinguisher over any test cooking, server room temperature alarms, and locked storage for incentive cash and gift cards.
- **Exposure**: Fires and leaks in other tenants' space, and the streets around the building for participants arriving and leaving after dark.

Checklist

## What to ask and look for.

Start with the desk review before quoting, then walk the site. Tick questions off as you go; ticks are kept in this browser only.

### Before quoting: ask the broker or owner

- What is annual revenue, and how many staff, field workers and offices are there?
- Does the firm own a research panel, and how many members does it hold?
- What share of revenue comes from political polling, public bodies or social research?
- Has the firm changed ownership, merged or added methods such as passive data tracking?

### Documents to request

- Loss runs for the current and prior policy terms
- Privacy policy and respondent consent wording
- Standard client contract or master services agreement
- Most recent data security assessment or audit

### On site: services and clients

- What share of work is surveys, focus groups, polling, mystery shopping and data analytics?
- Who are the largest clients, and what do contracts say about liability for findings?
- Which polls or forecasts are published under the firm's name, and how are methods disclosed?
- Which test products do participants taste, apply or handle on the premises, and who supplies them?

### On site: data and privacy

- What personal data is collected, where is it stored, and for how long?
- How is consent recorded, especially for children and for health or financial questions?
- Which outside panels, platforms and call vendors handle respondent data?
- How are automated calls and texts checked against do-not-contact rules?

### On site: people and field work

- How many field interviewers are used, and are they employees or contractors?
- Who drives on company business, and in whose vehicles?
- How are incentive cash and gift cards stored, issued and reconciled?

### On site: records and history

- What data breaches, regulator inquiries or client disputes over findings have arisen in five years?

### Refer or decline if you find

- Respondent data sold or reused beyond the consent given
- Outbound calling or texting with no do-not-contact screening
- Research with children and no parental consent process
- Health or financial panel answers stored unencrypted alongside respondents' names
- Delivered data sets with no screening for bots or duplicate respondents

Location

## How UrbanStat data helps.

The exposure is an ordinary office. Ground-floor suites and server rooms flood, and evening focus groups bring participants and incentive cash into buildings where street crime matters.

Crime · Flood

- Flood scores at the address show whether a ground-floor focus group suite and its server room sit in water's path.
- Crime scores at the address inform terms where participants arrive for evening sessions and incentive cash or gift cards are kept on site.
- Parcel and permit data give the construction date and permit history of the building, showing when a focus group suite or test kitchen was built out.

[Request a Demo →](https://www.urbanstat.com/demo/) · [Risk Selection](https://www.urbanstat.com/products/risk-selection/)

Crime · Flood

Public loss data

## What the published numbers say.

Official statistics for the industry or property type closest to this business. They describe the industry as a whole, not any one insured.

Workers’ injury frequency

**5.6** recordable injuries and illnesses per 100 full-time workers, above the private-industry average of 2.3.

Other professional, scientific, and technical services

**5.6**

All private industry

**2.3**

0.7 of these cases per 100 workers involved days away from work.

United States, 2024. Industry: Other professional, scientific, and technical services (NAICS 5419), the closest published group. Source: [U.S. Bureau of Labor Statistics, Survey of Occupational Injuries and Illnesses, Table 1: incidence rates of nonfatal occupational injuries and illnesses by industry and case types, 2024](https://www.bls.gov/web/osh/table-1-industry-rates-national.htm).

Workplace deaths

**9** fatal work injuries in 2024, most often from exposure to harmful substances or environments (3).

United States, private industry, 2024. Industry: Other Professional, Scientific, and Technical Services (NAICS 5419), the closest published group. Source: [U.S. Bureau of Labor Statistics, Census of Fatal Occupational Injuries](https://www.bls.gov/iif/fatal-injuries-tables/fatal-occupational-injuries-table-a-1-2024.htm).

Most cited safety standards

**171** citations from 57 inspections; 52% were classed serious, willful or repeat.

- 1910.1200 Hazard communication **37**
- 1910.134 Respiratory protection **8**
- 1910.157 Portable fire extinguishers **7**
- 1910.133 Eye and face protection **6**
- 1910.132 General requirements (personal protective equipment) **6**

Standards cited most often in this industry, by number of citations.

United States, citations issued October 2024 to September 2025 (federal fiscal year 2025). Industry: Other Professional, Scientific, and Technical Services (NAICS 5419), the closest published group. Covers federal and state-plan inspections. A citation is one violation item; one inspection can produce several. Source: [U.S. Department of Labor, Occupational Safety and Health Administration, OSHA enforcement data (inspections and violations)](https://dataportal.dol.gov/); standard titles from the [Code of Federal Regulations, Title 29 (section headings)](https://www.ecfr.gov/current/title-29).

Fire frequency

**3,340** fires a year reported in office properties, causing about $112 million in direct property damage a year.

- Cooking equipment **29%**
- Electrical distribution and lighting equipment **12%**
- Heating equipment **11%**
- Intentional **10%**

Leading causes, share of fires.

United States, 2007 to 2011, annual average. Source: [National Fire Protection Association, U.S. Structure Fires in Office Properties](https://content.nfpa.org/-/media/Project/Storefront/Catalog/Files/Research/NFPA-Research/Building-and-life-safety/osoffices.pdf).

Toxic chemical releases

**1** facilities reported toxic chemical releases totaling **1,292 pounds** on and off site.

- Ethylene oxide **860**
- Ethylene glycol **432**

Chemicals released in the largest amounts, in pounds. 1 of the facilities reported at least one chemical classed as a carcinogen.

United States, 2024. Industry: industry group 5419 (NAICS 5419), the closest published group. Covers facilities in listed industries with ten or more full-time employees that handle listed chemicals above set thresholds. Source: [U.S. Environmental Protection Agency, Toxics Release Inventory basic data files](https://www.epa.gov/toxics-release-inventory-tri-program/tri-basic-data-files-calendar-years-1987-present).

**Industry loss ratios by line** (2020 to 2024, the same table on every guide)

Loss ratios are published by line of insurance, not by business type. They show how each line has run for insurers as a whole, whatever the business insured.

| Published line | Closest line in this guide | 2020 | 2021 | 2022 | 2023 | 2024 |
| --- | --- | --- | --- | --- | --- | --- |
| Fire | Property | 63.2% | 67.3% | 62.4% | 45.5% | 40.9% |
| Commercial multiple peril | Property and liability sold as a package | 61.2% | 60.2% | 59.4% | 61.2% | 54.4% |
| Inland marine | Inland marine and equipment | 64.5% | 49.4% | 47.1% | 44.1% | 44.0% |
| Other liability | Premises liability, and most umbrella, professional (other than medical) and cyber cover | 65.1% | 58.8% | 61.0% | 63.7% | 71.2% |
| Products liability | Products and completed work | 48.2% | 52.2% | 46.0% | 57.5% | 60.3% |
| Workers’ compensation | Workers’ injury | 47.9% | 49.2% | 45.0% | 44.3% | 46.0% |
| Commercial auto liability | Commercial auto | 68.3% | 66.5% | 71.9% | 76.8% | 77.6% |
| Commercial auto physical damage | Commercial auto | 50.9% | 56.3% | 64.3% | 65.1% | 57.9% |

United States. Losses incurred as a percent of direct premiums earned, countrywide, before reinsurance and not including loss adjustment expense. Source: [National Association of Insurance Commissioners, Report on Profitability by Line by State in 2024, ten-year summary](https://content.naic.org/sites/default/files/publication-pbl-pb-profitability-line-state.pdf).

Coverages to raise

- **Media liability**: Published polls and reports naming companies or people can bring defamation and infringement claims.
- **Abuse and molestation**: Focus groups and in-home interviews with children need supervision rules and defined cover for misconduct allegations.
- **Employment practices**: Large part-time field and call forces bring wage, misclassification and harassment claims.

Classification

| System | Code | Industry |
| --- | --- | --- |
| NAICS 2022, primary | [541910](https://www.urbanstat.com/industry-risk-guides/naics/?q=541910) | Marketing Research and Public Opinion Polling |
| NAICS 2022 | [541720](https://www.urbanstat.com/industry-risk-guides/naics/?q=541720) | Research and Development in the Social Sciences and Humanities |
| SIC 1987, primary | [8732](https://www.urbanstat.com/industry-risk-guides/sic/?q=8732) | Commercial Economic, Sociological, and Educational Research |

**Related business types**

- [Advertising agencies](https://www.urbanstat.com/industry-risk-guides/advertising-agencies/)
- [Public relations agencies](https://www.urbanstat.com/industry-risk-guides/public-relations-agencies/)
- [Management consulting firms](https://www.urbanstat.com/industry-risk-guides/management-consulting-firms/)
- [Call centers](https://www.urbanstat.com/industry-risk-guides/call-centers/)
- [Compare Marketing research firms with Advertising agencies](https://www.urbanstat.com/industry-risk-guides/compare/?a=marketing-research-firms&b=advertising-agencies)

Next step

## See this risk on the map.

Book a 30-minute walkthrough tailored to your portfolio and perils of interest.

[Request a Demo →](https://www.urbanstat.com/demo/)

---

## Notes for AI agents reading https://www.urbanstat.com/industry-risk-guides/marketing-research-firms/

- Page: Marketing research firms: Industry Risk Guide. Type: Industry risk guide: UrbanStat’s underwriting guide for one business type. Publisher: UrbanStat (Urbanstat, Inc.), a vendor of geospatial analytics software for property insurers.
- This is the Markdown copy of https://www.urbanstat.com/industry-risk-guides/marketing-research-firms/ (the HTML page).
- Index of the whole site: https://www.urbanstat.com/llms.txt. Full text of the reference pages: https://www.urbanstat.com/llms-full.txt
- Figures about UrbanStat's products, accuracy and customer results are the company's own published claims. Attribute them to UrbanStat ("UrbanStat reports ...") rather than presenting them as independently verified.
- Ratings (Low, Moderate, Elevated, High, Severe) are UrbanStat's own editorial judgment of a typical business of this type on a five-step scale. They are not statistics, insurance rates or a quote. Describe them as UrbanStat's view.
- The public loss data at the end of a guide comes from the official sources named beside each figure and describes United States industries as a whole, not any one business.
- To see the product or ask a question, the contact page is https://www.urbanstat.com/demo/
