BORC.ee
EST. in the BalticsLV · LT · EE

Consumer
and media
research.

Baltic Online Research Company

A consumer and media research company serving organisations that need to understand audiences across Latvia, Lithuania and Estonia — three markets, three languages, one rigorous methodology.

3
Markets
4
Languages
~6M
Consumers
Start a projectExplore capabilities
(01) The remit

We turn consumer behaviour, media consumption and brand perception into decisions organisations can act on — in markets where most panels are too shallow, most samples too small, and most methodologies imported from elsewhere.

What we measure

Awareness and consideration funnels. Brand equity and image attributes. Price elasticity. Satisfaction and loyalty. Advertising recall and diagnostics. Media reach, frequency and attention. Purchase drivers and barriers. Segmentation and typology. Concept and product preference.

How we measure

Online quantitative (CAWI) as the backbone, supplemented with CATI for harder-to-reach audiences and CAPI for specific geographies or populations. Qualitative designs using focus groups, in-depth interviews, ethnography and asynchronous online communities. Analytical techniques from conjoint and MaxDiff through to latent class segmentation.

Where we work

Latvia, Lithuania and Estonia — the three Baltic EU member states. We field in Latvian, Lithuanian, Estonian and Russian, with English reporting. Cross-Baltic studies are designed to be genuinely comparable: shared instruments, harmonised coding frames and synchronous fieldwork windows.

(02) Services

Eight core capabilities, one integrated practice.

Every brief is different. Most projects combine two or three of the capabilities below — segmentation paired with brand tracking, price sensitivity alongside product testing, qualitative exploration feeding a quantitative validation. The strength of a full-service research partner is not breadth in isolation, it is the ability to design studies where one method sharpens the next.

01

Market Segmentation

Dividing a market into meaningful groups based on demographics, attitudes, needs, behaviours and lifestyle.

Segmentation answers the foundational question: who, inside a market, should we actually be talking to? A well-built segmentation gives every part of an organisation — product, brand, media, commercial — a shared vocabulary for describing the people they serve. The deliverable is not a slide deck; it is a typology, a set of personas grounded in real data, a sizing of each segment in the market, and an operational way to identify segment members in CRM data or survey flags.

Methods

  • —Latent class analysis
  • —K-means / hierarchical cluster
  • —Factor analysis on attitudinal batteries
  • —Typing tool development
  • —Segment sizing and profiling
02

Brand Research & Tracking

How well is the brand recognised, how is it perceived, how often is it chosen — tracked over time.

Brand tracking is the continuous measurement of a brand's health in the minds of its target audience. A tracker typically follows the brand funnel from awareness through consideration, preference and purchase, pairs it with an image grid of attributes relevant to the category, and reports equity as a composite metric. The value of a tracker compounds: the second wave shows movement, the fourth wave shows patterns, the eighth wave shows whether strategy is working.

Methods

  • —Spontaneous and aided awareness
  • —Brand funnel (awareness → consideration → preference → purchase)
  • —Image attribute grids
  • —Brand equity indices
  • —Wave-on-wave trend analysis
  • —Competitive benchmarking
03

Customer Satisfaction & Loyalty

NPS, CSAT, CES and loyalty drivers — linked to behaviour, not just declared intent.

Customer experience measurement has matured beyond the single satisfaction score. A proper CX programme distinguishes between relational measurement (how do you feel about us overall) and transactional measurement (how was that specific interaction). It pairs declared metrics — NPS, CSAT, CES — with driver analysis to understand what actually moves the score, and where possible links the survey data back to behavioural outcomes in the client's own systems.

Methods

  • —Net Promoter Score (NPS)
  • —Customer Satisfaction (CSAT)
  • —Customer Effort Score (CES)
  • —Relational and transactional programmes
  • —Driver analysis via regression
  • —Closed-loop feedback design
04

Product Testing

Testing products before launch or iterations of existing products, in controlled or natural conditions.

Product testing sits at the border between R&D and marketing. Done well, it protects organisations from launching things the market will not embrace and from killing things that would have worked with adjustment. Designs range from blind monadic tests where each respondent sees only one variant, through sequential monadic where respondents see several, to paired comparison. Fieldwork can happen in Central Location Tests (CLTs) with controlled conditions, or in Home-Use Tests (HUTs) where realism matters more than control.

Methods

  • —Monadic and sequential monadic designs
  • —Paired comparison
  • —Central Location Tests (CLT)
  • —Home-Use Tests (HUT)
  • —Blind vs. branded testing
  • —Concept, product and packaging iteration
05

Price Sensitivity Analysis

How much are buyers willing to pay, and what happens to demand when the price moves.

Pricing is one of the most leveraged decisions an organisation makes — and one of the most frequently made on gut feel. Quantitative pricing research replaces guesswork with measured willingness to pay, optimal price points, and demand curves at different price levels. Different tools answer different questions: the Van Westendorp Price Sensitivity Meter locates acceptable price ranges; Gabor-Granger measures demand at specific price points; conjoint analysis places price alongside other product attributes to reveal trade-offs.

Methods

  • —Van Westendorp Price Sensitivity Meter
  • —Gabor-Granger
  • —Choice-Based Conjoint (CBC)
  • —Brand-Price Trade-Off (BPTO)
  • —Price elasticity modelling
06

Qualitative Research

Focus groups and in-depth interviews — understanding the 'why' behind the numbers.

Qualitative research is the discipline of listening carefully to small numbers of people in enough depth to understand the motivations, language, frames and tensions that quantitative instruments only register as numbers. It is the right tool when the question is exploratory, when the category language is unfamiliar, when emotional drivers need unpacking, or when a concept needs to be stress-tested before going wide. Properly designed, it generates hypotheses that quantitative work can then confirm at scale.

Methods

  • —Focus groups (in-person and online)
  • —In-depth interviews (IDIs)
  • —Ethnographic and observational research
  • —Online bulletin-board communities
  • —Diary studies
  • —Shop-alongs and usability sessions
07

Quantitative Research

Surveys at scale — understanding 'how many' with statistical confidence.

Quantitative research is the measurement discipline: large samples, structured instruments, statistical inference. The purpose is not to replace human judgement but to size it — to tell decision-makers whether a pattern they have observed qualitatively exists at meaningful scale, or whether it is an artefact of the loudest voices in the room. A well-run quantitative study begins with sample design, proceeds through rigorous instrument development and cognitive testing, and ends with analysis that respects both significance and effect size.

Methods

  • —Representative online panel (CAWI)
  • —Telephone interviewing (CATI)
  • —Personal interviewing (CAPI)
  • —Mobile-first and push-to-web surveys
  • —River sampling for hard-to-reach audiences
  • —Weighting to census and demographic targets
08

Advertising Research

Pre-testing creative before launch; post-testing campaign impact after it runs.

Advertising research works on both sides of the launch line. Pre-testing — typically done on animatics, key frames, or finished-film in low resolution — catches creative that will not land before the media spend is committed. Post-testing measures the campaign in market: has it been seen, has it been remembered correctly, has it moved the attitudes and behaviours it was briefed against. Diagnostics go further, identifying which creative element or message is doing the work and which is being ignored.

Methods

  • —Pre-test: concept, animatic, finished creative
  • —Post-test: recall, recognition, brand linkage
  • —Campaign impact on awareness and consideration
  • —Creative diagnostics
  • —Multi-wave in-market tracking
(03) Quantitative

Quantitative methodologies.

Mode choice is not neutral. The decision to field online, on the phone, or in person shapes who is reached, what can be shown, how long the instrument can run, and how the data must be weighted. We design studies mode-first — picking the right field approach before the questionnaire is drafted.

Field modes

CAWI

Computer-Assisted Web Interviewing

Online self-completion surveys, delivered to a recruited panel or to a client's own customer list. CAWI is the default mode for most consumer research in the Baltics in 2026 — it is fast, cost-efficient, supports rich media and complex logic, and reaches the majority of the adult population. Quality is managed through pre-screening, trap questions, open-ended response coding, speeder and straightliner detection, and device fingerprinting.

Best for
Everyday consumer research, high-volume tracking, long instruments, media-rich stimuli.
CATI

Computer-Assisted Telephone Interviewing

Telephone interviews conducted by trained interviewers reading from a scripted instrument. CATI reaches audiences who are under-represented online — typically older demographics, certain rural populations, and some B2B audiences. Call centres use predictive diallers, recorded calls for quality assurance, and full supervisor monitoring. The mode is slower and more expensive than CAWI, but remains essential for nationally representative samples in certain target groups.

Best for
Older demographics, politically sensitive topics, B2B decision-makers, short-to-medium instruments.
CAPI

Computer-Assisted Personal Interviewing

Face-to-face interviews conducted on tablets, typically in respondents' homes or at controlled central locations. CAPI is the gold standard for nationally representative samples, for studies requiring physical product exposure, and for audiences where trust needs to be built before disclosure. The mode is costly and logistically intensive, but delivers data quality that other modes cannot match when stimuli must be controlled or observed behaviour must be captured.

Best for
Product testing with physical stimuli, nationally representative samples, populations without reliable internet access, studies needing observation.
MOBI

Mobile-first surveys

Instruments designed specifically for the mobile form factor, typically 3–7 minutes in length and using interaction patterns native to smartphone usage: sliders, swipes, image grids, short-form video. Mobile-first is particularly strong for in-the-moment measurement, location-triggered surveys, and audiences where the smartphone is the primary or only connected device.

Best for
Younger audiences, location or moment-triggered research, short tracking pulses, diary-style studies.

Analytical techniques

Choice-Based Conjoint (CBC)

Respondents make a series of choices between product profiles that vary across attributes. The analysis derives the utility respondents attach to each attribute level, enabling market simulation and 'what-if' pricing.

Adaptive Conjoint (ACA)

An interview-like conjoint design that adapts the questions shown to each respondent based on their previous answers. Efficient when many attributes need to be measured.

MaxDiff scaling

Respondents identify the most and least important (or appealing, or preferred) item from small sets. Produces a clean, ratio-level ranking free of scale-use bias — particularly strong for prioritising feature lists or message testing.

Van Westendorp Price Sensitivity Meter

Four questions about price thresholds (too cheap, cheap, expensive, too expensive) are combined into four intersecting curves that reveal an acceptable range and optimal price point.

Gabor-Granger

Respondents are shown a price and asked purchase intent, repeated at multiple prices. Produces a direct demand curve and a revenue-maximising price.

TURF analysis

Total Unduplicated Reach and Frequency. Used to build product ranges, flavour line-ups, or message portfolios that maximise cumulative reach across the target.

Key Driver Analysis

Regression-based identification of which attributes actually drive a dependent outcome — satisfaction, loyalty, intent. Separates what respondents claim is important from what statistically predicts behaviour.

Latent Class Segmentation

A model-based alternative to k-means clustering. Identifies latent groups whose answers to attitudinal batteries follow a distinct pattern, with probabilistic membership rather than hard assignment.

(04) Qualitative

Qualitative methodologies.

Qualitative design is about matching the method to the question. A group is not a small survey; an interview is not a long group. Each format reaches a different kind of understanding, and the craft of a qualitative practice is knowing which one the brief actually needs.

Focus groups

01

Moderated discussions with six to eight respondents, typically running 90 to 120 minutes. Well-suited to exploring shared category language, generating reactions to stimuli, and surfacing group dynamics around social or public topics. Conducted in person with a viewing facility, or online via video platforms with screen-sharing for stimulus presentation.

Duration
90–120 min
Setup
6–8 respondents

Mini groups / triads

02

Smaller groups of three to four respondents. Used when the category is sensitive, when individual voice needs more air-time than a full group allows, or when the client audience is specialised enough that a full group cannot be recruited comfortably.

Duration
75–90 min
Setup
3–4 respondents

In-depth interviews (IDIs)

03

One-on-one conversations, typically 45 to 90 minutes. The interview is where depth lives — a skilled interviewer can follow threads into territory that a group would not surface, and respondents disclose things they would not share in front of peers. Standard for B2B decision-maker research, for category experts, and for sensitive personal topics.

Duration
45–90 min
Setup
1 respondent

Ethnographic research

04

Researchers observe respondents in their natural environment — the home, the workplace, the store — and document behaviour in context. Ethnography is the right tool when declared behaviour and actual behaviour diverge, when context matters more than opinion, or when a category has unspoken rituals that respondents cannot easily articulate in an interview room.

Duration
2–6 hours
Setup
Observational

Online bulletin-board communities

05

Asynchronous online spaces where recruited respondents answer structured tasks over several days. Participants can upload photos, videos, and reflections, and the researcher can probe in between. Particularly strong for diary-style topics, for category immersion before a creative brief, and for international studies where a synchronous group is impractical.

Duration
3–10 days
Setup
20–40 participants

Diary studies

06

Respondents document an aspect of their life — a category use occasion, a media diet, a purchase journey — over a defined period. Diaries can be structured (daily prompts) or open, and can use mobile apps for in-the-moment capture of moments otherwise lost to recall bias.

Duration
7–30 days
Setup
15–40 participants

Shop-alongs & accompanied journeys

07

A researcher accompanies a respondent through a real task — a supermarket shop, a car-buying visit, a booking journey — observing decisions in the moment and probing the reasoning while memory is fresh. Particularly powerful for category and channel research.

Duration
1–3 hours
Setup
1 respondent

Usability and concept testing

08

Moderated sessions focused on a specific artefact — a product prototype, an app flow, a concept board, packaging. Tasks are observed and narrated aloud, with the researcher documenting friction points, misunderstandings, and the delta between designer intent and user perception.

Duration
45–75 min
Setup
1 respondent
Analytical approach

Qualitative analysis is not just quoting the loudest respondent. Transcripts are coded against a framework that evolves with the data; counter-evidence is sought as actively as confirming evidence; and findings are triangulated across sessions, researchers, and — where the design permits — across methods. Deliverables typically include thematic analysis, journey or ecosystem maps, need-state frameworks, verbatim libraries, and a video reel of key moments.

(05) Capabilities

Specialised capabilities.

Beyond the core service categories, these are the specialised practice areas where the research design matters most — where methodology choice has real consequences for the decisions the study is meant to inform.

01

Brand tracking architecture

A brand tracker is built around the brand funnel — spontaneous awareness, aided awareness, familiarity, consideration, preference, purchase, repeat — paired with an image grid of attributes specific to the category. Standard frameworks include Millward Brown's Brand Dynamics pyramid, the Ehrenberg-Bass approach focusing on mental and physical availability, and the Keller Customer-Based Brand Equity model. We design trackers to carry forward across waves: question wording is locked, sampling is held constant, and weighting schemes are documented so that movement between waves reflects reality, not methodology.

02

Segmentation methodology

A segmentation begins with the strategic question — what decisions is this segmentation going to support? — and works backwards from there. Needs-based and attitudinal segmentations travel best across categories and countries; behavioural and demographic segmentations are more operational but often less discriminating. Typical sample sizes for a robust segmentation are 1,500 to 3,000 per market. The output includes segment profiles, sizing, a typing tool (a short battery of questions that can assign any respondent to a segment), and activation guidance for each segment.

03

NPS and CX frameworks

The Net Promoter Score, at 11 points from 0 to 10 with a standard 'likelihood to recommend' wording, is the industry's most widely used loyalty metric. NPS is best treated as a tracking figure paired with open-ended why questions, driver analysis, and a closed-loop process that routes detractor feedback to the teams that can act on it. We also deploy CSAT (Customer Satisfaction) for transactional measurement and CES (Customer Effort Score) for operational friction — the three metrics answer different questions and are often most informative together.

04

Advertising pre- and post-test

Pre-testing catches the campaign before the media spend is committed. Animatics, key frames, and finished-film are tested against category benchmarks for attention, comprehension, likeability, and persuasion, with diagnostic questions identifying which scene, message, or brand cue is doing the work. Post-testing measures the campaign in market: proven recall, branded recall, message take-out, and — for multi-wave trackers — the campaign's effect on brand health metrics. Creative diagnostics, moment-by-moment attention curves, and multi-format testing round out the offer.

05

Product concept and optimisation

Product testing is staged across the development lifecycle. Early-stage concept screens reduce a long list to a shortlist; detailed concept tests compare finalists on purchase intent, uniqueness, believability, and fit with need; post-prototype product tests capture experience with the physical product in Central Location Tests or Home-Use Tests; and package testing isolates the role of packaging from the product itself. Designs choose between monadic (one variant per respondent, cleaner read, bigger sample) and sequential monadic or paired comparison (direct comparison, smaller sample, risk of order effects).

06

Media and advertising effectiveness

Media research goes beyond reach and frequency. We measure channel consumption by time of day and context, attention paid to different formats, trust in different sources, and — in the advertising effectiveness space — the causal link from media exposure to behavioural outcomes. For digital-native campaigns this involves browser-based exposure measurement and matched control designs; for traditional campaigns it involves survey-based exposure recall paired with in-market behavioural data.

07

Price sensitivity and demand modelling

Price research combines direct methods (Van Westendorp, Gabor-Granger) with indirect, trade-off methods (choice-based conjoint, brand-price trade-off). Direct methods are fast and well-understood; indirect methods reveal how price interacts with brand, features, and competition. The deliverable is usually a price simulator — an interactive model the client can use to test 'what if we price at X, what happens to share and revenue across scenarios'.

08

Youth, under-18 and sensitive audience research

Research with under-18 audiences, with respondents in vulnerable circumstances, or on sensitive topics is designed under elevated ethical standards: parental consent for minors, carefully constructed instruments that avoid harm, safeguarding protocols for interviewers, and clear data-minimisation in reporting. We follow the ICC/ESOMAR International Code principles on interviewing children and young people and the GDPR-aligned guidance for special category processing.

(06) Geography

The Baltics are not one market.

Latvia, Lithuania and Estonia are routinely bundled in planning decks and then surprised when the data comes back with three different shapes. The three markets share EU membership, Euro currency, Schengen access and a similar digital consumer baseline — but differ in language, media diet, consumption patterns, and the role of Russian-speaking audiences. Our default is to design for differences first, commonalities second.

Latvia

LV
Capital
Riga
Population
~1.9M
Languages
Latvian · Russian
EU since
2004
Currency
Euro

Latvia's media landscape is bilingual in practice, with distinct Latvian-language and Russian-language channels reaching different audiences. Riga is the commercial centre, accounting for a disproportionate share of consumption; regional differences between Riga, Vidzeme, Kurzeme, Zemgale and Latgale are often material enough to design for explicitly. Latvia adopted the Euro in 2014 and is part of the Schengen Area.

Sampling notes

Representative national sampling requires careful quota management on Latvian- and Russian-speaking populations, with oversampling in Riga where brand and media consumption differ measurably from the regions.

Lithuania

LT
Capital
Vilnius
Population
~2.9M
Languages
Lithuanian · Russian · Polish
EU since
2004
Currency
Euro

Lithuania is the largest Baltic market by population and consumer spend. The media landscape is primarily Lithuanian-language, with Russian- and Polish-speaking minorities concentrated in specific regions. Vilnius, Kaunas and Klaipėda form the three urban anchors, each with distinct economic and consumption profiles. Lithuania adopted the Euro in 2015.

Sampling notes

National samples are typically structured around the three largest cities plus a representative rural cross-section. Polish-speaking respondents in the southeast and Russian-speaking respondents in specific districts warrant consideration depending on the category.

Estonia

EE
Capital
Tallinn
Population
~1.4M
Languages
Estonian · Russian
EU since
2004
Currency
Euro

Estonia is the smallest Baltic market by population but has the highest digital penetration and the most mature e-commerce and digital services ecosystem of the three. The Russian-speaking population — concentrated in Tallinn and the north-east — is a distinct consumer segment with its own media diet. Estonia adopted the Euro in 2011 and is a world leader in digital government.

Sampling notes

Online panels reach the bulk of the adult Estonian population at quality. Russian-language fieldwork is typically designed as a parallel strand in Estonian national studies, with separate recruitment and fielding.

Cross-Baltic design.

Genuinely comparable

Cross-Baltic studies require more than translating a single instrument. Question wording, answer scales, response cultures, and scale-use tendencies all differ. We design comparability in from the start — shared instrument logic, harmonised coding frames, synchronous fielding windows, and joint analytical frameworks — so that differences in the data reflect differences in the markets, not differences in the method.

Language coverage

Studies are designed and reported in English, fielded in Latvian, Lithuanian, Estonian and Russian. Translation is handled via back-translation and cognitive testing for sensitive instruments rather than single-pass translation.

Sampling frames

Online panels provide the backbone for most consumer research. For nationally representative studies we blend sources and weight to census demographic targets (age, gender, region, education) with additional quotas on language group and urbanity where the category demands.

Regulatory alignment

All three markets operate under EU regulation — GDPR for data protection, EU Consumer Protection directives, the Digital Services Act. Research design, consent flows, data retention and cross-border transfer are designed for single compliance model across the three markets.

(07) Panel & data quality

Sample quality is the silent variable.

Two otherwise-identical studies can produce two different answers purely because of how the sample was recruited, profiled, incentivised and cleaned. Methodological choice usually dominates methodology discussions; panel quality sits underneath, quiet, and moves the numbers more than most buyers of research realise. These are the principles that govern how we run a panel and source the sample behind every study.

01

Recruitment

Panel recruitment is blended across source types — online advertising, partnership-based referral, and targeted recruitment for specialist profiles. No single source is allowed to dominate, since monoculture in recruitment produces monoculture in answers. Every new panellist goes through double opt-in consent, profile capture, and an initial calibration survey before being released to live fieldwork.

02

Profile depth

Panellists are profiled at sign-up on demographics, household composition, employment, and a set of category-level questions that enable targeted invitation rather than mass broadcast. Profile data is refreshed on a rolling basis — panellist life circumstances change, and a panel that never re-profiles drifts silently away from accuracy.

03

Quality controls at fielding

Every live survey runs a layered quality control stack: attention checks, trap questions with obvious right answers, speeder detection (respondents completing in less than a third of median time), straightliner detection (identical answers across a grid), open-end response quality review, device fingerprinting to catch multi-accounts, and IP geolocation to flag out-of-market respondents. Failures remove the respondent from the dataset before weighting.

04

Weighting and representativeness

Data is weighted to national census targets on age, gender, region and — where material — education and language group. We document weighting schemes transparently and, where effective sample size is materially reduced by weighting, we report it honestly rather than hiding it in a footnote. Weighting cannot fix a bad sample, and we resist the temptation to treat it as though it could.

05

Panel hygiene

Active panel size is managed. Panellists are rested between studies to avoid response fatigue, inactive panellists are pruned, and participation frequency is capped — both to protect data quality and to respect the panellist as a human being rather than a source of data. Panel incentives are benchmarked to local market norms, paid reliably, and sized not to induce professional-respondent behaviour.

06

Hard-to-reach audiences

Some audiences are under-represented on any online panel — older demographics, certain rural populations, low-income households, and some minority language groups. For studies requiring full national representativeness in those categories, we blend online panel with CATI, CAPI, or river sampling, and design the analysis to reconcile mode differences.

(08) Process

A study is seven stages, not one.

How a research project is staged matters almost as much as what is measured. These are the stages every BORC engagement moves through — compressed for small studies, expanded for large ones, but rarely skipped.

  1. 01

    Discovery & briefing

    Every study starts with a conversation about the decision it will inform. What choice is on the table? What does the organisation already know, and where does the knowledge stop? Who will consume the research, and in what form will they act on it? A good brief is half the research — we invest time here rather than rushing to fielding.

  2. 02

    Research design

    Design translates the decision into a research architecture: sample plan, mode choice, instrument structure, analytical approach, and reporting format. Where the brief can be answered by a single study we design one; where it cannot, we design a sequence — often a qualitative exploration feeding a quantitative validation, or a quantitative landscape feeding targeted qualitative depth.

  3. 03

    Instrument development

    Questionnaires and discussion guides are drafted against the hypotheses from the brief, reviewed with the client, and — for complex or sensitive instruments — cognitive-tested with a small number of target respondents before going live. Translation into Latvian, Lithuanian, Estonian and Russian is done with back-translation and local review, not single-pass machine translation.

  4. 04

    Fieldwork

    Live data collection, with daily monitoring. We watch for response distributions that diverge from expectation, for speeders and straightliners, for drop-out patterns that suggest a problematic question. Fieldwork is not set-and-forget: if something is going wrong, catching it in hour eight is a lot cheaper than catching it in week three.

  5. 05

    Data processing & analysis

    Cleaning, coding, weighting, analysis. Open-ended responses are coded to a frame developed against the actual data. Quantitative analysis pairs significance testing with effect sizes — a statistically significant but directionally tiny difference is usually not actionable. Analytical techniques (regression, factor, cluster, conjoint modelling) are run transparently, with assumptions documented.

  6. 06

    Reporting & interpretation

    The report is built around the decisions the study was commissioned to inform, not around the order the questions were asked. Findings are stated plainly, implications are drawn, and open questions — the places where the research could not reach a confident conclusion — are named honestly rather than buried. Deliverables range from full report decks to summary memos to interactive dashboards, depending on what the audience will actually use.

  7. 07

    Workshop & activation

    Research that stays in a deck delivers a fraction of its value. Where the engagement permits, we run an activation workshop: a working session with the people who will act on the findings, translating insight into decisions, priorities and next steps. The workshop is the moment at which the study converts from knowledge into behaviour inside the organisation.

(09) Standards & ethics

Research is a profession.

The difference between research and opinion is rigour, ethics and disclosure. These are the standards we operate under — published so the buyers of our work can hold us accountable to them.

ICC/ESOMAR International Code

The International Chamber of Commerce and ESOMAR Code on Market, Opinion and Social Research is the global ethical framework for the research profession. It governs respondent rights, duties of the researcher, relationships between researchers and clients, and the treatment of children and vulnerable respondents. BORC designs studies to be compliant with the Code in its current revision.

GDPR compliance

All fieldwork across Latvia, Lithuania and Estonia operates under EU General Data Protection Regulation. Lawful basis for processing, data minimisation, storage limitation, respondent rights (access, rectification, erasure), and cross-border transfer are addressed in standard operating procedures. Respondent data is separated from contact data wherever methodology permits, and personal data retention periods are defined and enforced.

Research ethics principles

Informed consent is explicit, not implied. Respondent participation is voluntary at every stage, and withdrawal at any point is honoured. Deception — where methodologically unavoidable, as in certain advertising pre-tests — is debriefed at session end. Sensitive topics are approached with appropriate safeguarding; vulnerable audiences (minors, respondents in distress) are protected by elevated standards.

Data security

Respondent-level data is held on EU-resident infrastructure, encrypted at rest and in transit. Access is role-based and logged. Sub-processors (panel partners, field agencies, transcription services) are contracted under data processing agreements aligned to GDPR Article 28. Data retention is capped by project, and end-of-project purges are documented.

Transparency

Every report discloses sample source, sample size, fielding dates, mode, weighting approach, and any limitations material to the interpretation of the findings. Methodology notes are not buried — they are part of the read. Where a finding depends on a methodological choice that reasonable researchers could disagree about, we say so.

Dual professional standard

BORC-led studies adhere both to the ICC/ESOMAR Code and to the norms of academic social research (informed consent documentation, reproducibility of analytical procedures, disclosure of conflicts of interest). Research is a profession, and a research buyer should expect the same standards from a commercial agency that they would expect from a university research unit.

(10) Questions

Frequently asked.


A simple single-market quantitative study — sample of 500 to 1,000, 10-minute instrument — typically runs 3 to 5 weeks from brief to final report. Cross-Baltic studies add roughly a week to accommodate translation, harmonised fielding windows, and comparative analysis. Qualitative studies run 4 to 8 weeks depending on recruitment difficulty and sample size. Segmentation and conjoint studies, which require more design and analytical work, typically run 6 to 10 weeks. Trackers run continuously, with wave-on-wave reporting cadence defined at programme kickoff.

Sample size is driven by the analytical cuts the study needs to support, not by a default number. For a national representative reading with headline results, 500 to 1,000 per market is typical. For sub-group analysis — segments, age cohorts, brand users — sample is sized so that the smallest cell of interest has a usable base of at least 80 to 100 respondents. Segmentation studies typically require 1,500 to 3,000 per market. B2B and specialist-audience studies may operate with much smaller samples where the population itself is small; in those cases the analytical approach adapts accordingly.

Yes. Russian-language fieldwork is a standard strand in Latvia and Estonia, where Russian-speaking populations are a significant share of the consumer base, and available in Lithuania where the category requires it. Russian-language instruments are designed with local linguistic review rather than generic Russian, and analysis reports the Russian-speaking strand separately where sample sizes permit.

Both. A significant share of research value comes from measurement repeated over time — brand trackers, customer experience programmes, media audience trackers, category trackers. We design trackers with wave-on-wave comparability as the first-order constraint: instrument wording is locked, sampling is held constant, weighting schemes are documented, and changes that need to be made between waves are flagged transparently rather than absorbed silently into the data.

Yes. Many studies combine primary research with data the client already holds — CRM records, transactional data, digital behavioural data, past survey responses. Where GDPR permits and respondents consent, survey data can be matched back to behavioural data to test whether declared intent actually predicts behaviour. This is particularly valuable in loyalty programme design, churn diagnostics, and channel effectiveness work.

Sensitive categories are handled with elevated design care. Instruments avoid leading or judgemental wording. Respondent anonymity is emphasised more explicitly than in neutral categories. For politically sensitive research we pay particular attention to social desirability bias — the tendency of respondents to give the answer they think is acceptable rather than the answer that reflects their actual view — and design indirect measures where appropriate. For health and finance categories we operate under GDPR special category processing rules and elevated data security standards.

Report format is chosen against how the organisation will consume the findings. Standard deliverables include full report decks (40 to 100 slides for major studies), executive summary memos (10 to 15 pages), dashboards for tracker programmes, verbatim libraries for qualitative work, video reels of key moments, and interactive simulators for conjoint and pricing studies. Most engagements include a presentation session with the research audience, and where the engagement warrants it, an activation workshop to translate findings into decisions.

Yes. B2B research in the Baltics operates under specific constraints — smaller universes, decision-making units with multiple stakeholders, harder-to-reach respondents, and smaller effective sample sizes. Designs rely more heavily on in-depth interviewing and on targeted CATI/CAPI recruitment than on online panels. Analytical interpretation leans on pattern-reading across smaller samples rather than statistical inference at scale.

Research pricing scales with sample size, fielding mode, instrument length, number of markets, analytical complexity, and reporting depth. We provide transparent line-item costing against every brief, so buyers can see what is driving cost and make informed trade-offs between scope and budget. Where a research question can be answered with a smaller or less complex design, we say so — resisting the agency habit of scope-expansion where it does not serve the decision the study is meant to inform.

Yes. A significant share of our work is delivered in partnership with global and European research agencies who need rigorous local execution in the Baltic markets. We operate comfortably as a fieldwork partner, as a co-designer, or as a full-service lead, and we are transparent about the role we are playing on any given engagement.

Our primary geographic focus is the three Baltic markets. For studies requiring broader European coverage we work in partnership with established research networks, leading the Baltic strand and coordinating cross-market harmonisation where our clients run the programme directly. The Baltics sit naturally alongside Nordic and Central-Eastern European strands in many pan-regional research designs.

Standard industry survey platforms for CAWI fielding, video conferencing and specialised online community platforms for qualitative, analytical software (SPSS, R, Python) for statistical analysis, specialised packages (Sawtooth) for conjoint and MaxDiff, and BI tools for tracker dashboards. Platform choice is driven by the demands of the study, not by a single internal preference.

(11) Engage

Start a
conversation.

A short brief is all we need to start. The question you want answered, the decision it will inform, the markets in scope, the timeline you are working to. A thirty-minute conversation on Friday turns into a design proposal by the following Wednesday.

New research briefs
borc@borc.ee
Partnerships
partners@borc.ee
Panel enquiries
panel@borc.ee
Tell us

The decision the research will inform.

The markets

Latvia, Lithuania, Estonia — one, two or all three.

The timing

When the decision needs to be made, not just when the report is due.

What you know

Prior research, hypotheses, constraints, prejudices — all useful.