You have three flavor directions, a packaging redesign, or a new SKU, and a stage-gate meeting on the calendar. The question isn’t whether to test the concept. It’s which CPG concept testing software gets you a defensible answer before that meeting, at a cost your innovation budget can actually absorb.
“CPG” and “FMCG” describe the same category of buyer here; UK and global teams searching for FMCG concept testing tools and US teams searching for consumer goods testing tools are solving the same problem, and every tool below serves both. If you need the same comparison outside the CPG/FMCG context, our broader concept testing software guide covers SaaS, agency, and startup use cases.
We evaluated each tool on five criteria that matter specifically for CPG concept work: validated accuracy, turnaround time, cost per concept, methodology depth (does it explain why, not just score), and fit for the CPG innovation funnel, from early idea screening to the final volumetric forecast before a launch is greenlit.
If you want the fuller picture of how each stage of that funnel works, our guide to concept testing methods covers the process end to end. Every tool below is judged against the same five criteria.
Concept testing tools for CPG brands, compared
| Tool | Best for | Key differentiator | Price |
|---|---|---|---|
| Articos | Early screening before a formal quant study | Only tool here with a published, peer-reviewed accuracy validation | $8–$20/study; $79–$199/mo |
| NielsenIQ BASES | Volumetric sales forecasting at the launch gate | 500,000+ forecasts delivered, MASB-audited model | Custom, project-based |
| Zappi | Automated testing at enterprise scale | Deep historical CPG normative database | Enterprise, not published |
| Suzy | Fast quant reads with a live qual layer | Owned always-on panel plus “Suzy Live” interviews | ~$99K/yr median (annual license) |
| quantilope | Conjoint, MaxDiff, and pricing tradeoffs | 15 automated methods, AI co-pilot “quinn” | From ~$22K/yr |
| Attest | Global reach on a flat, published rate | 150M+ consumers across 59 countries | Flat-rate, published tiers |
| Upsiide | Screening many concepts fast, mobile-native | Patented Market Simulator for share-of-choice | Custom, project or subscription |
What tools test product concepts before a CPG launch?
Seven categories of tool cover the CPG concept testing job, and most teams end up using two or three of them at different stages: a synthetic, AI-first tool like Articos for same-day early screening; a normed automated platform like Zappi, quantilope, or Upsiide for scaled quantitative reads; a flexible global panel like Attest or Suzy for broad reach or a qualitative layer; and BASES for the audited volumetric forecast at the final gate. The full breakdown, with pricing and honest limitations for each, is below.
1. Articos
What it is: Articos runs concept tests against AI-generated synthetic personas instead of a recruited panel, part of Articos’s broader AI user research platform, delivering a full research summary, purchase-intent read, and thematic breakdown in under 30 minutes.
Best for: Teams that need to screen five or ten early-stage directions this week, before deciding which two or three earn a real BASES or Zappi study. See the consumer goods use case for CPG-specific examples.
Strengths: Articos is the only tool on this list with a published, peer-reviewed validation study behind its accuracy claim: 86% recall against expert-published research findings across 46 studies in nine domains, benchmarked against Baymard Institute and Nielsen Norman Group research. A study costs $8 to $20 and returns results the same day, so a team can test a claims wall, a flavor concept, and a package redesign in the time a traditional panel takes to field.
Limitations: Articos doesn’t run recruited human panels or produce a MASB-audited volumetric sales forecast, so it isn’t a substitute for BASES or a comparable quant study at the final go/no-go gate. Treat it as the fast pre-filter that narrows the field before the expensive study, not the study itself.
We tested it:
We ran a real concept test in Articos: three directions for the same protein snack bar, each led by a different claim (high protein/low sugar, gut health/prebiotic fiber, clean label/no artificial ingredients), against nine synthetic shoppers built to match health-conscious grocery buyers, 28 to 45, who shop at Target or Whole Foods and read the nutrition panel before buying. The report came back with 8 themes across 12 questions and named high protein/low sugar the clear winner, because it was the claim shoppers could verify fastest by flipping to the back panel.
Every participant named protein as what got them to pick up the bar, and every participant said they’d check the back panel before trusting the claim. One synthetic shopper, Kayla Donnelly, put it this way: “20g protein will make me stop… But that only gets it into my hand. Then I flip it over and see what the tradeoff is.”
The gut health claim wasn’t rejected outright; participants reacted with curiosity, but all of them wanted specific proof (fiber source, actual amount) before trusting it, which is why the report flagged it as fragile rather than dead. The honest caveat came from the report itself, not from us: the winning claim “can collapse if sweeteners or ingredients feel engineered.” The report’s actual recommendation reflects that risk: lead with high protein/low sugar, back it with a clean-label secondary claim, and treat gut health as a claim worth a dedicated follow-up test rather than a launch bet.
Price: $8 to $20 per study on demand. Starter plan runs $79/month for 10 studies; Pro is $199/month for unlimited studies. See the full concept testing platform for setup details.
2. NielsenIQ BASES
What it is: BASES is the industry-standard concept testing and volumetric forecasting system for CPG launches, built on a database of over 500,000 forecasts and 3,600+ historical launch plans.
Best for: The final go/no-go gate on a major launch, where a board or executive team needs a sales forecast, not just a directional read.
Strengths: BASES is the only forecasting model to pass the Marketing Accountability Standards Board’s audit protocol (MMAP), and NielsenIQ reports average forecast accuracy within plus or minus 9%. For a launch backed by real production capital, that track record is the reason BASES functions as a required gate at most large CPG companies: concepts below category norms typically don’t advance.
Limitations: Fielding a full BASES study runs multiple weeks, and NielsenIQ’s own volumetric forecasting page describes annual engagements in the seven-figure range for a full program. That’s the right investment for a handful of major launches a year. It’s the wrong tool for iterating on ten early concepts before you know which one deserves that budget.
Price: Custom, project-based. Individual modules like BASES Price Advisor turn around in as little as two weeks; full volumetric forecasting engagements are priced separately and scale with program size.
3. Zappi
What it is: Zappi runs automated, survey-based concept and ad testing with normative percentile scoring against a proprietary CPG benchmark database.
Best for: Enterprise innovation teams running a high-volume pipeline who want every concept scored against the same category norms.
Strengths: Zappi’s normative database, built from years of past CPG studies, is what makes a raw score meaningful: a concept that lands in the 45th percentile means something specific relative to your category history. Turnaround on an established Zappi account runs 24 to 48 hours once a study is scoped, and the platform supports concept, ad, and packaging testing under one system.
Limitations: Third-party buyer estimates put per-study cost in the $3,000 to $25,000-plus range, and Zappi doesn’t publish pricing, so budgeting requires a sales conversation. The percentile score also tells you where a concept lands, not why: teams still need a separate diagnostic pass to understand what would move the number.
Price: Enterprise, quote-based. Not publicly listed.
4. Suzy
What it is: Suzy is an on-demand consumer research platform built around an owned, always-verified panel, combining quant surveys with live qualitative tools like “Suzy Live” and in-home usage testing.
Best for: Teams that want a single vendor for concept testing, creative testing, and package testing, with the option to jump into a live consumer conversation when a survey score alone isn’t enough.
Strengths: Suzy’s panel is pre-verified and available on demand, so studies can go live the same day. The addition of live interviews inside the same platform gives insights teams a qualitative layer without stitching together a second vendor for follow-up conversations.
Limitations: Suzy sells an all-inclusive annual license rather than per-study pricing, with reported ranges from roughly $35,000 to $184,000 a year and a median around $99,000. That structure makes sense for a brand running dozens of studies annually across methodologies, but it’s a heavy commitment for a team that mainly needs concept testing.
Price: Annual license, reported median around $99,000/year.
5. quantilope
What it is: quantilope is an automated research platform built around advanced quantitative methods, including conjoint analysis, MaxDiff, and Van Westendorp pricing, with an AI co-pilot called “quinn” that helps design and interpret studies.
Best for: Teams that need to understand tradeoffs, not just a purchase-intent score, such as which combination of flavor, price, and pack size maximizes share of choice.
Strengths: quantilope automates methodologies that traditionally required a statistician or an outside agency. One client, Zurich Insurance, reported saving roughly €50,000 to €70,000 a year in agency conjoint fees after bringing the work in-house on the platform, and results populate in real time as responses come in rather than after a full fielding period closes.
Limitations: The depth that makes quantilope powerful for tradeoff analysis also makes it a heavier tool than most teams need for a simple early-stage concept screen, and pricing is quote-based rather than published, starting around $22,000 a year for the Business tier.
Price: Custom annual subscription, from roughly $22,000/year.
6. Attest
What it is: Attest is a global survey and concept testing platform with flat-rate, published pricing and a panel spanning more than 150 million consumers across 59 countries.
Best for: Challenger and mid-sized CPG brands that need multi-market reach without an enterprise research budget.
Strengths: Attest’s flat per-response pricing holds regardless of how granular the targeting gets, which makes budgeting predictable in a category where agency quotes usually require a call first. Templated concept testing modules and on-demand advisory support also help teams that don’t have a dedicated research function get to a usable read quickly.
Limitations: Attest’s panel and methodology are built for broad survey-based reads rather than the deep diagnostic layer a full-service partner like BASES or a qualitative platform like Suzy provides, so brands making a major capital decision typically still pair it with a heavier study downstream.
Price: Flat-rate, published tiers; mid-market friendly.
7. Upsiide
What it is: Upsiide, built by insights firm Dig Insights, is a mobile-native concept screening tool with a swipe-based interface designed to mimic real shelf decisions rather than a traditional rating scale.
Best for: Innovation teams that need to screen a large batch of ideas, up to 50 concepts in a single study, before narrowing to the handful worth developing further.
Strengths: The swipe-based format and Upsiide’s patented Market Simulator model share of choice, cannibalization, and incrementality across a concept set, which gives portfolio-level answers most single-concept tests don’t. CPG-specific templates cover flavors, packs, and claims out of the box.
Limitations: Upsiide is built for breadth across many concepts rather than depth on any one, so teams that need to understand the reasoning behind a score, not just the ranking, typically add a qualitative or diagnostic layer separately.
Price: Custom, project or subscription-based.
What methods do consumer goods testing tools use?
Most consumer goods testing tools run one of four designs. Monadic testing shows each concept to a separate group of respondents, which avoids comparison bias but needs a larger sample size to reach statistical confidence. Sequential monadic testing shows the same respondents several concepts in a row, which is more sample-efficient but introduces some order bias. Comparative testing puts two or more concepts side by side and asks which wins. Protomonadic testing runs a sequential monadic pass first, then adds a head-to-head comparison, combining both data types in one study.
The concept itself is usually shown as a concept board: a headline, a short description, and an image or claims wall, standing in for the eventual pack, shelf placement, or ad. Later-stage CPG work builds on the same concept with a packaging test, a claims test, and a price sensitivity read, then rolls up into trial-and-repeat estimates for the volumetric forecast. Our concept testing questions guide breaks down which questions map to which stage.
What’s a good BASES alternative for early-stage screening?
Concept screening and volumetric forecasting solve different problems, and most teams reaching for a BASES alternative are really trying to avoid sending every early idea through a multi-week, high-cost forecasting system before it’s ready for one. Zappi and Upsiide both work as faster, less expensive quantitative alternatives for the screening stage, while a synthetic tool like Articos works as a same-day pre-filter before any recruited-panel study runs. Most CPG teams that adopt this approach keep BASES in the process for the launches that need it; they just stop sending every idea to it.
Are there AI concept testing tools built for CPG categories?
Yes, though the category splits into two different approaches. Zappi and quantilope use AI to automate analysis and reporting on top of a recruited human panel, which is where their normative benchmarks come from. Articos uses AI differently: it generates synthetic personas and runs the entire interview and analysis cycle without a human panel at all, which is what gets turnaround down to under 30 minutes instead of days.
For a broader look at how synthetic panels compare across categories beyond CPG, see our roundup of synthetic user research tools. Neither approach replaces the other; teams increasingly use the synthetic version to decide what’s worth sending to the panel-based one.
What’s the cheapest or free option?
Articos has the lowest entry point on this list: $8 to $20 per study, with a three-day free trial that includes two free studies and no credit card required. Attest offers a free self-serve survey builder for sending questions to your own contacts, though its full 150-million-consumer panel sits behind paid tiers. Every other tool here, BASES, Zappi, Suzy, and quantilope, sells through custom quotes with no free tier, and the annual-license platforms carry five- and six-figure minimums regardless of how many studies you actually run.
Should I combine synthetic and human research, not choose one?
For most CPG teams, yes. The pattern that’s emerged across insights teams adopting AI tools is a dual-platform approach: a fast, low-cost tool screens and narrows a wide set of directions, and a recruited-panel platform like BASES or Zappi runs the final volumetric or normative study on the two or three concepts that survive screening. That sequencing catches weak ideas before they cost anything and reserves the expensive, high-confidence study for the concepts that have already earned it, rather than treating any single tool as the whole process.
How to choose a concept testing tool for your CPG brand
Three questions cut through most of the noise:
What’s riding on this test?
If the concept is heading toward a major production commitment, you need a full-service partner with an audited forecasting model: BASES, or a comparable quant study through Zappi or quantilope. If you’re narrowing ten ideas down to three, cost and speed matter more than forecast precision. Some teams also weigh general-purpose research platforms like UserTesting against the CPG-specific tools here; our UserTesting comparison covers where that fits.
How often does your team test?
Teams running weekly or monthly screens benefit from low per-study costs and self-serve setup. Annual-license platforms make more sense for teams running dozens of studies a year across multiple methodologies.
Do you need a score or an explanation?
A percentile ranking tells you where a concept stands. A structured interview tells you why, and what would move it. Most strong innovation processes use both, in that order: understand first, then rank.
The criteria above apply regardless of which tool you land on. Every option in this guide will beat a coin flip. What actually sinks a launch is skipping the test, or rushing it through a single method because the calendar got tight.
