Brand strategy and consumer insights
Real consumer research and a strategy built on it, in four working days. Directed start to finish by a strategist with twenty years in the work, and for a good deal less than research like this used to cost.
What you get
A recommendation inside an organizing framework, so your stakeholders can see where the business sits, the idea that moves it forward, and how each action connects to the outcome in your brief.
Findings that name the motivation underneath what people say. Category givens are useful context and they are labeled as context, so the insights stay insights.
A PDF for presenting and an editable deck your team can lift slides from. The discussion guide, the screener and the panel come with it, and every claim traces to something you can read.
What it answers
A study is scoped to one decision. These are the shapes that decision usually takes.
If your question is not on this list, it is probably still the kind of thing this answers. Send it over and you will get a straight read.
Send your questionThe practice
Twenty years of turning consumer research into strategy that grows a business, across spirits and beer, food and beverage, retail, technology and financial services.
That experience runs from ethnography, sitting in people's homes and following them through their markets, through inventing products at an innovation consultancy, running innovation at a fast-growing food brand as it moved from startup to category leader, building a flagship retail experience, and leading commercial strategy inside a bank. The work has been run from the agency side and commissioned from the client side.
That is the part worth hiring. Knowing which finding is a real insight and which one everybody in the category already knows. Knowing when a respondent is telling a story about themselves. Knowing what a recommendation has to look like to survive a room full of stakeholders.
No study is handed to a junior team, and no part of one is left to run on its own. The screener and the discussion guide are written here, every transcript is read here, and your material is handled by one person. BABAHALA Group is led by Brian Armentrout, Principal Strategist.
Work has included
Across twenty years, at agencies and innovation consultancies, inside several of these companies, and since 2021 through BABAHALA Group.
The four days
A real sequence, and you sit at two points in it.
The discussion guide is written and the scan of current public evidence runs. You review the guide.
The panel is built and you review who is in it. Fieldwork runs overnight.
Transcripts are coded, themes and insights are built, and the market scan is finished.
The strategy is written, put through review rounds, and delivered as a presentation.
Have a decision coming up? Send it over and you will get a straight read on how this would work for it.
Send your questionHow it runs this fast
The speed comes from the panel. The thinking is still done the long way.
Each respondent starts from a real, anonymized US Census household record, drawn in proportion to how common people like them are among the buyers you screen for. Your buyers, rather than the general population.
Attitudes come from a scan of current public evidence, run fresh for every study and weighted to the past twelve months. Each one carries the source it came from and the date it was retrieved.
A moderator interviews each respondent for ninety minutes and keeps probing past the first answer, because the first answer a person gives is usually a rationalization. A stated reason never counts as a finding.
Why the work holds up
Twenty years teaches you the handful of places research goes soft, and how to keep it from happening. Every study is checked against those points before it reaches you, and the results go in the deliverable.
DepthWhether the moderator probed a must-cover topic or accepted the easy answer and moved on.
AgreementWhether respondents pushed back on the moderator's framing or simply adopted it.
Invented depthWhether a motivation a respondent stated rests on something real in their profile.
TraceabilityWhether every claim in the analysis points at an interview turn that says it.
StabilityA share of interviews is rerun, to confirm the same people hold the same positions.
CompositionHow the panel that was built compares against the screener it was built to meet.
Straight about the method
This is qualitative work, and it does what excellent qualitative work does. It tells you why people choose what they choose, and what would change their minds. It gives you a direction and the reasoning behind it.
Synthetic respondents model patterns of reasoning rather than individual people, so findings are directional and are never presented as projectable. Before a major spend, confirm the direction with a small live check.
Every deliverable states its method, its panel, its sources and its limits, right where your stakeholders can read them. Work that is straight about what it is tends to get trusted with the next decision.
Who it is for
For teams who commission research for a living and need an answer inside the planning cycle rather than after it. You will recognize the screener, the guide and the coding, because they are the ones you would have written.
For a restaurant group, a regional retailer, a local bank. Work at this depth has been out of reach for businesses this size, and it is not any more. You do not need to know what a discussion guide is to brief it or to use what comes back.
Each study is scoped to the decision in front of you, and it costs a good deal less than research at this depth used to. Ask, and you will have a number in the first reply.
Ask what it would costQuestions
Every respondent starts from a real, anonymized US Census household record, and every attitude they hold comes from a scan of current public evidence that carries its source and the date it was retrieved. What is synthetic is the conversation, not the evidence underneath it.
Nothing reaches your deck on the model's word. Quotes are pulled out of the transcript by turn number, so a quote cannot drift from what was said, and a claim that cannot be traced back to a turn is dropped before you see it.
A chatbot gives you the average of everything ever written about your category. It is confident, it is fluent, and it tells you what you already know.
This builds a panel of specific people who disagree with each other, screened to your actual buyers, then interviews each one for ninety minutes and reports where they split. The disagreement is the finding. An average cannot produce one.
From Census household records, drawn in proportion to how common people like them are among the buyers your screener describes. Your buyers, not the general population. A study on premium dog food is not full of people who do not own a dog.
Each person gets a life that fits their record: work, household, money, routines, what else they buy in the category and what they have never tried. You review the panel before anyone is interviewed.
Because a panel that agrees with itself is worthless, and it is the failure this method has to be built against.
Every study works out how buyers in that category genuinely divide, from that category's own evidence, and each respondent is built on a position, including what they do not believe. Somebody built to walk away over price does not drift into saying the cheap one is good enough. How far the panel agrees is measured and reported in the deliverable.
Yes. The full transcripts come with the study, along with the discussion guide, the screener and every panel profile. Any number in the deck can be followed back to the interviews behind it, and any quote to the moment it was said.
It will not give you a projectable number, and it is not the instrument for sizing a market or setting a forecast. Findings are directional, and they say so on the page.
Before a major spend, confirm the direction with a small live check. If your question needs a number rather than a reason, you will hear that in the first reply rather than after you have paid for a study.
A paragraph on the decision you are trying to make is enough to start. Four working days from approval to presentation, and you sit at two points in it: the discussion guide on day one and the panel on day two.
Between those, the work runs. You are not managing a fieldwork vendor or waiting on a recruiter.
Your brief is used to run your study and nothing else. It is never used to train a model, and it is not pooled with anyone else's work.
Your name is not used as a reference without your say-so, and a case study only happens if you agree to it in writing.
Start a conversation
A paragraph on the decision you are trying to make is plenty to start with. You will get back what the work would look like, what it would cost, and an honest read on whether this is the right instrument for your question.