From fieldwork to lasting business capability.FORWARD-DEPLOYED ENGINEERING
Agent Cases
← All casesFIELD NOTE 08 / Marketing & Growth

Iris—

From surveys and buying decisions,
to launch content, bundle recommendations and a review.

Research report and post-launch review delivered

An imaging device that could switch between modes of use was approaching launch. The brand team had collected survey responses, but still needed to decide which audience to address first, how to explain the product and how to discuss product bundles. Iris organized use cases, purchase intent and concerns from the raw responses, then delivered a research report, content directions and commercial recommendations. After launch, Iris compared the actual offer with the research and turned its unanswered questions into topics for the next survey.

Start with the value ↓See a working example ↘
A person holding a pocket-sized imaging device outdoors
From liking a concept to choosing the product you need.
Existing survey responses became product and audience assessmentsResearch extended into page and video content ideasThe launch prompted a review of bundle assumptions and future questions

WHY THIS WORK MATTERS

What interests users, and what still makes them hesitate before buying? Which launch tasks should each answer inform?

BUSINESS VALUE

Give the team the findings, assumptions behind the judgments and the researcher’s corrections together. Content teams know which situations to explain, product teams know which bundles to test, and the next study can continue from the questions already uncovered.

  1. 01Check questions, valid responses and sample composition
  2. 02Read interest, context, intent and hesitation separately
  3. 03Translate findings into selling points and content tasks
  4. 04Organize bundle recommendations and assumptions to test
  5. 05Compare the launch offer and revise the next survey questions

WHAT THE TEAM BRINGS

Raw survey responses, question definitions, confirmed product information, launch communication needs and the bundle assumptions used at the time. Add the final launch offer for the review.

WHAT THE WORK LEAVES BEHIND

  • Survey analysis with question definitions and sample scope
  • Product and audience assessments, plus page and video content directions
  • Pricing and bundle recommendations with their assumptions
  • Post-launch comparison and questions for the next study

HOW TO ASSESS A PILOT

  • Actionability: whether each key finding leads to a content task, product discussion or further research
  • Usability: whether the team can turn the recommendations into page messaging or a filming brief
  • Research continuity: whether the next study checks earlier assumptions and unresolved purchase concerns
Suggested evaluation criteria—not measured results claimed for this case.
START SMALL

Start with a collected survey that has not yet informed a decision. Choose one important purchase question, deliver a table of findings, judgments, content actions and items to validate, and have the team that will use it review it in practice.

See the actual process

THE METHOD IN A WORKING EXAMPLE

From “What do you like?” to “How would you buy?”

Iris separates interest, use contexts, purchase intent and concerns, then brings the findings into content and business discussions. The post-launch review asks what the original questions missed.

FIGURE 01Case illustration · English board

Research should make the next round of choices clearer.

A working board connecting user research to action. It contains no real survey, prices or preference distributions.

What to look for

Survey analysis first produces content tasks and conditional bundle recommendations. After launch, the actual offer reveals purchase paths omitted from the original questions, informing the next round of research.

The work in this case

Iris analyzes an existing survey and translates findings into messaging priorities and research recommendations. After launch, the actual offer is used to revisit assumptions and formulate more precise questions.

01 / THE ASSIGNMENT

Start where the business
gets stuck.

For brand and product marketing teams that have collected user feedback and are preparing a product launch. The survey contains feature rankings and attitudes toward purchase, but the team still needs to know how those findings should change the page, content and product bundles.

01

Liking a product does not mean being ready to pay

A new format attracts attention, but a purchase still depends on where someone would use it, whether an existing device could do the job and whether they need the full set of features.

02

Existing customers do not represent the entire market

The sample was more familiar with the brand and category, and skewed toward particular groups. Ignoring that composition could turn existing customers’ reasons to upgrade into assumed reasons for new customers to buy.

03

Ask about the wrong unit of purchase and the recommendation misses a level

Treating the complete kit as the only purchase option misses the route of buying the core capability first and adding extensions later. A single price range cannot answer the bundle question.

02 / THE CONTINUOUS WORK

Each deliverable must help
the next task move forward.

Expand each stage to see its inputs, decisions, actions, outputs and handoff. Each step explains how the next person could continue.

  1. 01Check what the survey can answer
  2. 02Break feature interest into purchase decisions
  3. 03Translate findings into content tasks
  4. 04Make recommendations around purchase bundles
  5. 05Use the launch offer to check the original research
01

Check what the survey can answer

A checklist linking each question to its definitions and sample composition.

Input
Responses already collected by the brand, the question structure and the raw spreadsheet.
Decision
Confirm the valid responses, single-choice or multiple-choice definitions and respondent groups for each question. One overall sample size cannot stand in for every question.
Action
Clean the data by question, organize options and missing values, and note the share of existing users and audience skew to define where later findings apply.
Output
A checklist linking each question to its definitions and sample composition.
Handoff
Every later percentage or ranking must trace back to its question. Figures with different denominators cannot be combined into a single funnel.
02

Break feature interest into purchase decisions

A four-layer view of purchase decisions and a summary of audience opportunities.

Input
Questions about feature preferences, use cases, purchase intent and concerns.
Decision
A popular feature is only the first layer. The practical task and the barriers to paying need separate interpretation; a neutral response cannot be treated as an imminent purchase.
Action
Organize findings into four layers: what attracts people, where they expect to use it, their attitude toward buying and what makes them hesitate. Record both supporting and conflicting signals.
Output
A four-layer view of purchase decisions and a summary of audience opportunities.
Handoff
Communication starts with relevant use cases, then addresses questions about learning, image quality and bundles, instead of making the new format the entire proposition.
03

Translate findings into content tasks

Research report, content directions and creative recommendations.

Input
Purchase assessments, product information and the brand’s launch communication needs.
Decision
Emphasizing ease of use is not yet an executable task. The brief needs to specify which scenes or content would help someone understand it.
Action
Organize the product narrative, primary and secondary selling points, content directions and filming scenarios. Turn functional benefits into actions that can be shown, and set aside unsupported claims for verification.
Output
Research report, content directions and creative recommendations.
Handoff
Brand and creative teams can use the work to plan pages, short-form content and video topics. The teams decide what to adopt and execute.
04

Make recommendations around purchase bundles

Commercial recommendations with explicit premises and items to confirm.

Input
Survey responses on purchase attitudes and concerns, plus the bundle assumptions in use at the time.
Decision
Users may want the core capability first without buying everything at once. The original survey did not sufficiently distinguish the bundle options.
Action
Include pricing and bundle recommendations in the report while preserving the assumptions behind them. Do not present recommendations as confirmed prices or extrapolate them into sales forecasts.
Output
Commercial recommendations with explicit premises and items to confirm.
Handoff
The recommendations enter team discussion. Their relevance needs to be reassessed against the final sales offer.
05

Use the launch offer to check the original research

Post-launch comparison and questions for the next study.

Input
The public launch page, actual product bundles and the original report’s conclusions.
Decision
The final offer separated a core device from expansion bundles, showing that the original one-price-for-the-whole-kit question did not cover every buying path.
Action
Distinguish alignment in direction, unconfirmed adoption and gaps in the original assumptions. Translate the review into future survey questions about willingness to pay for the core capability, interest in extensions, bundle choice and alternatives.
Output
Post-launch comparison and questions for the next study.
Handoff
The research leaves both conclusions and changes to the way questions should be asked, avoiding another round of calculations built on the same incomplete assumption.

03 / THE DECISIONS

The decisions matter
as much as the outputs.

When a draft, a source or the actual behavior differed from expectations, how did the Agent locate the problem and choose the next step?

01

Separate novelty from reasons to pay

The signal
The new format attracted attention, but ease of use and everyday shooting experience also mattered.
The choice
Read feature interest alongside context, intent and hesitation, then assign different communication tasks to each.
What it changed
The page can first answer what the product does for the reader, then explain its technical capabilities.
02

Keep existing-customer findings within their scope

The signal
Participants were relatively familiar with the existing brand and category.
The choice
Label findings as directional observations from this sample. New users still require separate validation.
What it changed
Avoid substituting upgrade needs for new-customer education or inventing market-size and purchase forecasts.
03

Turn a judgment gap into a better research design

The signal
The actual launch offer had more distinct bundle levels than the original pricing question.
The choice
In the next study, test the core product, extension capabilities and bundle choices before discussing price ranges.
What it changed
The survey moves from asking what price people accept to asking which part of the offer they are paying for.
Iris / IN PRACTICE

A user’s interest is where research begins. Understanding why they have not bought yet brings it closer to the business decision.

04 / WHAT CHANGED

Compare the work.
Understand what changed.

Compare the starting situation, the Agent’s actions and the resulting change. Read these alongside the working examples to see what was actually delivered.

01Feature rankings → layers of purchase decisions

BEFORE / THE STARTING SITUATION

Put the most popular feature at the top of the page and assume it is the strongest reason to buy.

IRIS / THE WORK

Analyze use cases, purchase attitudes and hesitation together.

AFTER / THE CHANGE

Explain value through situations and address concerns through content. Each finding has an explicit downstream task.

02One complete-kit price → several purchase paths

BEFORE / THE STARTING SITUATION

The study largely formed price expectations around the complete kit.

IRIS / THE WORK

Compare the launch bundles and separate the core capability from extensions.

AFTER / THE CHANGE

The next study can compare buying the core product first, choosing a bundle and upgrading later.

03Report delivery → review the report itself

BEFORE / THE STARTING SITUATION

Treat sending the report as the end of the research.

IRIS / THE WORK

Use the public launch offer to examine the assumptions and reasoning used at the time.

AFTER / THE CHANGE

Retain useful directions and turn weaknesses in question design into checks for the next study.

05 / FEEDBACK INTO THE WORK

Feedback changed
what happened next.

Checking the sample changed where the findings applied

Feedback was no longer presented as a rule for the whole category. The report retained audience scope and question definitions.

Content moved from what the product has to how it is used

Abstract selling points needed to become actions that could be filmed and situations people could understand.

The launch offer changed the next set of questions

The original pricing recommendation should not be reused on its own. It needs to be considered alongside the final product bundles.

06 / THE DELIVERY

Work the team can use.
A handoff it can continue.

The brand team designed and collected the survey. Iris handled cleaning, analysis, recommendations, report delivery and the subsequent review. Similar wording on public pages is not treated as proof that the recommendations were adopted, and public crowdfunding performance is not attributed to this report.

01

Survey analysis reference

Question definitions, sample composition, preferences, intent and hesitation organized by question.

02

Product and audience assessments

User tasks, communication opportunities and research limitations brought together in one report.

03

Content and creative directions

Survey findings extended into page messaging, content topics and video scenarios.

04

Commercial recommendations

Pricing and bundle recommendations state their assumptions, with unresolved items shown separately.

05

Post-launch review

Comparison with the final offer separates alignment in direction from unproven attribution.

06

Next research questions

Bundle choice, interest in extensions and alternatives become part of the next round of questions.

The Agent sustains the work. The team owns the delivery.

BUSINESS CONTEXT

Brand and research leads

Design and collect the survey, provide product facts, and decide the product bundles and final launch offer.

AGENT CONTRIBUTION

Iris

Cleans and analyzes the data, organizes the report and content recommendations, then reviews the reasoning after launch.

TEAM COLLABORATION

Marketing and creative teams

Produce pages and content and execute the launch using the final offer.

GO DEEPER

Implementation, supporting diagrams and retrospectives

Explore interactive methods and supporting handoff detail. The business process and deliverables are presented above.

EXPLORE THE METHOD

Supporting diagrams and methods

Iris separates interest, use contexts, purchase intent and concerns, then brings the findings into content and business discussions. The post-launch review asks what the original questions missed.

FIGURE 02Method board · Example content

What decisions do four types of research information support?

A review-method board, not a statistical chart from the original survey.

What to look for

Feature interest informs the messaging entry point; use contexts inform content tasks; intent and concerns inform validation and business discussions.

The work in this case

Keep the denominators and meanings of different questions separate. Attention cannot be treated directly as sales.

FIGURE 03Method board · Example content

Ask what people would buy before asking what they would pay.

An example question structure for the next research round, not a validated pricing plan.

What to look for

Replace one question about the whole package with three purchase paths: the core product, an extension and the complete bundle.

The work in this case

The improvement is in the questions: people willing to buy the core product may not need every extension.

07 / THE REUSABLE WORK

Turn project experience
into a reusable method.

Brand and product marketing teams launching a product and looking to turn user feedback into positioning, bundles and content tasks.

01

Four layers of purchase decisions

Inputs
Raw survey responses and question definitions
Method
Analyze interest, context, intent and hesitation separately, and verify the denominators
Outputs
An assessment table for content and commercial discussions
Checks
Interest is not a purchase, and findings from one sample do not represent the entire market.
02

Product bundle research

Inputs
Core capability, extensions and feasible bundles
Method
Compare purchase paths, then test intent and concerns for each
Outputs
Bundle-choice matrix and a follow-up survey
Checks
Make clear whether respondents are evaluating a single product, a bundle or an extension accessory.
03

Research translated into creative work

Inputs
Survey findings and verified product information
Method
Write one situation, one action and one question to validate for each benefit
Outputs
Task cards for pages and videos
Checks
The scene should explain the benefit, and the copy should not expand the technical claim.

A PRACTICAL START

A practical place to start.

Begin with a collected survey that has not yet been translated into decisions. Build a table of findings, judgments and next actions, then decide what further research is needed.

Explore AIelsen Agent and FDE services ↗

Apply this case’s methods to your own work

Download the workbook ↓Editable Markdown · No original client material

KEEP EXPLORING

Explore another workflow.

All 9 cases ↗