| Release (Features) | Status | Date of Release |
|---|---|---|
| New data source addition | Live | Jul-05-2026 |
| Multi Segment Survey | Live | Jul-05-2026 |
| Multi Creative Assessment | Live | Jul-05-2026 |
| Creative Evaluation | Live | Jul-05-2026 |
| Standard Surveys gets Analytics Tab | Live | Jul-05-2026 |
| Brand Tracking Overtime | Live | July-14-2026 |
| LinkedIn New Summary Card | Live | Jun-29-2026 |
### Why App Reviews Are Different
App Store and Play Store reviews capture something that most other Mentions sources do not: the operational truth of the brand. They show where customers are delighted, where the experience breaks, what features are valued and what emotional language consumers use when speaking about a product they use regularly. Because these reviews are written about a specific, current version of the app by people with direct experience of it, they reflect the consumer relationship as it actually exists today rather than as it is perceived from the outside.
### Why This Matters
For app-led and service-led businesses, this update closes a meaningful gap in how Mentions reports are built. The most direct consumer signal available to these brands is now surfaced at the point of setup rather than left to the user to locate. The result is a stronger Mentions report, richer consumer memory for AI Twins and more grounded action points for the business.
## Multi-Segment Survey
A real audience is never one mind. The people, a brand needs to understand are rarely uniform in how they think, what they value or where they are in their relationship with the brand. Some are loyal. Some are considering. Some are aware but unconvinced. Some have churned. A survey that can only speak to one of those groups at a time does not reflect how an audience actually works.
Until now, a Quant survey on consumr.ai could only be run against one segment at a time. If a research team wanted to understand how three different audience types responded to the same set of questions, they had to create three separate surveys, run them independently and reconcile the results afterward. That process introduced operational overhead and made direct comparison harder than it needed to be.
### What’s New
When setting up a Quant survey, users can now select multiple segments and assign a proportion to each. consumr.ai allocates respondents across the selected segments in those proportions and runs the survey in a single pass. The results come back together, with each segment's responses clearly attributed, making comparison direct and immediate rather than something that has to be constructed after the fact.
### No More Imposed Labels
When a Quant survey is distributed, the system needs to know how to frame the respondent's perspective before they answer. Previously, consumr.ai used Stages of Conversion as that framing mechanism. A user setting up survey distribution could assign a Stage of Conversion to the audience, such as Loyal Customer, Churned or Considering, and the respondents would answer from that assigned position.
The problem was that Stages of Conversion are external labels, not memories. A segment built from observed behavioral data showing low brand affinity could be assigned the Loyal Customer label and asked to respond accordingly. The segment's own signals said one thing. The label it was given said another. The responses that came back were shaped by the label rather than by what the segment actually represented, which introduced a layer of distortion into the output that was difficult to detect and easy to overlook.
With this release, Stages of Conversion have been removed from the survey distribution workflow entirely. Respondents now answer from the mindset embedded in their segment.
### Why This Matters for Research Teams
This makes Quant surveys more realistic, more flexible and easier to manage. Teams no longer need to create separate surveys for every segment they want to study. They can bring multiple segments into one survey and compare responses within the same run. Because each segment responds from its own mindset rather than an imposed label, the output is grounded in what each audience type actually represents. Where segments agree, that agreement is real. Where they diverge, that divergence is real. Teams can see which opportunities exist across the full audience and which are specific to one group.
## Multi Creative Assessment
When a team has several creatives in hand, the question they are really trying to answer is not whether each one works. It is which one works best, and why. That is how creative decisions are made in practice: by putting options side by side, comparing them against each other and understanding what separates the stronger from the weaker. Until now, Multi Creative Assessment provided feedback on individual creatives but did not produce a true head-to-head comparison. A team could receive reactions to each asset separately but still struggle to identify which one should win and why.
This release brings that logic into Multi Creative Assessment. All uploaded creatives are now evaluated together in a single comparative pass. The AI Twins see every creative at once and react to them relative to each other, the way a real person actually decides between options, rather than rating each one in isolation. And for the first time, that comparison is built on a structured evaluation framework rather than on open-ended reactions.
### What’s New
The assessment begins with the user selecting a campaign objective. That objective determines which parameters the creatives are judged on and how those parameters are weighted. An awareness creative is judged on attention and recall. A conversion creative is judged on offer clarity and call-to-action strength. Each creative is then classified as Strong, Average or Weak against each parameter, producing a structured, weighted score that supports the comparison and makes the final verdict something the team can see the reasoning behind, not just accept.
### From Comparison to Direction
Once the assessment is complete, consumr.ai produces delta reasoning: a comparative explanation of why one creative performs better than another at the parameter level, not just a declaration of a winner.
From there, consumr.ai defines a Gold Standard. Based on the collective feedback from the AI Twins across all uploaded creatives, the Gold Standard identifies what the ideal next creative should contain: what to retain from the stronger assets, what to improve from the weaker ones and what to add. New creative variations can be generated directly from the Gold Standard output within the same workflow, moving the assessment from diagnosis to production in a single step.
### Why This Matters for Creative Teams
Creative feedback is most useful when it is structured enough to brief from. The objective-led parameter framework means every creative is judged on what it is actually meant to achieve, not on a generic checklist. The scoring matrix creates an audit trail that makes the verdict easier to understand and easier to defend. The Gold Standard gives the team a concrete next step rather than a set of observations to interpret. The path from assessment to the next creative brief is shorter and more direct than before.
## Creative Evaluation
Consumer opinion and market performance are related signals, but they measure different things. A creative can resonate strongly with an audience in a research setting and underperform in a live campaign. A creative that consumers find unremarkable can quietly convert at a strong rate because it communicates something timely and clear. The most trustworthy creative feedback is the kind that holds both signals at once: what the audience responds to and what the market has already confirmed works for this brand.
### What’s New
Creative Evaluation extends Multi Creative Assessment by grounding the evaluation in the brand's own real advertising history before the AI Twins assess anything.
The user selects the level at which to bring in performance context: account, campaign or ad set. The system reads the real performance data from that selection and identifies two of its best-performing and two of its worst-performing creatives. Those four assets are loaded into the AI Twins' memory as proven reference points before the new creatives are assessed making their feedback more specific to the performance metrics.
The campaign objective continues to determine which parameters are evaluated and how they are weighted. The proven performance signals shape the scoring and recommendations from within that framework, so the output reflects both what the audience responds to and what the brand's own campaign history supports.
### Why This Matters for Research and Media Teams
Research teams and media teams working on the same brief are typically working from different evidence. One from audience intelligence, the other from campaign performance data. Creative Evaluation brings both into a single assessment. The Gold Standard Action Steps are specific enough to brief from directly because they are derived from what has already driven results for this brand in this market. The creative variations generated from those steps are pointed in a direction the data already validates, which shortens the path from assessment to a confident production brief.
\
Analytics
---------
A survey produces responses. What research teams need is not the responses themselves but the understanding that comes from them. Who is this audience really? How do they see the brand relative to its competitors? Where do they spend their time and attention, and where are they most open to being reached? Until now, answering those questions required the team to do the interpretive work themselves after the survey came in.
This release changes that. consumr.ai now introduces an analytics layer that does that interpretation for you, turning the same set of survey responses into three kinds of insight.
### Segmentation
Segmentation groups respondents who answered similarly into distinct audience types and generates a complete profile for each one. Each segment receives a name, a classification, a demographic breakdown, a market size and a behavioral story covering what motivates that segment, what they want from the category and what gets in the way. The profiles are ready to read, present and act on without a manual analysis step in between. Users can also create an AI Twin directly from any segment output, moving from audience understanding to further research in the same workflow.
### Brand Track
Brand Track auto-constructs a brand health study from the brand and competitor list provided at setup, supporting up to five brands in a single run with no manual questionnaire construction required. The output measures each brand across a six-stage conversion funnel from Awareness through Familiarity, Consideration, Preference and Intent to Endorsement, showing precisely where brands gain and lose ground relative to each other through the funnel. A Brand Perception Matrix scores all brands side by side across qualitative dimensions including Customer Support, Innovation, Premium Feel, Trust and Value for Money, and classifies each brand as Leader or Follower based on its overall comparative score.
### Media Consumption
Media Consumption measures how the audience consumes media, when they consume it, on what devices, through which platforms and where they are most receptive to advertising. The output covers overall media intensity, channel and streaming platform distribution, device preferences, daypart behaviour and brand discovery patterns, with plain-language summaries accompanying each section so the findings are ready to present without additional interpretation.
### Why This Matters for Research Teams
Each survey type converts responses into structured, labelled outputs that are ready to brief from, present to stakeholders or feed into further research. The interpretive work that previously sat with the analyst now sits with the platform. Teams running segmentation, brand tracking and media planning as part of the same research programme can now initiate all three from Standard Survey and receive outputs that are consistent in methodology and ready to use.
## Brand Track Over Time
A single Brand Track run tells you where a brand stands today. It shows how the audience perceives the brand across six funnel stages, how it compares to competitors and where it is strongest and weakest relative to the competitive set. That reading is useful. What it cannot tell you is whether anything is moving, in which direction or why.
Brand health is not a fixed condition. It is shaped by campaign activity, competitor moves, category shifts and market events. A brand that scores well on awareness today may be losing ground on consideration without that erosion showing up anywhere. A campaign investment may be driving real shifts in preference or intent that never get measured because there is no wave run after the campaign ends. The number only becomes strategically meaningful when you can see whether it is climbing or slipping over time.
### What’s New
Brand Track can now be run as a recurring study across multiple waves. Each wave runs the same questionnaire on the same audience at a defined interval. Each wave is saved as its own reading, stamped with the date it was collected, and the readings are stitched together into trend lines for every metric and every brand. Awareness, Familiarity, Consideration, Preference, Intent and Endorsement can now all be charted across time rather than read as a single snapshot. Perception dimension scores across Customer Support, Innovation, Premium Feel, Trust and Value for Money are tracked across waves in the same way.
Running a wave before and after a campaign makes it possible to measure the direct impact of that campaign on brand health metrics in the same audience, as the audience composition is kept consistent from wave to wave. Not as an inference drawn from separate studies with different respondent pools, but as a measurable before-and-after difference in the same people, on the same framework, at two points in time.
### Why This Matters for Research Teams
Brand Track Over Time turns a point-in-time diagnostic into a continuous measurement system. Teams can track the health of their brand and up to five competitors across time, understand whether the trajectory is moving in the right direction and isolate the effect of specific campaigns or market events on brand perception. The before-and-after wave structure gives campaign measurement a level of precision that a single-run study cannot produce.
## LinkedIn New Summary Card
When a team builds a LinkedIn audience report, they are studying people in a professional context. The dimensions that make that study meaningful are the ones LinkedIn captures natively: what these people do, what industries they work in, where they are based and what they care about professionally. The summary card at the top of the LinkedIn audience report has been rebuilt to do exactly that.
### What’s New
The summary card now reads LinkedIn's own native data directly across four sections: Top Job Titles and Key Industries, Age and Gender, Top Interests and Location by City. Each section is sourced from LinkedIn's own breakdown rather than approximated through meta's taxonomy. Interests in particular are now shown as genuine LinkedIn interests, reflecting how the professional audience actually identifies on the platform.
Because LinkedIn data is not always complete depending on the size and composition of the audience being studied, the summary card has been built across a set of defined rendering states. If all four sections are fully populated from LinkedIn, the card renders in its complete state. If LinkedIn data is entirely unavailable, the card renders a corresponding empty state. For partial availability, the card renders intermediate states that display only what LinkedIn data genuinely exists. What the user sees always reflects the true state of the LinkedIn data, nothing more and nothing less.
You can find this in the Pro tab, inside any LinkedIn Behaviour Report created for a professional or B2B audience.
### Why This Matters for B2B Research
Professional audience research is only as credible as the data it is built on. When a B2B team presents a LinkedIn audience report to a client or a decision-maker, the audience description needs to reflect professional identity as LinkedIn defines it: what these people do, where they work, what level they operate at and what they care about in a professional context. The rebuilt summary card ensures that every section of that description is sourced directly from LinkedIn's own data, so what reaches the decision-maker is an accurate, unambiguous picture of the professional audience. For B2B research, where the audience definition is the foundation everything else is built on, that source-level accuracy is what gives the report its authority.