Command Center
Sunday 24 August 2026 · 6:42 PM IST · All districts · Wave 3, day 28 of 55
Where responses are coming from
Live submissions and party favourability, by district · Telangana
Live response feed
Straight from the field, as agents press submit
Response volume
Daily submissions, last 14 days · all surveys
Sentiment, day by day
Daily share on the govt · neutral not shown
What people are talking about
Issue mentions across open answers · Wave 3
Vote intention, if polls were tomorrow
Q22 · n = 17,311 answered · illustrative sample data
Undecided voters concentrated in Nalgonda, Khammam and Mahabubnagar, all three sit below zero on favourability. That is the persuasion battleground.
District watchlist
Largest field footprint first · tap a row for the map
| District | Responses | Coverage | Net sentiment |
|---|
Field operations health
Data pipeline and quality, right now
Live Map
Every submission, geo tagged · 119 assembly constituencies · 35,658 polling booths
Constituencies
All 119 · ranked by net favourability to us
Constituency detail
Tap any constituency on the map
Constituencies
All 119 assembly seats · 33 districts · 17 Lok Sabha seats · 35,658 polling booths
Can the sitting parties hold their seats
2023 result versus where this wave's data points, seat by seat
Constituency register
Click any seat to open it on the live map with its polling booths
| AC | Constituency | District | Held by | 2023 margin | MLA | Party net | Member vs party | Projected | Call | Booths | Worked | Responses |
|---|
AI Issue Detection
Patterns mined across every response, constituency, booth and the 2023 result
If you do only three things this week
Ranked by electors affected and how fixable the cause is
Voter Bank & List
The electorate at every level, from the whole state down to a named elector
Register
AI Scheme Mapping
Who qualifies, who is actually getting it, and the question that measures the gap
Telangana
Whole state roll
Income tier of this electorate
Modelled, and what it implies for who qualifies
Where the gap is widest
Ranked by households missed weighted by how badly the scheme is failing
Scheme register
Click any scheme to open its eligibility rule, its field question and the exact booths to ask it in
| Scheme | Level | Benefit | Who qualifies | Eligible | Receiving | Reach | Gap |
|---|
Eligibility is estimated from electorate composition and a modelled income tier. Receipt is what respondents report, not an administrative count. Scheme names, benefits and eligibility rules are real.
Smart questionnaire for this scope
Generated from the widest gaps here, in the three languages the field app speaks
AI News Research
Telangana coverage read every day and attributed down to the constituency
Headlines, links and images belong to their publishers and are shown with attribution. Tone is an automated read of how a story lands for the party in office, and should be confirmed by a human before it drives a decision.
Today's Brief
Where you stand
Where it is worst
What they are saying
Issues raised most, and which way they are moving
The arithmetic
Do these things today
In this order, and nothing else until they are done
In the press this morning
Also worth knowing
Only what has something to say today
Needs you first
Campaigns & Waves
Plan the rounds, see which households have been visited how many times, and push each wave to the field
Waves
Each wave is one full round of the constituency, with its own questionnaire and deadline
Contact depth
How many times each household has actually been reached
Where the next round should go
Wards with the thinnest contact, worst favourability first
Wave roster
Who is assigned, what they cover, and how far through they are
| Agent | Covers | Households | Done | Progress | Last sync |
|---|
AI Speech Builder
A speech brief for any level, from a single village up to the whole state, built from every layer underneath it
What to build it from
Turn a layer off and the brief is rewritten without it
Lead with this
Open here, and give it the most time
Do not raise
Every one of these will cost more than it returns here
The speech
Read it as written or take the bullet points
Local names and facts to use
Naming the specific place and the specific number is what makes it land
Casework
Grievances raised on the ground and lifted out of the news, tracked to closure
| Case | Where | Category | Source | Raised | Age | Status |
|---|
A case lifted from the news is created automatically when a story names a grievance and resolves to a seat. It stays unverified until an agent confirms it on the ground.
AI Media Scorecard
Which outlets carry us, which run against us, and what each one keeps returning to
Outlet scorecard
Every masthead we read, ranked by how it treats us
| Outlet | Stories | Tone | Net | Runs most on | What to do with them |
|---|
Subjects the press keeps returning to
Volume and tone by subject across every outlet
Seats getting the most coverage
Where the press is paying attention, and how it reads
Expenditure
Lawful campaign spend against the Election Commission ceiling, ready to file
By head of expenditure
The heads the Commission's own return asks for
Spend by week
Against the notified campaign period
Register
Every entry, with the voucher reference the return needs
| Date | Head | Detail | Where | Amount | Voucher |
|---|
Log Spend
Enter an expense once, against the head the Commission's return asks for
Against the ceiling
Every entry needs a head, an amount and a voucher reference, because the Commission's return needs all three. Cash entries above the reporting threshold are flagged for the agent's attention.
Survey Integrity
Proof that the fieldwork actually happened, agent by agent
Agent scorecard
Every check that can be made without taking anyone's word for it
| Agent | Covers | Responses | GPS verified | Median duration | Duplicates | Score | Flags |
|---|
What is checked
Every one of these is measured, not asserted
AI Political Analyst
Every layer collapsed into one position: what to lead with, what to leave alone, and the citation behind each call
Constituency
Lead with this
Talk tracks the data supports here
Do not raise this
Where the ground or the news will turn it against you
Where each argument lands
The seats and booths to take it to first
AI Questionnaire Generator
Build a field questionnaire from everything the AI layers already know about a seat
What to build
Pick the seat and the layers to draw from
Every question carries the number it came from, so the set can be defended in a review.
Your questionnaire
Choose a seat on the left and generate
Nothing generated yet. Pick a constituency, choose which layers to draw on, and hit generate.
AI Digital PR
What the internet is saying about us, by channel, topic and audience
Channel by channel
Volume and tone, and what each channel needs to go live
Who is talking
Audience cohorts and how each one reads, tied to where they live
What they are talking about
Top topics by volume and tone
Keywords in play
The language people actually use about us
Posts, and how the comments landed
Sentiment on the post itself, and on every comment under it
AI Rival Watch
Every other party's online position, and the openings it leaves us
All parties, side by side
Volume, tone and momentum across every party we track
Topic board
Who owns which subject, and where they are exposed on it
| Topic | Owned by | Mentions | Reads | What it means for us |
|---|
Where they are strong
Do not fight them here without a better answer
Where they are exposed
Openings the conversation is already giving us
Their posts, and how their own audience reacted
A post reading well while its own comments read badly is the opening
AI Audience
Follower base health: who actually follows us, who is a bot, and who is a rival account
Follower quality by platform
A base padded with bots produces engagement that never becomes a vote
| Platform | Followers | Real and active | Dormant | Suspected bot | Rival accounts | Engagement |
|---|
Generational split
Share of the follower base, and how each cohort reads
Where that cohort actually votes
Seats with the highest share of this age band on the roll
What this means on the ground
Audience data crossed with the electorate, seat by seat
Surveys
Create once, collect by app, shared link, or scanned paper form
Wave 3 · Household Pre Poll Survey
Door to door · 24 questions · all 33 districts
Farmer Distress Deep Dive · North
Adilabad, Nirmal, Nizamabad, Jagtial · 14 questions
Urban Youth Employment Pulse · GHMC
Shared WhatsApp link · self serve · 10 questions
Post Rally Feedback · Warangal Public Meeting
Event exit survey · 8 questions · 21 Aug
Wave 2 · Household Pre Poll Survey
Baseline for comparison · closed 26 Jul
Candidate Recognition · LB Nagar
Draft · 12 questions · awaiting sign-off
Deploy
Collection modes
Quality & privacy
Speed-run and duplicate-location checks flag suspicious entries for supervisor review automatically.
Responses
Every submission, app, link and paper, in one stream
| ID | Time | Respondent | Location | Agent | Mode | Sentiment | Vote intent |
|---|
AI Insights
What the ground is saying, analysed the moment it arrives
Overall sentiment on the state government
18,462 responses scored · Wave 3 to date
Net sentiment −1, up 4 points from Wave 2 (−5). The recovery is broad, Siddipet (+14) and Nizamabad (+11) lead, but drinking water keeps Khammam (−16) and Warangal Rural (−13) deep in the red.
Sentiment by issue
How people feel when they raise each issue
Sentiment by demographic
Net sentiment toward govt · Wave 3
What's working for the govt
Positive points auto-extracted from open answers
What's hurting the govt
Negative points, ranked by frequency × intensity
In their own words
Representative voices · translated from Telugu / Hindi
Constituency league table
Assembly segments where the swing is largest since Wave 2
| Constituency | District | n | Net sentiment | Δ vs Wave 2 | Top negative issue | Call |
|---|
Field Teams
102 agents · 6 teams · productivity and data quality, live
Agent roster
Sorted by responses today · flagged rows need a supervisor's eye
| Agent | Team · District | Today | Wave 3 | Avg time | Last sync | Status |
|---|
Team leaderboard
Responses this week
Needs attention
Reports
Auto written from live data, no more waiting months for the paper pile
Daily Situation Report · 24 Aug
Auto generated 6:00 PM · sent to 14 people
Wave 3 · Week 2 Rollup
Generated Sun 11:00 PM · trend vs Wave 2
Constituency Deep Dive · Warangal East
On demand · requested by A. Rao, 23 Aug
Post Rally Feedback · Warangal
Event report · closed survey
Daily Situation Report · Sunday 24 August 2026
Executive summary
The recovery is holding: net sentiment −1 (Wave 2: −5), led by Siddipet (+14), Nizamabad (+11) and Medchal (+9). The race is INC 34 vs BRS 36, inside the margin of persuasion with 8% undecided. The one drag is drinking water, which keeps Khammam (−16) and Warangal Rural (−13) deep red.
Sentiment, day by day
Where the ground moved this week
Most raised issues today
Next 72 hours · recommended actions
Data quality
12 submissions flagged (speed run and GPS cluster checks); 24 paper scans in the OCR queue; GPS lock on 97.4% of submissions. Flagged records are excluded until review.
Field Agent Mobile App
What your ground team carries instead of a clipboard
Paper took months. This takes seconds.
Agents open the assigned survey, ask the questions exactly as scripted, and submit. The dashboard updates before they've reached the next house.
Villages with no signal are no problem, responses queue on the device and sync automatically when the network returns.
Each submission carries its coordinates and timestamp, that's what draws the survey trail on the Live Map and stops made-up entries.
The agent flips language per respondent. Voice answers in any of the three are transcribed and sentiment scored automatically.
Teams that prefer printouts photograph the filled form, OCR reads the handwriting and files it like any other response, same dashboard, same charts.
Every survey also has a public link and QR, share it on WhatsApp groups and let people answer on their own phones.
Try the real thing on your phone
Scan to open the live field app, fill a survey and watch it land in this dashboard's live feed within a second. Switch on airplane mode to see offline capture; reconnect and it syncs automatically. Add it to your home screen and it behaves like an installed app.