One general. One message. One number that decides it. That is the whole theory of political consulting, and almost nobody who buys it has read it. This page reads it out loud: what the consultants sell, why candidates buy it, how the 96.8 crore people on the 2024 roll actually think, and what the result does to every contract, licence and business plan in the state for five years.
We are the digital, data and AI layer under a campaign, not the general. Fixed fee in writing after a scoping call. One candidate per constituency per cycle, in writing. Scope: one seat, one legislator, or one consultancy that needs a technical partner.
A candidate learns the booth-level maths of a constituency in the last twenty days, from the people who want something from the candidate. The consultants learned it a year earlier, from the data. That gap is the whole business.
Four hundred volunteers, nine groups, no list of who is in which, and nobody can say how many homes were visited on Tuesday. The campaign is running. It is just not being counted.
Pre-certified ads. Declared social handles. A 48-hour silence. A Rs 40 lakh ceiling that an expenditure observer can audit line by line. Enthusiasm is not a compliance plan.
Not victory. Nobody can sell that. They sell a smaller chance of losing stupidly.
Let us start with a number. Rs 1,35,000 crore. That is what N. Bhaskara Rao of the Centre for Media Studies estimated the 2024 Lok Sabha election cost, all in, parties, candidates, government and the Election Commission together (CMS estimate, April 2024). About Rs 1,400 per voter. More than the 2020 American presidential election. The most expensive democratic exercise ever held on this planet. And here is the thing. The legal spending limit for a Lok Sabha candidate in a larger state is Rs 95 lakh. For an assembly seat in a larger state such as Karnataka, Rs 40 lakh; smaller states carry Rs 75 lakh and Rs 28 lakh (Election Commission revision, January 2022). Those ceilings cover one thing: a candidate's own spend between nomination and the declaration of the result. Parties have no ceiling. Governments spend on the conduct of the poll. Political advertising, by one trade estimate quoted by the consultancies themselves, ran past Rs 3,000 crore in the financial year of the general election. Put together, the ceilings and the total are describing different things. Party spend has its own disclosure rules, and whatever a party spends on a specific candidate is attributable to that candidate under Section 77. The consulting industry lives in that larger, party-level number. This page is an attempt to describe it honestly. It is not legal advice; the rules section links the originals.
Political consulting in India did not begin in 2014. It got a name in 2014. A former public-health professional ran a campaign that treated an election like a product launch: survey first, message second, candidate third. He built an organisation around it, ran it for governments in Bihar, Punjab, Andhra Pradesh, Delhi, Tamil Nadu and West Bengal, and then, in the most Indian plot twist possible, walked out of the consulting room and into the ring with a party of his own. Prashant Kishor is the name. The method is what matters. And the method is simple enough to write on one slide.
There can be no war without a general. One face, not twelve. One message, not a manifesto. One number, not a mood. The number is the swing you need in the booths you can actually move. Everything else, the rallies, the songs, the hoardings, the door-to-door, is spend in service of that number. A consultant who cannot tell you the number is an event manager. An expensive one.
So what does a consultancy really sell? Think about it from the buyer's side. A candidate has paid, in the currency that matters, for a ticket. The ticket is a lottery with a five-year prize and a 50 per cent chance of total loss. The buyer is afraid. The consultancy sells a reduction in variance. Fewer dumb mistakes. A message tested before it is printed. A list of the 40,000 homes that decide it. Proof that Tuesday happened. That is the product. Certainty is not on the menu. Anyone offering it is selling you something else.
Every constituency in India is three numbers. Turnout. Split. Swing. Take an illustrative assembly seat of 2,40,000 voters, invented for the arithmetic. Turnout 70 per cent, so 1,68,000 votes get cast. The top two take, say, 45 and 40. That is a gap of 8,400 votes. Now watch what the consultant does. They do not try to convince 1,68,000 people. They find the 60 booths, out of maybe 280, where the margin was under 60 votes last time, where the roll grew fastest, where the local intermediary changed sides. They work those. A four-point swing in the right 60 booths moves more than a ten-point swing in the wrong 100. That is not magic. It is arithmetic with a map. Most candidates have the map. They have never opened it.
Where does the map come from? Public data, mostly. Form 20, the booth-wise result the Returning Officer publishes after every count. The electoral roll, by part and by age band. The census, by ward. Past turnout by booth. Scheme beneficiary counts by panchayat, which the state itself publishes. None of this is secret. All of it is scattered across forty PDFs in three scripts. The consultant's first product is simply a spreadsheet that puts it on one screen. We build that spreadsheet. It is the least glamorous and most valuable thing on this page.
Here is where the film changes reel. Everyone who has lost an election has a theory about voters. Almost all of those theories are about the consultant's own neighbourhood. India does not vote as one mind. It votes as at least eight.
The beneficiary. The most important voter in modern India and the least discussed in English. A household that receives a direct transfer, a gas connection, a house, a ration, a pension, knows exactly who signed it and when the money stopped. This voter is not ideological. This voter is transactional in the most literal and legitimate sense: a promise was made, a credit arrived, the ledger is kept. Campaigns that talk only about vision lose this voter to campaigns that talk about the next instalment.
The community voter. Caste and community are not a bug in Indian politics. They are the oldest trust network in the country. A community votes as a bloc when it believes the bloc is how it gets heard. The consultant's job is not to pretend this away but to understand which blocs in which booths are movable, which are settled, and which intermediaries actually carry them. A community leader who switched sides in March is worth more than a rally in April.
The aspirational young voter. First-time and second-time voters, roughly a fifth of the roll in many seats, who do not remember the scheme that won the last election and are not grateful for it. They want a job, a road to the job, and a leader who does not embarrass them on their phone. They are reached on YouTube and Instagram and lost by a single clumsy clip. They are the least loyal and most persuadable segment on the roll, which is why every consultancy over-spends on them and under-measures them.
The woman voter. The quiet revolution. Women's turnout now matches or beats men's in many states, and women vote differently from the men in their own house more often than the men know. Direct transfers to women's accounts, safety, liquor policy, cooking fuel: these decide seats, and they are decided in the kitchen, not at the rally.
The anti-incumbent. A voter who is not for anyone. Who is against whoever is there. Five years of small humiliations at a government office add up to a vote. This voter is reached with one question: what exactly changed in your lane? If the answer is nothing, the consultant's job is to make sure the challenger asks it in every booth.
The urban absentee. The IT park, the gated community, the family that holds strong opinions and a voter card registered in a village 400 km away. Loud online, missing at the booth. The mistake is to campaign to them because they are visible. The discipline is to count them once and move on.
The local-strongman voter. Votes as the panchayat president, the cooperative boss, the contractor or the temple committee tells them to, because that person fixed the water last summer. Reached through the intermediary or not at all.
The split-ticket voter. The same person who votes one party for Parliament and another for the assembly, and a third name for the corporation, and sees no contradiction. India's voter is more sophisticated than India's pundits. Campaigns that assume one election predicts the next lose the next.
Eight minds. One roll. The consultancy that treats a constituency as a single audience is advertising, not consulting.
Now the money. Not the campaign money. The money after. A state government in India signs, directly or through its boards and corporations, contracts for roads, water, housing, metro lines, power, hospital supplies, software, school meals, sand, liquor, mining leases and advertising. Public procurement in India is routinely estimated at around a fifth of GDP. The Union Budget alone set capital expenditure at Rs 11.21 lakh crore for 2025-26 (Budget speech, February 2025). The Government e-Marketplace crossed Rs 4 lakh crore in annual transactions in 2023-24 (GeM, April 2024). Every rupee of that has a signature somewhere near a politician.
So here is the honest version of what an election does to the economy of a state. For the six months before it, decisions slow, because nobody signs a file that a new minister might reopen. For the six weeks of the Model Code of Conduct, they stop. The day after the count, a thousand contractors learn whether their bills will be cleared or re-examined. Projects get renamed, retendered, or quietly continued. Land that was a residential zone becomes a corridor. A licence that took four years takes four months, or the reverse. This is not a scandal. It is the operating system. Businesses that understand the electoral calendar plan their working capital around it. Businesses that do not, discover it.
The repercussions run downstream for years. A contractor on the wrong side of a change in government waits for arrears and sometimes for an inquiry. A supplier on the right side grows faster than the market can explain. The Supreme Court struck down electoral bonds on 15 February 2024, and the State Bank of India data the Court ordered published showed more than Rs 16,000 crore had flowed through them since 2018; that one spreadsheet told the country how closely business and politics sit. That is the context any honest page about political consulting has to admit. Elections are not a side show to the economy. In a state, they are the economy's quarterly results.
Which is why businesses also buy political consulting, quietly. Not to win seats. To read the room: which way a state is leaning, what a manifesto commits a government to, which district will get the industrial corridor, and when to bid. We do not lobby. We do not fix. We can tell a business what the public data says about where a state is going, and we will put that in writing.
Consultancies publish their own prices now, which is new. One Indian firm's public rate card (thinkpolitically.com, read October 2026) runs from Rs 25,000 plus tax for a constituency visit and audit, to about Rs 4 lakh for a corporation or local-body seat, to Rs 16 lakh plus tax for a full-service assembly engagement, with top-tier Lok Sabha work quoted at Rs 50 lakh and above per constituency. A Delhi consultancy lists itself on IndiaMART at Rs 1,00,000 a month. The Association for Democratic Reforms has put the cost of simply maintaining a constituency between elections at Rs 3 lakh to Rs 4 lakh a month (ADR, 2023, as cited by the same rate card). At the top of the market, one figure is on the public record: the YSR Congress Party's expenditure statement filed with the Election Commission showed Rs 37.57 crore paid to I-PAC for consultancy out of Rs 85 crore spent on the 2019 Andhra Pradesh campaign (The News Minute, 2019). The spread is not dishonesty. It is the difference between a spreadsheet and several hundred people on the ground for a year.
And read the ceiling properly. Section 77 counts a candidate's spend from the date of nomination to the declaration of the result. The survey done ten months earlier is not in that window. The consultancy's fee usually sits with the party or an organisation, not the candidate. What is spent inside the window and attributable to the candidate must sit in the register, to the rupee, because an expenditure observer will read it against the platforms' own reports. The firms that know this say so in the engagement letter. The ones that do not are the ones that end up in the news.
What you are paying for, in order of what it is worth: the booth map, the message test, the field discipline, the dashboard, the compliance. What you are usually charged most for: the rally. Buy accordingly.
The Model Code of Conduct is a code, not a statute, but the laws underneath it have teeth. Section 77 of the Representation of the People Act caps candidate spending and the Election Commission audits the ledger. Section 126 names television and similar apparatus for the 48-hour silence before the close of poll; the Commission's instructions and the platforms' 2019 Voluntary Code extend it to paid digital, and we switch everything off 48 hours before the close of poll regardless. Political advertisements on television, cable and, by the Commission's direction, on social platforms must be pre-certified by the Media Certification and Monitoring Committee. Candidates declare their social media handles at nomination. Google and Meta require advertiser verification for election ads in India and publish the spend. Surrogate advertising through meme pages and unverified handles is documented and monitored by the Commission and the platforms. And WhatsApp, the channel every campaign runs on, is closed to campaigns at the platform level: Meta's WhatsApp Business policy prohibits political parties, politicians, candidates and campaigns, and the firms that serve them, from using the WhatsApp Business Platform. Any vendor selling a candidate "official API WhatsApp outreach" is selling a number that will be banned. We do not build on it, and we say so before the first call.
None of this is a reason to go quiet online. It is the reason to have someone on the campaign who has actually read it.
The last reel, and the one that matters to a sitting MLA. Elections are won in the four years between them. A legislator whose constituency can find them on Google, whose office answers a helpline within the hour, whose grievance register is a database rather than a diary, whose work is on YouTube in the local language before the opposition's version is, does not need a miracle in the last month. They need a counting machine, running quietly, all year. That is the part of political consulting that is simply good operations. It is also the part we are best at, because it is the same machinery we build for businesses: search presence, a Business Profile that answers, a service desk, a CRM, a dashboard, content that keeps coming. The politics is yours. The plumbing is ours.
We are in Bengaluru, so one local reel. The city has not elected a council since the old corporation's term lapsed in September 2020. The Supreme Court has set 31 December 2026 as the deadline for polls to the five Greater Bengaluru Authority corporations, after a special revision of the rolls that runs to November, and has said there will be no further extension. That is hundreds of wards, thousands of aspirants, and a campaign window measured in weeks, in a city where turnout routinely trails the state by ten points. The Karnataka assembly follows in 2028 and Parliament in 2029. Ward polls are run by the Karnataka State Election Commission under its own expenditure rules, so the Election Commission of India caps quoted above do not apply to a ward candidate; the discipline does. A ward campaign is an assembly campaign in miniature: the same map, the same three numbers, a smaller roll and far less time. The people who start now with the map and the helpline will spend December counting. The rest will spend it printing.
We are not the general. We are the digital, data and AI layer the general should have had a year ago. Each item is a thing we have built for businesses and can build for a constituency.
Form 20 booth results, roll growth by part, census and scheme data on one screen, with the 60 booths that decide the seat marked and reasoned. Public data only, assembled properly, updated as new rolls publish.
Three messages, small budgets, real audiences in the constituency, measured by saves, shares and replies rather than applause. The campaign prints the one that moved people, not the one the room liked.
A phone line, missed-call number, SMS and web form into one grievance CRM: logged, routed to the office, acknowledged within the hour, closed with a photo. Consent recorded. Not built on WhatsApp, because Meta's WhatsApp Business policy bars political campaigns from the Business Platform, and we will not put a campaign on a number that gets banned.
Search, Business Profile, YouTube and short video in the local language, built so the constituency finds the legislator's own account of the work before it finds anyone else's. Runs between elections, when it costs least and matters most.
Homes visited, by booth, by day. Complaints opened and closed. Ad spend and reach by ward. Volunteer activity. One screen the candidate reads at 10 pm. If Tuesday is not counted, Tuesday did not happen.
Ad accounts owned by the candidate or party with Apex as manager, so the "Paid for by" disclaimer, the verification and the expenditure trail all sit in the right name. Google and Meta advertiser verification, Media Certification and Monitoring Committee submissions, the declared-handle list, the 48-hour switch-off, and an expenditure file that reconciles to the platforms' own reports.
We publish political work only with written permission, and none is published yet. We will not invent a case. What we can show is the machinery, already running, for businesses.
The free tools on /free-tools, from a schema validator to a malware scanner, each built in-house and open to anyone without a login. Run one on your own site before you call us. The same hands build the constituency desk.
Indraa Kumar D, on his own page, with the method written down. You will know whose judgement you are buying before the first call.
The ads practice and AI automation pages describe systems that run today for Indian businesses, on the official WhatsApp Business API, with the compliance built in.
Three kinds of people land on this page. We can help two of them.
An aspirant or a sitting legislator, assembly or municipal, who wants the digital, data and operational layer of the campaign built by people who do this for a living, under a strategist of their own choosing or under their own political judgement. You bring the politics. We bring the counting machine. Yes.
A political consulting firm that has the strategy and the ground but not the technical bench: the dashboard, the grievance CRM, the ad compliance, the constituency data on one screen. A white-label partner that will not poach your client. Yes.
Anyone who wants fake accounts, bought followers, surrogate pages, scraped voter data, paid trolls, deepfakes, or a promise that the seat is won. No. Not at any fee. The refusals below are not a marketing line. They are the engagement letter.
Refusals cost nothing to verify. These are the three we hold to on every engagement.
Public election data is public. Scraped rolls with phone numbers, bought databases and any list a voter did not consent to are not. We work from Form 20, the published roll and opt-in lines, and nothing else.
No bot networks, no bought followers, no surrogate meme pages, no unverified ad accounts, no unlabelled AI content and no synthetic video of anyone. If it would embarrass the candidate on the front page, we do not do it in the back room.
We will tell you the number, the booths and what the data says. We will not tell you that you have won, and we will not write a line that asks for a vote on religion, caste or language, which Section 123 of the Representation of the People Act treats as a corrupt practice.
The same discipline we bring to a business. A scoping call, a fixed fee in writing, and a first deliverable inside the first month.
Which seat, which cycle, who holds the strategy, what exists already. We say on the call whether we are the right fit and what we would build first. One candidate per constituency per cycle, confirmed in writing before any work starts.
Public data assembled on one screen with the decisive booths marked. This is delivered first because every other decision depends on it.
Helpline and grievance CRM, dashboard, ad accounts verified, presence built, the message test run. Each piece demonstrated working before the next is started.
Daily counting through the campaign, the compliance file kept current, and the 48-hour switch-off executed to the minute. After the count, the legislator keeps the machine.
A fixed fee in writing after the scoping call. We do not publish a political price because no two seats are the same job and because the fee is a line in an expenditure file that someone will read.
The first step is small and named: a 14-day constituency digital readiness check, which reads the seat's public data, the candidate's search presence and the state of the ad accounts, and sets up the official channels in the right names. It ends with a written scope and a fixed fee for whatever comes next, or with nothing, if nothing is needed. The market bands are above in chapter five, published by firms that list them. Our scope is narrower than a full-service consultancy and priced accordingly: we build and run the digital, data and AI layer; we do not field six hundred people. Where a consultancy engages us as its technical partner, the fee is to the consultancy and the client relationship stays theirs.
Two things are always true of our quote. It is a fixed fee for a written scope, so the campaign knows its number on day one. And it is invoiced to the entity that can lawfully bear it, with the work inside the Section 77 window itemised separately from the work before it, which is the paperwork an expenditure observer would expect. If that sounds dull, good. Dull is what compliance looks like.
Every figure above traces to a public source. Read them before you quote us.
Election Commission of India: candidate expenditure limits revised January 2022 (Rs 40 lakh and Rs 28 lakh for assembly seats, Rs 95 lakh and Rs 75 lakh for Lok Sabha seats, by state size); Form 20 booth-wise results are published by each state Chief Electoral Officer. Centre for Media Studies estimate of Rs 1.35 lakh crore for the 2024 Lok Sabha election. thinkpolitically.com rate card and its ADR citations, read October 2026. IndiaMART listing at Rs 1,00,000 a month. WhatsApp Business policy, section on government and political use. Supreme Court deadline of 31 December 2026 for the Greater Bengaluru Authority polls. Union Budget 2025-26 capital expenditure and the GeM annual figure are from the Government of India's own releases. Reviewed 10 October 2026.
A candidate, a legislator's office or a consultancy. Send the constituency and the cycle and we will reply within 24 hours with the first thing we would put on one screen. No bot, no pressure. Your name and constituency stay on our own server; we never publish, list or reference political clients.
No spam. No calls without consent. Just a clear plan.