The Coach Selection Index — India
Every "top coach" list in this country is written from the seller's side. This one is written from yours.
You tell it two things — who is paying and at what leadership level — and it scores the coaches you are considering out of 100, across seven pillars and twenty-one indicators, every point of which has to be backed by evidence you can open.
The weights are not our opinion. They come from what 38 corporate sponsors and 54 coachees in South and Southeast Asia said they actually select on, published cell by cell with its provenance. The arithmetic runs in your browser, not in a language model, so the same inputs give the same number on Claude, ChatGPT, Gemini, or a sheet of paper.
Read the full methodology · machine-readable spec · reference implementation · the AI extraction prompt
Who is buying, and for whom
These two answers select one of twelve published weight vectors. You can see exactly what they do to the weights before you score anybody — which is the whole point.
If your brief is unusual — a turnaround, a first-generation promoter, a leader moving country — drag the sliders and the model rebuilds and re-ranks live. Your slider positions are exported with your result, so a tuned run is just as checkable by somebody else as the published one.
The weights this produces
Named coaches, scored on public evidence
Find them, then ask them
This index does not hold a list of coaches, and that is deliberate — a directory would rank whoever paid to be in it. You assemble the candidates; it scores them. These two generators do the work you would otherwise do by hand, and both are built from the weights you just set.
A · Build the widest list you can stand
One click. The search runs here, grounded in real web results, and returns named coaches with their publicly available details — city, years practising, credential, sector focus, and a source link for each. It is shaped by what you set above: your level, your sector, your city, and the three pillars that weigh most on your brief.
Free to run and free to use — there is no paid tier of this tool and no coach can pay to appear. Anything the search cannot source it leaves blank rather than guessing, and nothing it returns is scored: a candidate arrives on your scorecard blank, and every point is earned from what the coach actually sends you.
What comes back is a candidate list, not a ranking. Nothing here can tell you which coach is better, because nothing here has seen the evidence — that is what step B and the scorecard are for. Treat the names as a starting set and nothing more.
Then widen it. Whatever a search returns is already shaped by who publishes most, and sponsors rank visibility eighth of nine at 2.6%. Add names from your own network, your HR business partner, the ICF or EMCC chapter for your city, and anyone a colleague has actually worked with. Six candidates beat three. The instrument cannot reward a coach it is never shown.
B · Ask each of them for the evidence
This is the step most buyers skip, and it is the one that decides the hire. The letter asks for exactly the twenty-one things this index scores, with the pillars that weigh most for your brief marked as priorities. Send it to every candidate on the same day and give them the same deadline.
Then bring what comes back to step three. Paste each coach's reply into any AI along with the extraction prompt and use paste AI codes, or set the sliders yourself — it takes about four minutes per coach once the evidence is in front of you.
Score the coaches in front of you
Add each shortlisted coach and work through the seven pillars. For every indicator you set a code (what the evidence shows) and an evidence tier (how well you can prove it). A claim with nothing behind it scores 70% of a documented one; a claim with no evidence at all scores zero.
Short of time? Paste the coach's evidence pack into any AI along with the extraction prompt, then use paste AI codes to fill the form in one go. The AI only reads and codes — it never scores.
The result
Scores update as you type. Where two coaches' confidence intervals overlap they share a position — this instrument will not invent a difference the evidence cannot support.
Bands, gates and the confidence interval
Bands
| Band | CSI | What it means for you |
|---|---|---|
| A | 80.0 + | Documented, contextually matched, measurable. Hire with confidence. |
| B | 70.0 – 79.9 | Strong. Check the weakest pillar before you sign. |
| C | 60.0 – 69.9 | Workable, with gaps you should name in the contract. |
| D | 50.0 – 59.9 | Thin. You are buying on faith somewhere. |
| E | below 50.0 | Do not proceed on this evidence. |
The five gates
A weighted sum alone lets a coach buy the top of the table on cheap dimensions. These five screens run first, and two of them are absolute.
| Gate | Rule | Consequence |
|---|
Why some scores carry a wide interval
κ (kappa) is the share of your evidence that is verified or corroborated rather than merely claimed. The interval is ±(1 − κ) × 12. A coach documented at every point gets a precise number; a coach who is mostly assertion gets a wide band — and you can see at a glance that the number is soft.
What this does not do
It scores the coaches you bring it. It says nothing about the ones who never reached your shortlist — and wherever those names came from, that source has already applied its own bias before the first indicator is coded. Search ranks who markets best; this ranks who can produce receipts, but it can only rank who it is shown. So widen the list before you score it. On the evidence in the underlying study, the coaches you are most likely to be missing are the ones sponsors would rank highest and every discovery channel ranks lowest: visibility is worth 2.6% to a buyer and close to everything to a search engine.
The Engagement Readiness Check
The largest single predictor of coaching success in the underlying study is not the coach at all. Coachee readiness was cited by 67.6% of sponsors and 57.4% of coachees; organisational culture by 35.1%; manager support by 21.6%. None of these is an attribute of the coach, so none of them is scored against a coach here.
They are still the difference between a good hire and a wasted year. Answer these five honestly before you commission anyone.
- 1Is the coachee choosing this, or receiving it? Coaching delivered as a verdict rarely survives session four.
- 2Can you write the goal in one sentence a stakeholder would recognise? If not, no coach can hit it.
- 3Has the line manager agreed to a three-way contracting conversation? Clear goals agreed with the sponsor tied for the top success factor at 67.6%.
- 4What will you measure, and when did you agree it? After the engagement starts is too late.
- 5Will the organisation tolerate the behaviour change it is paying for? Systems defeat intentions. Diagnose the structure, not the person.
Fewer than four clear yeses and the problem is not your shortlist. Changing coaches will not fix it.
How to be discovered without spending a rupee
You have the credential. You have the years. You are good in the room. And almost nobody finds you — so you post daily, boost a few, refresh the site, try a directory, hire an SEO consultant. Each works for about six weeks. None compounds. And the coaches who are visible do not seem better than you, which is the part that stings, because it is usually true.
By the time a serious buyer is choosing between three coaches, visibility has already done its only job — getting you into the room. Every question after that is evidentiary: can you show it? Most coaches, at that point, cannot.
What changed, and why it favours you
Buyers used to search. Increasingly they ask — ChatGPT, Gemini, Claude, Perplexity, Google's AI overview. That is a different game, and the rules are unusually kind to a coach with substance and no budget:
- 1You cannot buy the answer. There is no ad slot in a synthesised response.
- 2Volume stops working. More of the same thin content is actively penalised now, not rewarded.
- 3Specifics get cited; adjectives get dropped. "1,900 logged hours, ICF register" survives summarisation. "Highly experienced transformational coach" does not.
- 4Consistency decides whether you exist. Answer engines resolve entities. A coach whose site, LinkedIn, credential register and book jacket disagree about their own hours is not a trusted entity. They are noise.
- 5Corroboration beats volume. One dated third-party mention outweighs a month of your own posts.
Now read that list against the evidence tiers this index scores you on — T1 verified, T2 corroborated, T3 your own word for it, T4 nothing. They are the same thing. A serious buyer and an answer engine are asking one identical question: can this be checked? Which means it is one piece of work, not two.
The honest catch
This will not make you discoverable next week, and anyone promising that is selling you the last decade's game. It makes you checkable, so that when the evidentiary question arrives you are one of the few with an answer. It also cannot manufacture substance you do not have: if you have 200 hours, publishing that you have 200 hours is honest and will score as 200 hours. The index is a mirror, not a marketing tool with extra steps. Several items will be things you cannot claim yet — those are your development plan, not your copywriting brief.
The full checklist — all 21 items, what each is scored on, and what each does for an answer engine → Free, complete, and generated from the same specification buyers score you against.
What we charge, and what we will never charge for
The checklist is published free and complete. There is no withheld tier and no private advice. Follow it yourself and your score is identical to the score of a coach who paid somebody. Nirvedha does sell coaches help implementing it — assembling an evidence pack, writing a method document, fixing entity consistency. That is labour, and it is fenced:
- ·Paying cannot move a code. Any coach who is, or was within 24 months, a paying client has their register entry coded by the independent adjudicator — never by us.
- ·The relationship is published on the entry, displayed, not footnoted.
- ·No fee, ever, for listing, scoring, re-scoring, correction or removal. Corrections are free permanently. The moment position can be bought, this becomes the search engine it was built to replace.
Selling labour anyone could perform themselves, from instructions anyone can read, is a service. Selling access to information that determines rank is a racket. The difference is whether the instructions are public — so they are, above, in full.
The author takes his own test, and takes his place in it
This index is published by Nirvedha Executive Coaching Solutions, whose founder is himself an executive coach competing in this market. That is a conflict. It is managed rather than disclosed and ignored — and the management is not abstention. An author absent from his own table has avoided the test, not passed it.
So score him, and rank him. CSI-IN has no idea whose name is in the box. Put Sudhakar Reddy Gade through all twenty-one indicators exactly as you would anyone else. Nirvedha's own scorecard is published in full — every code, every tier, and the evidence pack behind it — and the commitment to publish it was made before the result was known. A mid-table finish by the author would be the most credible thing that could happen to this instrument.
Two things make that safe to offer. He does not code his own entry — an independent adjudicator sets Nirvedha's indicators from the same evidence pack any other coach submits, and the entry says so. And you can re-rank the whole table against your own weights in ten seconds, with the sliders above. If you suspect the weights were shaped to flatter their author, that is a suspicion you can settle yourself, on your own screen, without asking anyone's permission. A ranking nobody can recompute needs its author removed from it. This one does not.
No coach can pay to be listed, pay to rank, pay to be removed, or buy advertising against it. Any coach may inspect their own codes and evidence citations and submit documented corrections. The specification, the weights, the code and the conformance fixtures are published under CC BY 4.0 — fork it, criticise it, publish a better one.
Read §12, Governance and conflict of interest · §13, Limitations — including the small sponsor sample, the two imputed cells, and the fact that the leadership-level tilt is declared judgement rather than survey data.
Nothing you type here leaves your browser
There is no server, no account and no transmission. Your working is kept in this browser's local storage so you can come back to it, and clear everything removes it. Coach names you enter are yours alone — we never see them.