Comparable search
Every comparable from the right database, with its reason.
India searches Prowess and every other jurisdiction searches Orbis, and a search never swaps one for the other. Each company accepted or rejected keeps a coded reason that survives the next run.
The routing
The jurisdiction picks the database.
Decided in one place in the code, so no screen can send an India study anywhere else.
- IndiaProwess (CMIE)
- UAETP Catalyst®
- GermanyTP Catalyst®
- SingaporeTP Catalyst®
- United KingdomTP Catalyst®
No fallback
An India study never gets foreign comparables.
Prowess and Orbis are separate pipelines with their own vendor, tables and sync. If Prowess cannot answer, an India search stops and says so rather than quietly filling the set from Orbis.
India only
Prowess (CMIE)
Prowess — Centre for Monitoring Indian Economy (CMIE)
- Companies
- 112,189 Indian companies
- Figures
- Revenue in ₹ crore
Every other jurisdiction
TP Catalyst®
TP Catalyst — Bureau van Dijk (a Moody’s Analytics company)
- Companies
- 31.6M active companies
- Figures
- Revenue in thousand USD
When Prowess cannot answer, an India search says
India comparable search is temporarily unavailable (Prowess). Please retry — Indian results are never substituted with non-Indian sources.
Where to start
Begin from last year, not a blank page.
A search can start from scratch, from the prior-year TP study your client uploads, or from a study already in your vault.
- A new searchIndustry, size and screens against the database for the study’s jurisdiction.
- Last year’s TP studyUpload it with the IRD and its search strategy is read and offered as the starting point.
- A study from the vaultEdit it or copy it, choosing the new financial year and PLI up front.
Count first
The count you see is the set you get.
Step 1 counts the companies that match your strategy before anything runs. The run starts from that same set, and the two are held equal. Outside India, each study also names the TP Catalyst release it searches.
- Each Orbis company records the database update it came from.
Step 1 · Count
4,812
Companies matching the strategy
Step 2 · Run
4,812
Companies the funnel starts from
Reasons that stay
A decision belongs to the company, not the run.
Every rejection carries one of the coded grounds, with the evidence it rests on. Overturn one and your reason is required, then applied again each time the search is re-run.
- The current run and the one before it are both kept.
- The Accept/Reject Matrix is built from the same records.
Run 1 · the funnel’s decision
Vantor Plant Services Ltd
Rejected: Functions, assets and risks differ
Ground:FAR-MISMATCH
Described as taking contract risk on fixed-price installation projects. The tested party is paid on cost and carries no project risk.
Accepted by practitionerReason required
Note 21 shows 94% of revenue is cost-reimbursed maintenance; the fixed-price work was one contract that ended in FY 2023-24.
Run 2 · the decision is kept with the company and applied again.
Coverage
What it covers.
- Orbis, outside India
- 31.6M
- Active companies
- Prowess, India
- 112,189
- Indian companies
- Routing
- India to Prowess
- Never to Orbis as a fallback
- Rejection grounds
- 16 codes
- Each with its evidence source
- Runs kept
- Current and previous
- Overrides carry into each re-run
- Database release
- Chosen per study
- TP Catalyst studies
Questions
Before you ask.
- What happens if an Indian search cannot use Prowess?
- The search stops and says so. An Indian tested party is never benchmarked against Orbis companies as a fallback.
- Can I overturn a decision the funnel made?
- Yes. Accept or reject any company with a reason. The reason is stored against the company and carries forward when the search is run again.
- Which database release did a study use?
- Outside India, each study records the TP Catalyst release it searched. Changing the release after a run marks the results out of date.
- Does the number before the run match the result?
- Yes. The Step 1 count and the set the run starts from are the same companies. The funnel then narrows that set, with a count at every filter.
