Help Center
Use Capital Almanac with the evidence in view.
Find practical guidance for search, companies, public and private markets, Capital AI, screening, market maps, monitoring, APIs and team rollout. Each article explains the research judgment around a feature as well as the product behavior.
12 guides
Getting started
Plan the first research workflow
Start with a specific job such as company screening, diligence, public-company review or market mapping and identify the sources that job requires. A focused first workflow makes it easier to verify coverage, permissions and evidence behavior before the team expands into additional research surfaces.
Open related surface →Search
Search without losing source context
Use search to find the entity, document or research theme you need, then inspect the source context behind material facts instead of treating the result list as the conclusion. When the connected source universe does not support a query, Capital Almanac should show an unavailable state rather than seed the answer with invented research.
Open related surface →Companies
Read a company profile carefully
Review reporting periods, freshness and source attribution alongside financial or company fields so values are interpreted in the right context. For private companies in particular, an unavailable field should not be read as zero or assumed to match an estimate from another source.
Open related surface →Public markets
Check quote and financial freshness
Market observations can have different freshness depending on provider rights and exchange coverage, so confirm whether a value is real time, delayed or end of day before using it. Financial statements and earnings data should also be read with their reporting period and source attached.
Open related surface →Private markets
Work with incomplete private-company data
Private-company records are naturally less complete than public filings because disclosure varies and many terms are not public. Keep unknown values unknown, use sourced funding or transaction records where available and avoid turning gaps into implied valuations, ownership or performance claims.
Open related surface →Capital AI
Review an AI-assisted answer
Treat Capital AI as a way to find and organize relevant evidence, not as an authority that can create facts absent from the source set. For a material conclusion, inspect the cited company, document, reporting period or transaction context and narrow the claim if the evidence does not support the broader answer.
Open related surface →Screening
Build a defensible company screen
Choose filters that correspond to fields supported by the connected data universe and document the criteria that matter to the research question. A screen is easier to defend when a teammate can understand why a company was included and which source supports each key filter.
Open related surface →Market maps
Turn a sourced universe into a market map
Begin with a company universe built from explicit screening or research criteria, then use relationships and categories supported by the underlying records. Do not add companies, investors or edges merely to make a visual map look complete when no legitimate source supports the relationship.
Open related surface →Watchlists and alerts
Monitor change without confusing it with meaning
A watchlist or alert can tell you that a tracked record changed or that a configured condition was met, but it does not explain why the event matters. Open the underlying evidence, compare it with prior context and decide whether the change is relevant to the thesis or workflow you are following.
Open related surface →API
Handle unavailable values downstream
Applications using the Capital Almanac API should distinguish a legitimate zero from a null, missing or unavailable field. Preserve source and freshness information where the endpoint provides it so users of the downstream system can understand the limits of the data they are seeing.
Open related surface →Methodology
Understand the evidence model
Capital Almanac is designed to retain provenance and communicate when a field is unsupported instead of replacing it with a fabricated fallback. Use the methodology page when you need to understand how sources, normalization, missing values and market-data freshness are expected to behave.
Open related surface →Enterprise
Prepare for a team rollout
Define the research use cases, user groups, source coverage, integrations and identity requirements that matter before access is provisioned broadly. Security, SSO, API and administrative capabilities should be confirmed for the planned deployment rather than inferred from general marketing language.
Open related surface →More guidance