Reuters

What Reuters Actually Is and How to Use It

Reuters is a news agency. That's the simple version. In practice, it's one of the three wire services that finance and media organizations rely on for real-time information. The other two are Bloomberg and AP. If you work in any field where timely factual reporting matters — trading, research, journalism, compliance — Reuters feeds are basically infrastructure. You don't notice them until they're gone. Most people encounter Reuters through the consumer-facing website or app. That's not what this is about. This is about the professional side: Reuters Eikon, the Refinitiv platform, and how to actually extract useful data from it without wasting half your morning.

Accessing Reuters Professional Data

You need a subscription. The consumer site is free but it's lagged content with ads. The professional terminal costs roughly $25,000 to $30,000 a year per seat. If your organization already pays for it, get on the Refinitiv Learning platform and spend an afternoon on the basic modules before you start asking colleagues for help. It'll save everyone time. The platform rebranded from Thomson Reuters to Refinitiv after the spinoff, then LSEG acquired Refinitiv in 2023. The UI still shows "Eikon" in a lot of places because nobody moved fast enough to rebrand everything. Don't let that confuse you. It's the same product.

How to Pull Historical Data Efficiently

The most common task is grabbing historical price or fundamental data for a list of securities. There are several ways to do this and most people pick the slowest one. Using the Excel add-in is fine for small requests. It's terrible for anything beyond a few hundred instruments. I once tried to pull five years of daily close data for about 2,000 European equities through the Excel plugin. It took 47 minutes and crashed twice. Switched to the RDP (Refinitiv Data Platform) REST API and the same request ran in under three minutes. The key is using RICs (Reuters Instrument Codes) correctly. A RIC like "AAPL.O" or "0DOV.L" is the standard identifier. Don't try to search by company name in bulk — the search endpoint is slow and returns ambiguous results. Build a mapping table first. I keep a local CSV that maps ticker symbols, ISINs, and RICs for the universes I track. When I need to pull data, I already have the correct RICs. That one step cut my typical workflow from 90 minutes to about 20. For Python users, the refinitiv.data package (formerly eikon) is the standard approach. The function rd.get_data() handles bulk requests. The important detail most tutorials skip: set raw_output=True if you're pulling lots of fields. Otherwise the library does extra transformation work that slows things down noticeably.

Common Pitfalls That Waste Time

Here's what nobody tells you about Reuters data: it's not always consistent across asset classes. Currency conversion factors, corporate action adjustments, and holiday calendars vary by market. I learned this the hard way when backtesting a cross-asset strategy. The EUR-denominated bond data had different trading day conventions than the USD equity data. My model assumed both used NYSE business days. It was off by several days per quarter and the discrepancies compounded. The fix was straightforward once I knew what to look for. Every dataset in RDP has metadata fields. Check DATASOURCE and CALENDAR parameters for each instrument type. Cross-reference the settlement conventions before you trust any backtest result that mixes asset classes. Another issue: real-time data has a subscription tier attached to it. If you're on a delayed or snapshot-only plan, requests for live quotes will either fail silently or return stale data. I once spent two hours debugging what I thought was a code error before realizing our subscription tier didn't include real-time US equities. The RDP dashboard shows your available permissions under Account Settings. Check it before you write a single line of code.

Reuters in News Research

If you're using Reuters for news rather than market data, the News API through RDP is the professional route. The consumer search is adequate for casual use but it lacks the filtering precision you need for compliance or research workflows. The News API supports full-text search, date range filtering, and topic categorization. The tricky part is understanding how Reuters indexes stories. A single event often generates multiple articles — an initial breaking news piece, a follow-up analysis, a corrected version. If you're counting articles or tracking sentiment over time, deduplicate by using the firstCreated timestamp and the unique uniqueName field. Without deduplication, your counts will be artificially inflated. I handle this by querying for the earliest article per event cluster and then pulling follow-ups separately if needed. It adds a step but it's the difference between having usable data and having to manually clean a messy export.

When Reuters Isn't the Right Tool

Be honest about what Reuters does poorly. It's not great for private company data. Coverage is thin outside of publicly traded entities and a narrow set of large private firms. If you need information on small private businesses, you're better off with Dun & Bradstreet or local registry sources. Cryptocurrency data on Reuters is also limited compared to dedicated crypto data providers. You'll get prices for major coins but the depth, order book data, and on-chain metrics aren't there. For anything crypto-related, use a specialized source alongside Reuters rather than expecting it to cover the gap. Emerging market data quality varies. Some markets have sparse coverage or significant reporting lags. Before relying on Reuters data for an emerging market position, verify the coverage depth against the local exchange or regulator directly. I made the mistake of trusting Reuters coverage for a small-cap Brazilian stock during an earnings window. The data was three days behind the local ticker. By the time the Reuters article appeared, the market had already moved.

Practical Setup Recommendations

Start with a clean environment. Set up your RDP credentials once using rd.set_session() in Python and store them securely. Don't hardcode them in scripts. Use environment variables or a credentials manager. Batch your requests. The API has rate limits and individual requests for hundreds of instruments will hit them. Break large pulls into groups of 100-200 and add a brief delay between batches. It's slower but it won't get your session throttled. Cache your results locally. Market data changes predictably — end-of-day snapshots don't refresh until the next trading session. Store what you've already pulled and only request updates. I use a SQLite database keyed by RIC and date. Most of my queries hit the cache and only miss on fresh data that I haven't seen yet. This usually reduces actual API calls by 80 percent or more on routine workflows. The Reuters ecosystem is functional but it rewards people who understand its quirks. The documentation is thorough but scattered across multiple pages that don't always reference each other. The data itself is reliable when you know what you're looking at. Spend the time upfront learning the metadata fields and subscription limits, and everything else becomes much simpler.