Revenue intelligence platform: the three data sources it scores, and the one to fix before you buy

What a revenue intelligence platform actually does

A revenue intelligence platform is software that captures what happens in your deals (emails, meetings, calls, CRM changes), scores each deal on its chance to close and turns the result into a forecast and pipeline views for sales and RevOps. The short definition of revenue intelligence lives in our glossary. This guide is about what the platform scores, which of those inputs you can fix yourself before you buy one, and the questions to ask in the demo.

Sales and RevOps teams buy revenue intelligence software for one reason: the forecast in the CRM is built on what reps type, and reps type late, optimistically or not at all. The platform replaces part of that typing with captured activity. The part it cannot replace is the data about the account and the people in it, and that is where revenue intelligence projects can quietly lose accuracy.

Activity capture

The platform connects to email, calendar and the phone system and logs every interaction against the right deal, without the rep entering anything. This is the foundation: every other function reads it.

Forecasting

From activity and deal fields, the platform predicts which deals will close this period and rolls them up into a forecast by rep, team and region, often next to the forecast the reps submit, so managers see where the two disagree.

Deal scoring and pipeline visibility

Each deal gets a health score: engagement is rising or falling, the next step is dated or not, the right people are involved or not. Pipeline views show which deals moved, stalled or slipped since last week. The stages themselves come from your CRM; our guide on sales pipeline stages covers how to define them.

Buying group visibility

The platform shows which people at the account are engaged in the deal, and flags deals where only one contact is active. This is the function that depends most on data outside your team's activity: the platform can only show the committee members it knows exist.

Who uses revenue intelligence, and for what

The platform has four kinds of users, and each reads a different part of it. Knowing who will open it every week tells you which features matter in the demo.

• The CRO or head of sales reads the forecast: the committed number, the gap to target, and which deals explain the gap. Revenue intelligence gives them a second forecast, built from activity, next to the one the managers submit.
• Sales managers read deal health and pipeline changes before the weekly review: which deals slipped, which have gone quiet, which depend on a single contact. The platform replaces an hour of asking reps for updates with ten minutes of reading.
• RevOps owns the setup and the inputs: which activity counts, how stages map, which fields the scores use. They are also the first to see when a score is wrong because a record is stale, which makes them the natural owner of the account and contact layer.
• Reps use it least, and mostly to avoid typing: activity is logged for them, and the next-step reminders come from the platform. If reps get nothing back from it, adoption drops and the activity data thins out.

The common thread: every user trusts revenue intelligence exactly as far as its scores match what they know about their deals. The first time a score contradicts reality because the champion left months ago, that trust is spent.

Revenue intelligence vs CRM reporting vs sales intelligence vs conversation intelligence

Four categories overlap in vendor messaging. The table separates them by what each one captures and the question it answers.

Revenue intelligence and sales intelligence are complementary rather than competing: one reads your deals, the other describes the accounts behind them. The sales intelligence category as a whole is covered in our guide to sales intelligence tools.

The three data sources every revenue intelligence platform scores

Every one of these tools scores three kinds of input. They do not fail the same way, and only one of them can be repaired in a spreadsheet before you sign.

The third source is the one you can fix on your own, list by list, before paying for a single seat. With Derrick, each fact has its own feature:

Enrich Companies, Enrich Leads and Find a company's people read LinkedIn, so they need your LinkedIn account connected through the Derrick Chrome extension (Find a company's people does not need Sales Navigator, but it does need that connection). Find Duplicates runs in the Google Sheets sidebar.

What the platform scores wrong when the CRM is stale

A deal score is only as current as the facts under it. When the champion left the company three months ago, the platform sees silence and reads it as a stalled deal, when the deal is in fact dead. When the account was acquired or cut its headcount in half, the platform still scores it against a budget that no longer exists. When the champion's address bounces, the activity that should feed the score never arrives. B2B contact data decays at roughly 2.1% a month, about 22.5% a year, according to our CRM data quality report, and the deals in your pipeline are not exempt.

The forecast inherits the same problem. A forecast built on deal scores adds up every stale fact in the pipeline: a quarter's worth of deals whose champions moved on can look like a healthy commit until the close dates pass. The fix is not a better model. It is checking the accounts and contacts attached to open deals before the platform scores them, and again every quarter, so the forecast reads the company as it is today. Our guide to contact management covers how to keep those records current once the cleanup is done.

Revenue intelligence platforms compared, and the data layer under them

We do not sell a revenue intelligence platform, and we do not test them against each other, so this is not a ranking. It is a map of what the main platforms are built around, as described by their vendors, followed by the layer every one of them depends on. Check pricing and features with each vendor: both change often.

Many activity-capture platforms create contacts in your CRM from email signatures and meeting invites; ask each vendor how those contacts are verified and refreshed, because the deal scores depend on them.

The layer under all of them: account and contact data

Every platform above scores deals using facts about the account and the people in it. None of them promises that those facts are current. This is where Derrick sits: not as a revenue intelligence platform, but as the tool that fills and verifies the account and contact layer before and after you deploy one. Enrich the accounts on open deals, check that champions still hold their role, list the rest of the committee and verify the addresses, in the web app, in the Google Sheets sidebar, from Claude through the MCP or through the API. Whatever platform you pick, it scores better on data that is true today.

How to choose a revenue intelligence platform: six criteria

1. Where your deals happen. If most of the selling happens on recorded calls, a conversation-first platform reads more of it. If it happens by email and in meetings, activity capture matters more than transcripts.
2. Your CRM. Some platforms are built for one CRM and read it deeply; others connect to several. Check how the platform writes back to your CRM, not only how it reads from it.
3. The inputs it scores. Ask for the list of fields behind the deal score. The more it relies on account and contact fields, the more your data quality decides its accuracy.
4. How it handles people changing jobs. A platform that notices a champion's departure only through silence will score dead deals as stalled ones for weeks.
5. The trial on your own pipeline. A platform that cannot show scores on your real deals during the trial is asking you to buy on sample data.
6. The work before go-live. Every vendor has a list of what to clean first. Get it in writing, estimate it, and compare it across vendors: it is part of the price.

None of these criteria is about the model. Most revenue intelligence platforms use similar methods; what separates the results in your company is the quality of the three inputs, and how much of that quality the platform leaves to you.

5 questions to ask in the demo

Vendor demos run on clean sample data. These five questions bring the conversation back to yours.

1. "What happens to a deal score when the champion changes jobs?" Ask whether the platform detects it, or only notices the silence weeks later.
2. "Which account fields does the score use, and where do they come from?" If the answer is "your CRM", ask how the platform handles fields that are empty or years old.
3. "Can we see the score on our own deals during the trial?" A score demonstrated on sample data tells you nothing about your pipeline.
4. "How does the forecast treat a deal with only one engaged contact?" Single-threaded deals are the ones that slip; the answer shows how the platform reads buying groups.
5. "What do we need to clean before go-live, and who does it?" A vendor that says "nothing" has not looked at your CRM. Get the list in writing, with the owner of each item.

Is your CRM ready for a revenue intelligence platform?

Tick what is already true for your CRM. The check gives you a readiness score, what to fix before you sign, and what the data part of that fix costs in credits.

A worked example. 500 accounts with open deals, one champion each. Refreshing account size, industry and country costs 1 credit per company, so 500 in all; checking each champion's current role costs 1 credit per profile, 500 more; both features are on the free and paid plans. Verifying the champions' emails adds 1 credit per email on the paid plans, up to 500. Listing five committee members per account costs 1 credit per person, so 2,500, also on the paid plans. That makes 4,000 in total, which fits in the Standard plan at 20 euros a month (10,000 credits) with room left; the free plan's 100 credits a month cover the two free steps, company data and champion roles, on a sample of 50 accounts first.

Fill the account layer before you buy: the app, Sheets, Claude and the API

• The Derrick web app is the default for a list of accounts: export the accounts and champions attached to open deals from the CRM, import them, add a column per fact and send the result back. Nothing to install.
• The Google Sheets sidebar does the same inside a spreadsheet, and is where Find Duplicates runs. Derrick works from the sidebar, not from formulas.
• Claude or ChatGPT through the Derrick MCP fits the deal review: before the forecast call, ask whether the champion on your biggest deal still holds the role and whether the account is hiring. MCP comes with the Plus plan and above; LinkedIn operations need the Chrome extension connected.
• The Derrick API fits the continuous version: a workflow re-checks the account and champion every time a deal enters a late stage, and writes the result into the CRM record the platform reads. API access comes with the Plus plan and above, and works with Zapier, Make and n8n. Our guide on HubSpot enrichment workflows shows the pattern.

Implementation: what breaks in the first 90 days

• Adoption. Reps stop looking at the platform if the scores contradict what they know about their deals. The first time a score is wrong because of a stale fact, trust drops. Clean the open deals first.
• The CRM as input. The platform reads your stages, amounts and dates. If stages have no entry criteria, the platform learns from inconsistent labels and scores accordingly.
• Personal channels. Deals run on personal phones or messaging apps leave no activity. Decide early which channels count, and tell the team.
• Too many dashboards. Pick the two views managers will use in the weekly pipeline review and start there. Everything else can wait for month three.
• No owner for the data. Someone in RevOps has to own the account and contact layer after go-live, or the decay returns within two quarters.

Revenue intelligence metrics to track after go-live

Revenue intelligence software promises a better forecast. Measure that promise, and measure the inputs that decide it.

The first metric is the one the vendor will show you. The last three are the ones that explain it. When forecast accuracy stalls, look at the contact layer before blaming the model: revenue intelligence cannot score a relationship with someone who no longer works there.

When a revenue intelligence platform is not the next step

Revenue intelligence platforms pay off when there is enough pipeline to analyze and enough activity to capture: several reps, dozens of open deals, a sales cycle long enough for patterns to show. For a team of two founders selling to twenty accounts, a well-kept CRM and a weekly review do the same job. And for any team whose CRM is mostly empty fields, the next step is the data, not the platform. Fix the account and contact layer first; then the platform has something true to score.

There is also a middle path. Before you commit to a platform, run the readiness check above on your real pipeline, fix the account and contact layer for the deals that matter this quarter, and hold a manual weekly review against those refreshed records for a month. If the review already changes which deals you trust, the platform will amplify that. If it does not, the problem is upstream, in how deals are qualified and staged, and revenue intelligence will only measure it more precisely.