AI for revenue operations
Give it the Salesforce exports, the comp plan document and the bookings file, and it hands back the pipeline review, the back-tested lead score, the comp model and the deal review. Every number names the file it came from.
Fernwell-deal-review.xlsxExample| Period | List | Contract price | Billed, 2 months free | Off list |
|---|---|---|---|---|
| Year 1 | $200,000 | $90,000 | $75,000 | 62.5% |
| Year 2 | $200,000 | $110,000 | $110,000 | 45.0% |
| Year 3 | $200,000 | $135,000 | $135,000 | 32.5% |
| 36-month total | $600,000 | $335,000 | $320,000 | 46.7% |
NoteYear 1 nets 62.5% off list after the free months, past the 50% line, so this deal needs CFO approval, not only the CRO’s.
Ask it for
“Build the Q3 QBR deck from the pipeline export, the ARR export from billing and the spend sheet: ARR waterfall, GRR, NRR, cohort retention and coverage.”
Q3-QBR-deck.pptx“Back-test our lead score on 12 months of leads using only fields that existed at creation, and size the hot, warm and cold bands to SDR capacity.”
Lead-score-backtest.xlsx“Model the FY27 AE plan: $850k quota, 50/30/20 credit split by deal type, 1.5x accelerator over 100%, clawbacks on churn inside 90 days, cost at 85, 100, 120%.”
FY27-AE-comp-plan.xlsx“Deal desk review of the Fernwell Regional Hospital renewal ($90k, $110k, $135k ramp over three years, 45% off list, net-60, first 60 days free).”
Fernwell-deal-review.xlsx“Map every field in the Dynamics export to Salesforce, list picklist values with no target, and write the reconciliation checks the sandbox load must pass.”
Migration-field-map.xlsx“Review the 8,200-contact June webinar list for consent by region, suppression across HubSpot and Mailchimp, and CAN-SPAM and CASL treatment of the resend.”
Webinar-consent-review.docx
Inside the Revenue Operations mode
Operating metrics and reviews
Builds the Monday pipeline report, the forecast roll-up and the QBR deck, plus a metric definitions doc that pins down ARR, GRR, NRR and coverage.
Funnel definitions scoring and routing
Back-tests the score on past leads, drops fields filled in after the outcome, and sets band thresholds from conversion rates and SDR capacity.
Territory quota and comp design
Costs the plan across attainment and checks its crediting, splits, earned dates and clawbacks against California Labor Code 2751.
Deal desk review
Reviews non-standard deals by computing the effective price over the term, testing the structure against the approval matrix and writing up the ASC 606 effects.
Data migration and interface mapping
A CRM move gets a field map, a load order and reconciliation checks, so the go or no-go rests on row counts, sums and spot checks on top accounts.
Marketing privacy and consent review
Checks consent records, the preference center and suppression before each send, and classifies triggered messages as commercial or not under CAN-SPAM and CASL.
Marketing mix and attribution
Campaign recaps set platform-reported conversions against closed-won revenue, so channel lift comes from holdout tests instead of platform claims.
- Office skills
- Excel, Word, PowerPoint, PDF
- Tools
- Read, edit and write files, Search files, Run code, Search and read the web, View images
- Connectors
- SharePoint, Google Drive, Gmail, and additional custom connectors
- House style
- Your fonts, layouts and rules, on every file
How it works
- 1
Open the RevOps folder
Save the Salesforce pipeline and user exports, the marketing automation list pulls, the bookings file from billing, the comp plan document, the deal approval matrix and the routing rules table. Keep them as they come out of each system.
- 2
Ask for the file
Name the back-tested score, the comp model at three attainment levels, the deal review, the field map or the deck. Each one comes back as a finished file with a source on every figure and a note on anything it couldn’t reconcile.
- 3
Review and decide
Every result traces to an export, so you check the numbers against the system of record before the forecast call or the board meeting. Counsel reads the comp plan, finance rules on the deal, and you decide what loads into the CRM.
Who it’s for
Revenue, sales and marketing operations staff, deal desk and sales compensation analysts, CRM and marketing automation administrators, sales planning managers, and the sales and revenue leaders who own the annual sales plan.
Questions
Our Salesforce data is a mess. Will its reports be garbage?
The output is only as good as the exports, so start with the data. Its skills tell it to flag what it could not find or tie out instead of guessing, and a report built on duplicate accounts or stale close dates lists those records rather than smoothing them over. It also builds the cleanup files first: dedupe lists, a field data dictionary and exception reports for deals missing amounts or next steps. Fix what it flags, then ask for the forecast.
Can it calculate commissions from our plan document?
Yes, when the plan is written down. Feed it the plan and the closed-deal export and it applies crediting, splits, accelerators, caps and clawbacks line by line. Each payment shows the deal that produced it, and side agreements nobody wrote down won’t appear anywhere in the calculation. A person who is not paid by the plan signs off before anything reaches payroll.
Will it approve a 45% discount for a rep?
No, approvals stay with the people your policy names. It works out the effective price over the term with the ramps, free months and payment terms, tests the structure against the approval matrix in the folder, and flags the ASC 606 effects for finance to rule on, because a ramp deal’s revenue is recognized ratably while cash follows the stepped billing.
Can it tell us which channel really drives revenue?
It compares platform-reported conversions against the closed-won revenue you actually booked, because the platform total over-counts shared customers. Meta’s 7-day click window and Google’s last click can both claim one sale, so summed dashboards read far above real revenue. It computes the summed total, the booked total and the gap from your exports, and drafts the geo-holdout design and lift calculations for a proper test. The budget call stays with you and marketing.
Where does our pipeline and customer data go?
Ultramarine Work changes files only in the project’s folder. When a request needs it, it can read files elsewhere on your computer, never passwords, keys or app data, and you can turn that off in settings. The model runs on Paper Instruments’ servers, so the content it works on is sent there to be processed. Paper Instruments is SOC 2 Type II, and our security and compliance details are on our compliance page.