Case study·CRE advisory

Lease and critical-date abstraction

From a four-week abstraction backlogto a six-day reviewed extraction with critical dates tied to Salesforce drafts

92%
Simple-field accuracy after reviewMeasured against the legal reviewer's spot-check on the bounded lease set
The value was not extraction alone; it was extraction tied to the next action.
Advisory Partner · UK CRE advisory & brokerage firm
01Pilot envelope
Pilot length5 weeks
First signal6 days
First ROI30 days
Team alongside5 seats · 3 colleagues

5-week pilot on a bounded lease set; Co-ordinator agreed the critical-date dictionary and Salesforce dry-run payload.

02What it owns
Reports toAdvisory Partner, dotted line to the Legal Reviewer on dictionary calibration and the Salesforce Admin on writeback.
Owns
  • Lease abstraction — read each PDF in SharePoint and extract the agreed simple-field set into a structured Word memo with citations
  • Critical date detection — surface breaks, rent reviews, expiries and option windows, with the source clause linked back to the lease
  • Field dictionary maintenance — keep the agreed simple-field set versioned with the legal reviewer and changelog visible to analysts
  • Salesforce task drafts — pre-fill task records in dry-run state for the admin to publish; no live writeback without approval
  • Audit trail — every extracted value carries a page, clause and reviewer reference recorded in SharePoint
Does not do
  • Interpret ambiguous clauses — escalate to the legal reviewer with the clause highlighted
  • Publish to Salesforce — the admin holds the publish button on every drafted task
  • Send landlord, tenant or client communications
Done looks like

Analysts open a lease and read a reviewed memo, not a blank template. Advisors see the next critical date in Salesforce, with the clause citation one click away and the admin's approval on the record.

03The team
AI teammates3
NaomiReads each lease PDF in SharePoint, extracts the agreed simple-field set, and mirrors it into a structured Word memo with page and clause citations.
ReubenSurfaces breaks, rent reviews, expiries and option windows from each abstracted lease and queues them for legal review with the source clause linked.
ElenaPre-fills Salesforce task records for each critical date in a dry-run state for the Salesforce admin to review and publish.
Human team5
  • 3 Advisory partnersAdvisory
  • 3 AnalystsAdvisory
  • Legal reviewerLegal
  • Salesforce adminOperations
  • Operations leadOperations
04Connected stack
SharePointAdobe AcrobatWordSalesforceOutlook
05What it returned
92%Simple-field accuracy after reviewMeasured against the legal reviewer's spot-check on the bounded lease set
43Critical dates foundBreaks, rent reviews, expiries and option windows surfaced with source citations
0Unapproved Salesforce writesTask creation staged in dry-run for the admin to publish
  • Day 0
    Co-ordinator sessionBounded lease set scoped with the advisory partner; first cut of the simple-field dictionary agreed with the legal reviewer.
  • Day 6
    First reviewed extractionSharePoint, Adobe PDF and Word joined; Naomi drafts the first batch of abstracted memos with page and clause citations.
  • Week 2
    Critical-date dictionary agreedReuben surfaces the first forty-three critical dates; legal reviewer signs off on the categorisation and notice-window rules.
  • Week 3
    Salesforce dry-run stagedElena pre-fills task records in a sandbox object; the admin reviews the payload shape before any production writeback is enabled.
  • Week 4
    Re-review on edge casesTwelve abstractions reopened by the legal reviewer; dictionary versioned and re-applied to the rest of the bounded set.
  • Day 30
    ROI reviewAdvisory partner signs off on accuracy, the critical-date list, and the dry-run payload; expansion to a second client scope agreed.
Get started

Want this for your team?

Each design-partner pilot starts the same way: one workflow, the minimum useful context, and a first ROI signal measured in days.