CRM Engine
Activity in. A defensible read on every deal out.
Macklore runs one loop across your whole pipeline: track what happens, score the engagement behind it, flag what is slipping, and suggest the follow-up that recovers it. Five parts of the product, one continuous mechanism.
The mechanism
Four stages, left to right, with nothing hidden in the middle.
Every box below is something you can open and inspect. There is no step where data disappears into a model and reappears as a verdict you cannot question.
Activity data
Mail, calendar, call logs and document views stream in and attach themselves to the right deal and contact.
Engagement scoring engine
Signals are weighted by recency, direction, depth and who was involved, producing a single health read per deal.
Risk flags and follow-up suggestions
Deals whose pattern matches stalling or churn get flagged, each with a specific next action and timing.
Pipeline and forecast reports
Stage movement, health distribution and a commit view weighted by engagement rather than stage probability.
The loop re-runs on every new signal, not on a nightly schedule.
When a proposal is opened at 11:20, the score moves at 11:20 and the board reorders itself before the rep refreshes. A health read that lags a day is a health read that arrives after the call.
- Recency weighted: a reply last week counts for less than a reply this morning
- Direction aware: inbound buyer activity outweighs outbound rep activity
- Depth aware: a forwarded proposal counts for more than a single open
- Thread aware: a widening gap between replies is itself a signal
Inside the product
Five surfaces reps and leaders actually work in.
Each one is a page of its own, with the detail of how it behaves rather than a feature bullet.
Trial
Reading about scoring is slower than watching it run on your own deals.
Import your open pipeline and the first health read lands the same day.
Design decisions
Three choices that shaped how the engine behaves.
They explain most of the differences you will notice against a traditional CRM with an AI feature bolted on.
One record, not four systems
The deal, the contact, the thread history and the health read sit in the same place. Nobody has to reconcile a spreadsheet against an inbox.
Scores you can interrogate
Open any score and the contributing signals are listed with timestamps. A number nobody can explain is a number nobody will trust in a pipeline review.
Built for a rep's morning
The first screen answers one question: what needs me today. Everything else is one click away rather than in your face.
Deal record: Hexfield Interiors
Illustrative view
Proposal sent
₹12,40,000
At risk
18
Suggested follow-up
Re-share the pricing section with the finance contact copied in, and ask one closed question about the approval route. Send today, before the reply gap passes ten days.
The engine recalculated 1,284 signals across this sample pipeline today.
Put your own pipeline through the loop and see which deals read differently than you assumed.
Most teams find two or three deals in their commit list that have had no buyer activity for over a week. Those are usually the ones that decide the quarter.
- Hexfield Interiors moved to at risk
- Varun Steelworks flagged hot
- 4 follow-ups queued for review
Pipeline forecast
Illustrative board view, scored on live activity
Qualified
2
Northbeam Logistics
₹4,80,000 · 6d in stage
Satara Agro Mills
₹2,15,000 · 11d in stage
Proposal sent
2
Hexfield Interiors
₹12,40,000 · 9d in stage
Loomcraft Textiles
₹6,75,000 · 4d in stage
Negotiation
2
Varun Steelworks
₹18,90,000 · 7d in stage
Pinecrest Dairy
₹3,60,000 · 3d in stage
Engagement recalculated on every new email, call and document view
7 open deals scored
This is a sample pipeline shown for illustration. Connect your own CRM data for real, live engagement scoring.