Muljat CRM — Deep Dive & Competitive Benchmark

Part 1 · Full Module Inventory — every screen, walked

I opened all 20+ sections live. Verdict on each: is it real, and what's the gap? The pattern is unmistakable — the build quality is high; the data is contaminated with demo records from the development process.

Module	What it actually does	Verdict & gap
Dashboard	90 Scorecard, Rocks, KPI tiles, pipeline-value chart, win rate, leases-expiring radar, "coming up" tasks	✅ Real — see the Review Checklist; numbers polluted by demo deals
Calendar	Month/Week/Day/Agenda, color-coded event types (Tasks, Close, Showings, Feasibility, Timeline), all-day rows	✅ Real & full-featured
Pipeline (All Deals)	Kanban board, 185 deals / $348M, stage columns, drag-drop, action-plan picker, KPI header	🟡 Real — deal-card + action-plan fixes pending (Checklist)
Lead Routing	Round-robin auto-assignment with a persistent rotation cursor, master on/off, one-click rebalance, brokers ranked by pipeline	✅ Real — beyond LeadSimple
Re-engage Leads	150 dead/dormant leads (Lost / Nurturing / Competitor), one-click revive → Nurturing + follow-up task	✅ Real — strong lead recycling
Tasks	42 action items across deals, owner filters, urgency sort, overdue tracking	🔴 Real but dirty — 15 fake-overdue; phantom users (Doug/Sara/Mike) in filters
Action Plans	36 trigger-based workflow templates, 227 steps, fire on stage change	🔴 The mess — "(copy)" duplicates w/ 0 steps shadow the real "[AP2026]" plans
Properties	163 assets · 522K SF · 240 units · 481 tenants · 89% occupancy, portfolio view, filters	✅ Real — far beyond LeadSimple
Map	163 assets across 17 cities, layered (Portfolio/Sale listings/Sold comps/In development), color by type	✅ Real & impressive — 40 of 123 geocoded
Access Hub	123 credentials — signs, keys, lockboxes, master + tenant codes — searchable, ties to property pages	✅ Real & populated — genuinely useful
Listings	38 active · $34.9M for-sale, For Sale / For Lease / Prepping / Archived, type + price + SF filters	✅ Real — "Package B/D" demo naming; 325 archived
Listing Copy	3 channels (LoopNet / Craigslist / Buyer email) × 3 voices × 3 lengths, drafts from a linked deal	✅ Real & clever — sources only "(copy)" demo deals
LOI Builder	24 clauses in the firm's voice, Tenant/Landlord smart defaults, advanced + retail provision sets, RCW citations, Word export	✅ Excellent — standalone; wire it to pull from a deal
Lease Abstracts	161 abstracts · $63.2M scheduled rent, AI "upload PDF → Opus distills terms," forward-rolloff chart, ledger	✅ Real (AI) — no NOI data; verify critical-date accuracy
Comps	199 closes · $58.8M sale volume, median $/SF + DOM + close-vs-ask, 12-mo trajectory, CSV export	🔴 Real but broken data — lease median reads $0.02/SF
Contacts	426 relationships, recency-ranked, "scan a card," role tabs, $2.4M lifetime GCI	🔴 Real but unsegmented — all 426 tagged "Tenant," 0 owners
Team	7-person roster, production-by-person (deals + %), tenure mix, active-pipeline ranking	✅ Real — this IS principal-grade reporting; demo-skewed
Deal Rooms	NDA-gated document sharing with signed/views tracking, shareable links	✅ Real — empty; test with a real OM + external viewer
Settings	Company · Regional & Fiscal · Notifications · Security · Branding tabs	✅ Real — security posture defined (see Part 5)
AI Assistant ("Grok")	Data-connected chat ("sees pipeline, deals, listings, contacts"), Grok ↔ Claude model toggle, sandboxed to CRE, context-aware prompts	✅ Real — accuracy of answers still to verify

The one finding that ties the whole CRM together

Every analytical surface is degraded by the same root cause: demo data from the build. Comps shows a $0.02/SF lease median; Listing Copy can only draft from "(copy)" deals; Contacts dumps all 426 people as "Tenant" with 0 owners; Action Plans stacks empty "(copy)" templates over the real ones; Tasks invents overdue items and phantom users. This is not 20 separate problems — it's one data-reconciliation problem (Card 1 on the Strategic Review) showing up in 20 places. Clean the data and most of this page turns green at once.

Part 2 · Competitive Benchmark — what brokers actually buy

The CRE-CRM market has consolidated hard around Buildout, which now owns both Apto (2022, no longer sold to new buyers) and Rethink (2021, its flagship). Here's the field, with sourced pricing.

CRM	Best for	Price / user / mo	Standout	Weakness
Buildout / Rethink+	Mid-large, marketing-heavy listing shops	~$129–159 + ~$275/mo platform fee + add-ons	OM/brochure generation; embedded ownership data + AI	Opaque stacked pricing; setup complexity; weak mapping
AscendixRE	Mid-large, incl. enterprise (on Salesforce)	$79 Foundations / $99 Enterprise (SF license incl.)	Stacking plans + AI lease abstraction + commissions natively	Salesforce admin overhead; paid config services
ClientLook	Solos & small teams, new agents	$129/mo or $1,086/yr	Human Virtual Assistants do your data entry	No reporting/dashboards; slowness; no marketing automation
RealNex	Budget independents (CCIM / RE/MAX)	$129–149	Cheapest all-in-one (CRM + analysis + marketing)	"Outdated UI, support non-existent"; can't invoice commissions
Apto	(Legacy — maintenance only)	~$89 historically	CRE-on-Salesforce depth	Dead to new buyers; admin-heavy
Reonomy / ProspectNow	Prospecting data (not a CRM)	$119–$400+ ($4,800/yr Reonomy)	National ownership data + "likely seller" scoring	Data feed, not a pipeline — sits on top
Muljat CRM	MGC's exact workflow	Custom build (no per-seat fee)	EOS + LOI + lease-distill + comps + map in ONE owned system	Data hygiene + single-developer governance

Where the Muljat CRM already wins ✅

Breadth in one owned system — pipeline + properties + map + comps + lease abstracts + LOI builder + listing copy + deal rooms + EOS. No single incumbent does all of this cleanly.
EOS built in — Rocks + Scorecard + Issues/To-Dos. No CRE CRM on the market does this (more below).
Lead routing & re-engagement more automated than LeadSimple.
Production-by-broker reporting (Team page) — the analytics incumbents charge extra for.
AI assistant with a Grok↔Claude toggle — newer than anything the incumbents ship.
No per-seat fee. Replaces $6K–$13K/yr of subscriptions.

Where the incumbents still win 🟡

Reliability of the basics — every button works in a 10-year-old product; the deal card doesn't yet.
Commission accounting depth — CommissionTrac/AscendixRE handle splits, draws, phased lease commissions, 1099s. The Muljat CRM estimates a number; it doesn't account.
Stacking plans — visual floor-by-floor occupancy. Not present; worth considering.
National ownership data — Reonomy/ProspectNow scale. (You substitute local CBA + CoStar + assessor — fine for your market.)
Battle-tested security & data governance — see Part 5.
Mature mobile — unverified here; incumbents are proven on phones.

Part 3 · The Cost Story — what this replaces

A 5-person commercial brokerage realistically spends, per sourced pricing:

Option (5 users)	Annual cost	Notes
ClientLook	~$5,400/yr	Cheapest credible all-in-one; thin reporting
AscendixRE Enterprise	~$5,940/yr	Best feature-per-dollar; + paid config
RealNex	~$8,000/yr	Cheap on paper; UI/support risk
Buildout / Rethink	~$11,000–13,000/yr	Per-user + ~$3,300/yr platform fee + add-ons
+ Prospecting data	+$11,000–24,000/yr	ProspectNow / Reonomy sit on top
Muljat CRM	Dev cost only	Replaces $6K–$13K/yr (CRM) — and the spreadsheets it kills

The internal ROI story writes itself: a custom tool that retires $6K–$13K/yr of recurring subscriptions (plus LeadSimple, plus Ninety.io at ~$12/user/mo) pays back fast — provided the maintenance and governance are handled (Part 5). The warning the literature repeats: with custom software, "the $100K isn't the build — it's the next five years."

Part 4 · Novel vs. Table Stakes vs. Risky

Where the research reframes what to build proudly, what to build to par, and what will actually bite.

🟢 Genuinely novel — build these proudly

EOS scorecard auto-populated from CRM data. The EOS-software market (Ninety, Bloom, TractionFlow) openly complains about "double data entry" — operators retype pipeline numbers into a separate scorecard tool. Your scorecard data already lives in the CRM. No CRE CRM does this. This is your single most defensible original idea — it could eventually retire Ninety.io ($12–14/user/mo) entirely.
Property-as-hub relationship mapping. Brokers' #1 unmet need: "show me everyone tied to this building" — owners, tenants, brokers, lenders, history. Every incumbent under-serves it. You're already close (Properties ↔ Pipeline ↔ Contacts). Nail the property-as-hub model and you beat them where they're all weak.
One owned workflow tuned to MGC. Off-the-shelf CRMs are "built for everyone, perfect for no one." Your idiosyncratic mix (EOS + tenant-rep + LOI + comps + the BellinghamCRE database) is exactly the case where custom wins.

⚪ Table stakes — build to par, don't oversell

AI lease abstraction is commoditized. Crexi Vault (24+ fields, ~2 min, free to users), LeaseLens ($25/export, 200+ fields), Kira (93–97% accuracy), Leverton/MRI, Prophia (99% w/ human QA), VTS AI all do this. Your feature is competitive only because it lives inside the deal workflow — not because abstraction itself is rare. Critical: every trusted incumbent pairs AI extraction with human validation — never ship distilled lease terms as authoritative without a person confirming them.
AI listing copy & LOI drafting — Crexi AI Script, Buildout, Genie AI already do these. Fine to have; not a differentiator.
Kill data entry. ClientLook hires humans to do brokers' data entry — because manual entry is the #1 reason CRMs fail to get adopted and brokers revert to spreadsheets. Your AI assistant + email auto-capture can do this in software. This is higher-leverage than any new feature.

🔴 Risky — the things that actually sink custom builds

Single-developer key-person dependency. "When one developer is the only party who can maintain the app and the only one with source-code access, the developer has become a significant risk to the business." This is your #1 risk — and it's cheap to fix now (Part 5).
Commission-accounting completeness. The benchmark is CommissionTrac (Yardi): unlimited splits, caps, multi-tier referrals, agent draws + draw recovery, phased/contingent lease commissions paid over time, vouchers, branded invoicing, 1099s, audit trails. Easy to underbuild and discover the gaps mid-close. RealNex literally "can't invoice" — don't repeat that.
Security posture for a system of record. Client-side password gating (fine for a marketing site) is not adequate for a system holding client and commission data. The CRM's Settings shows a real security model (12-char passwords, 2FA for admins, session timeouts) — confirm it's enforced, not just displayed.
Maintenance drag. The recurring custom-software warning: the build is the cheap part.

Part 5 · Governance — de-risk the system of record

Consensus best practices for de-risking a single-developer custom build that becomes your system of record. These are cheap now and expensive after the developer is gone.

Do these before you're fully dependent

Own the repository. Create a GitHub org in Muljat's name; Robert is a collaborator, not the owner. Immediate code access, no escrow fees — the practical move for most small businesses.
Source-code escrow as the fallback if he insists on holding the repo — release triggered if he can't maintain/support it.
Guaranteed data export. Contractually, you can pull all deals, contacts, commissions, and lease abstracts in open formats, on demand, independent of Robert.
Documentation as a deliverable — a runbook for deploy/restore, not logic locked in one person's head.
Second-developer readiness — at least one other person who can deploy and patch.
Real authentication & backups appropriate to client/commission data, with a tested restore — confirm the Settings security model is enforced.

The Bottom Line

What this review concludes

The Muljat CRM is real, broad, and in several places ahead of products that cost $6K–$13K a year — the EOS integration and the unified CRE toolkit are things no incumbent ships. It is not vaporware and it is not a LeadSimple clone; it's closer to a custom Buildout-plus-Ninety.io with your own LOI builder and market database bolted in.

The two things standing between it and a confident July 1 launch are not features:

Data hygiene — one reconciliation fixes twenty symptoms. Purge the demo records, segment the contacts, fix the commission/value math, and every analytical screen becomes trustworthy at once.
Governance — own the repo, guarantee export, document, and harden auth, so a system of record isn't a single-person dependency.

Get those right and the strategic plays the market hands you are clear: make the property the hub (the need every incumbent misses) and kill data entry with AI (the reason CRMs fail to stick). That's a CRM no one can buy off a shelf.

Muljat CRM Review

The whole CRM, and the whole market.