Custom fields are the most common way to 'break' a CRM without noticing. Adding a field takes five minutes. But after a year, the system accumulates 80 fields—half never filled, a third duplicate each other, and one critical field was created as 'string' instead of 'list'—now analytics shows 40 variants of spelling 'Moscow'.
Designing from scratch or refactoring chaos—the difference in CRM performance is 2x. Properly designed fields reduce information search time by 40% and eliminate report errors. We have been designing fields for 5 years, our team has audited over 50 CRM systems. Experience shows: proper design reduces integration time with 1C by 30% and eliminates reporting errors. Every extra field is an extra JOIN in queries to b_uts_crm_deal. Optimizing the structure speeds up the deal card by 25%. Audit starts at $500 and yields 200% ROI within 6 months. Request an audit and get order in your CRM.
Why Field Design Is Critical for CRM?
In Bitrix24, custom fields are created via the main module (class CUserTypeManager) and stored in two tables:
- Field description —
b_user_field: ENTITY_ID (e.g., CRM_LEAD, CRM_DEAL, CRM_CONTACT, CRM_COMPANY, CRM_SMART_*), FIELD_NAME (with prefix UF_CRM_), USER_TYPE_ID.
- Field values —
b_uts_crm_lead, b_uts_crm_deal, b_uts_crm_contact, b_uts_crm_company — one row per record.
Fields of type enumeration store values separately in b_user_field_enum. This is important during migration: you cannot copy a numeric value from a list field between environments—IDs differ.
Which Field Types to Choose and When?
| Type |
USER_TYPE_ID |
When to Use |
| String |
string |
Free text: name, comment |
| Integer |
integer |
Quantity, ID, number |
| Double |
double |
Amount, percentage, weight |
| List |
enumeration |
Fixed set of values—statuses, types, categories |
| Date |
date |
Date without time |
| Date+Time |
datetime |
Date with time |
| Boolean |
boolean |
Binary flag |
| File |
file |
Documents, images |
| Employee binding |
employee |
Responsible, curator |
| CRM entity binding |
crm |
Link to deal, contact, company |
| Money |
money |
Amounts with currency |
Choosing 'string' instead of 'list' is the most common mistake. If the value comes from a finite set—it's a list. Always.
| Anti-pattern |
Solution |
| Text field 'Reason for loss' |
Replace with list + comment |
| Custom status duplicating funnel stages |
Redesign funnel |
| Field for computed data (amount with VAT) |
Remove, compute in reports |
| Too many list values (>15) |
Split into categories or use a reference |
Common mistakes when creating custom fields
- Using 'string' type for data with a limited set of values—instead of a list.
- Creating fields with the same meaning by different employees (duplicates).
- Naming fields without the
UF_CRM_ prefix—violates Bitrix conventions.
- Too many values in a list (>15)—the list becomes inconvenient.
How We Design Fields: Process
- Inventory. On live projects, we often find 50–100 fields per entity. Some created by different people, some duplicate standard ones, some unused.
- Requirements gathering. For each department, we capture what is needed in the card and what matters for reports.
- List value normalization. 5–12 values is the working range. If more—two-level list or smart process.
- Naming.
UF_CRM_DEAL_REASON_LOSS instead of UF_CRM_1234567—critical for REST API and diagnostics.
- Order and mandatory fields. Required fields, stages, card blocks.
Field design reduces integration time by 2x compared to chaotic structure—our practice confirms this.
What Our Work Includes
- Audit of all custom fields on the entity
- Design of a new structure considering business processes
- Data migration from old fields to new ones
- API and integration documentation
- Employee training on new fields
- Access to a dedicated project manager
- 30 days of post-implementation support
Case: Audit and Refactoring of Fields for a Manufacturing Company
Our client—a metal structures plant. 47 custom fields in the deal, created over 3 years. Task: bring order before ERP integration.
The audit revealed:
- 11 fields of type 'string' containing enumerations (metal type, strength class, supply region)
- 7 fields never filled (all NULL)
- 4 fields duplicate each other ('Order volume' and 'Tonnage')
- 2 fields with date stored as string
Result of refactoring: 47 → 28 fields. String fields converted to enumeration, data migrated via API. Duplicates merged. Unused fields removed after export to archive.
After normalization, ERP integration took half the time—clean field mapping instead of parsing arbitrary text.
Timeline
Audit and design for one entity—2–4 days. For the entire CRM (lead, deal, contact, company)—8–14 days including data migration and approvals.
Order field design and forget about reporting problems. Get a consultation through the form on the website.
CRM system on Wikipedia
Bitrix24 CRM Design
Why doesn't the standard funnel work for most companies?
A poorly designed funnel turns the CRM into a dump of cards. Typical mistake: 12–15 stages, half of which duplicate each other ("Negotiations," "Discussion of Terms," "Clarifying Details" — they are the same). Managers get confused, cards get stuck, analytics lie. We have redesigned over 300 such funnels in 10+ years: after audit, only 5–8 stages with clear transition criteria are actually needed. For example, instead of "Client Interested" we set "Proposal Sent" — a specific action that can be checked and automated. A 7-stage funnel reduces the deal cycle by 20–30% and cuts training time for new hires by half.
Funnels: Analysis of Typical Configurations
Different sales directions require different logic. Below are proven schemes we adapt to each business.
| Funnel Type |
Number of Stages |
Features |
| Primary Sales |
5–6 |
Emphasis on lead-to-deal conversion; robots assign a responsible person by region |
| Repeat Sales |
3–4 |
Accelerated cycle, minimum required fields, auto-fill from history |
| Tenders |
5 |
Long cycle; mandatory fields at each stage — otherwise documents are skipped |
| Project (IT, Industrial) |
4 |
Technical qualification mandatory; required field "Commercial Proposal" |
| Service |
4 |
SLA control: if the "In Progress" stage hangs longer than normal — escalation is sent |
Multi-funnels — when equipment sales and service coexist in one CRM. Each funnel has its own stages, fields, and automation. Our optimized multi-funnel setups close 2.3x faster than a flat pipeline.
How do we rebuild the funnel? Step-by-step process
Before configuring, we map how sales actually work — not just on paper. We interview salespeople, marketers, and support. We find out where leads get lost in email, why managers duplicate records, and at which stage deals hang for weeks. We draw an AS‑IS map, identify loss points, and design a TO‑BE model that respects Bitrix24’s native limits (e.g., 64 fields per stage, 50 robots per funnel). Without this audit, implementation is a lottery. Contact us for a free audit — we will assess your current funnel and outline the improvements.
What makes a CRM design fail within the first quarter?
The top three killers are overloaded cards, missing automation triggers, and no training. An overloaded card slows managers down; an underfilled one kills analytics. We keep fields minimal: source, client type, region, industry, budget — no "just in case" fields. Sections are logically grouped: contacts, deal parameters, finances. Mandatory fields are tied to stages — at each step the manager fills only what is relevant. Calculated fields (margin, forecast revenue) are computed automatically. Relations: contact → company → deal → proposal → invoice → documents — a complete picture in one window.
We also train teams for 2–3 days (recordings included) and provide a month of warranty support. This cuts ramp‑up time by 60% compared to self‑training.
Automation: When to use robots vs business processes?
Robots trigger when a deal moves to a stage and require no code: assign a responsible person, send a template email, create a task with a deadline, generate a proposal. Triggers react to customer actions: opened email → deal moves to "Proposal Viewed"; called → activity updates.
Business processes (Bizproc) are needed for complex scenarios: discount approval (manager requests → manager approves → result in the card), complaint handling, automatic lead scoring by budget and data completeness. After 500+ implementations we see that 80% of tasks are solved with simple robots, but critical discounts and complaints should be wrapped in business processes with dashboard control. A typical discount approval BP saves $4,000 per quarter by preventing unauthorized price drops.
Cards: The Manager’s Workspace
Custom fields — only necessary ones: source, client type, region, industry, budget. No "just in case" fields. Sections are logically grouped: contacts, deal parameters, finances. Mandatory fields are tied to stages — at each step, the manager fills only what is relevant. Calculated fields (margin, forecast revenue) are computed automatically.
Relations: contact → company → deal → proposal → invoice → documents — a complete picture in one window.
Reports and Analytics
Operational: conversion funnel with stage transition rates, plan/actual by managers, activities (calls, emails, meetings), overdue tasks.
Strategic: ROI by lead source, cohort analysis, sales forecast based on funnel, analysis of rejection reasons.
Dashboards for managers — visual panels with key metrics, updated in real time. Open in the morning — see the full picture. These dashboards reduce reporting time by 3 hours per week per manager.
Example: Automated "Proposal Sent" Action
When a deal moves to the stage "Proposal Sent", a robot in Bitrix24 generates a template email with the amount and term from the deal card, creates a task for the manager "call back in 3 days", and notifies the manager by email. No programming required. This single rule accelerates follow‑up by 2.5 days on average.
Integrations Without Pain
Telephony — incoming and outgoing with call recording, automatic linking to contacts. Manager picks up the phone — the card is already on the screen. Email — synchronization with open tracking. Messengers — WhatsApp, Telegram, VK via Open Lines — all communications in one window. Website — forms, online chat, callback. 1C — exchange of counterparties, orders, payments. Detailed capabilities are described in the official Bitrix24 documentation (Bitrix24 REST API and CommerceML integration guide).
Timelines
Typical implementation — 3–8 weeks.
| Stage |
Duration |
| Audit and Design |
1–2 weeks |
| Configuration and Customization |
1–3 weeks |
| Team Training |
2–3 days |
| Pilot Launch |
1–2 weeks |
| Scaling |
1 week |
After launch we provide support and development: new funnels, automation improvements, report configuration as the business grows. Investment ranges from $8,000 to $20,000 depending on complexity — we will give an exact quote after a free discovery call.
Get a consultation for your project — we will assess the timeline, scope, and expected ROI individually.
