How Much Does Custom AI Automation Cost in India? Founder Ranges, Scope Tiers, and What Not to Buy

At NOW Media, custom AI automation engagements for Indian SMBs typically range between **₹5 lakh and ₹30 lakh** ($6K to $40K USD) when you buy a real operating system: workflow audit, architecture, dashboards, CRM or lead flows, integrations, and team adoption. AI automation is **not just chatbots**. A chatbot bolted onto a messy WhatsApp + sheets stack does not fix reporting, follow-ups, outlet blind spots, or founder visibility. Those are the jobs founders actually pay for.

This guide uses **NOW Media's engagement range**, as published on our AI automation service, and explains how scope changes the quote. These are our planning ranges, not a survey of India-wide prices. Bias declared: we sell custom systems we also run on. You still get a buy-vs-build decision you can use with any vendor.

Direct answer: NOW Media's range in 2026

Our published engagement range is ₹5L-₹30L (approximately $6K-$40K USD), with a typical overall timeline of 4-16 weeks. A discovery brief determines the quote; the scope examples below are not fixed-price packages.

Light workflow automation

One or two painful processes: lead routing, status syncs, simple reporting, and limited integrations. A narrow, clean scope usually sits nearer the lower end of our engagement range.

Founder ops dashboard

One operating view for revenue, leads, delivery, and expenses, with CRM or lead flows, Indian FY-aware reporting, and team onboarding. Data sources, roles, and reporting rules determine the estimate.

Multi-outlet systems and AI assistants

Branch comparison, role-based views, assistants connected to SOPs and project data, deeper integrations, and data cleanup. These requirements push scope and cost toward the upper end; larger requirements may need separate phases.

A CRM configuration and a founder operations platform solve different problems. Compare the workflows covered, integrations, ownership, and ongoing manual work rather than the headline fee alone.

What you are actually buying

A serious engagement is not a widget. It is a sequence.

1. **Business workflow audit.** How leads, servicing, reporting, money, and people actually move today (including the WhatsApp threads nobody wants to admit are the system of record).
2. Gap map. Missed follow-ups, delayed reports, outlet-level blind spots, repeated manual entry, decisions made without live numbers.
3. Custom architecture. Dashboards, portals, CRM flows, approvals, assistants, and integrations shaped to your business. Keep tools that work. Replace workflows that slow you down.
4. Build with demos. Working slices weekly, not a big reveal at week twelve.
5. **Training and improvement.** Onboarding, documentation, usage checks, and post-launch fixes based on what people actually click.

If a proposal jumps straight to "we will add AI" without mapping the work, you are buying a demo, not an operating layer. Our AI automation page spells the five inclusions and five phases the same way we sell them.

Every project we run starts with a **Discovery Blueprint**: where leads come from, how teams follow up, how outlets report, where data lives, and where founders lose the plot between tools. That document is the scope. Without it, the fee is a guess dressed as confidence.

SaaS vs custom vs hybrid

**Buy SaaS (Zoho, HubSpot, and peers) when:**

• One primary sales motion, one entity, clean data habits.
• Out-of-the-box pipelines and reports match how you sell.
• Your team will live inside that product without constant side spreadsheets.

**Build custom (or heavily customized) when:**

• Multi-outlet or multi-brand reporting is the real product, not a nice-to-have.
• Indian fiscal year, lakh formatting, multi-currency, or partner dashboards are first-class needs. Generic SaaS fights you here; we wrote that up in the Indian fiscal year problem with generic SaaS.
• Founders need one screen that mixes revenue, delivery, and ops, not three exports taped together on Monday.
• Your stack already includes WhatsApp, ads, forms, and local payment tools that need a coherent routing layer.

**Hybrid is the common adult answer.** Keep SaaS where it wins. Build the operating layer that SaaS will never care about: outlet P&L views, FY targets, scope builders, partner portals, assistants trained on your SOPs. We built our own ops command center for that reason (CRM, scope builder, project tracker, invoicing, Indian FY revenue targets, Claude assistants). Soft fact: we can build a version of that pattern for you. The longer story of why we refused a generic CRM is in why we built Bleep instead of buying a CRM.

Founders sometimes ask if custom means "throw away Zoho." Usually no. Custom means stop pretending a generic CRM is also your multi-outlet finance view and your partner portal.

Scope drivers that move the fee

These are the levers that push a project toward ₹5L or toward ₹30L:

• **Integrations.** Ads, website forms, WhatsApp, accounting, existing CRM, POS, booking tools. Each live source adds mapping and failure handling.
• **Outlets / entities.** Two branches is not ten. Roll-ups, permissions, and partner views multiply design and testing.
• Roles. Founder, GM, sales, franchisee, finance. Different jobs need different screens and different AI prompts.
• **AI assistants.** Assistants tied to documents, SOPs, and project data cost more than a generic chat box and deliver more. See Claude AI assistants in production.
• **Data cleanup.** Dirty sheets, duplicate leads, and years of informal naming are often the silent half of the project.
• **Adoption.** Training, change management, and post-launch iteration are part of the fee if you want usage, not shelfware.

A useful buying habit: ask every vendor to price the same list of sources, roles, and outlets. If one quote is half the other, check whether they skipped cleanup, assistants, or onboarding.

Timeline: 4-16 weeks explained

The windows below are indicative planning examples within our published range. Data readiness and integration requirements can change them.

4-6 weeks

What fits: Narrow workflow automations; limited integrations; one primary dashboard

6-12 weeks

What fits: Founder dashboard + CRM/lead layer + onboarding for a core team

10-16 weeks

What fits: Multi-outlet platforms, heavier integrations, assistants, partner views

Rushing sixteen weeks of work into four usually means skipped audit, skipped cleanup, or skipped training. The calendar is not vanity. It is how you avoid a system nobody opens after week two.

Weekly demos matter more than Gantt theatre. If you cannot see a working slice by week three or four on a mid-scope build, the project is drifting into slideware.

Red flags in AI automation pitches

Walk away, or at least renegotiate hard, when you see:

• **"AI automation" equals a chatbot widget** with no workflow map.
• No Discovery or audit phase. Only a fixed feature list invented before anyone sat with your team.
• **ROI percentages invented on the sales call.** If nobody has seen your data, they cannot know your recovery rate.
• **Unlimited scope language.** "We will automate everything" without owners, sources, or success criteria.
• **No training or adoption plan.** Tools die when the champion leaves.
• **Quotes that ignore Indian FY, GST-adjacent reporting needs, or multi-currency** when your business clearly has them.
• **Demoing a generic template as if it were your architecture.**

Declare the opposite of those flags in writing before money moves. Ask who owns the data model, who trains the team, and what happens in the first thirty days after launch.

What NOW Media ships

We are a Bangalore studio (NOW Media, a brand of Bleep Design) that builds custom AI systems as part of a wider stack: brand, website, content, automation. Industry patterns we see repeatedly:

• Hospitality groups: multi-outlet revenue, GM daily views, partner dashboards.
• **Fitness chains:** member lifecycle, lead response, branch comparison.
• Real estate teams: enquiry routing, partner pipelines, inventory velocity views.
• Healthcare operators: appointment load, reminders, multi-branch ops views (ops tooling, not clinical advice).

Proof of product thinking lives on our about page and on the stack essay branding agency stack 2026. The automation product itself starts at AI automation.

We are not a pure software shop quoting CRM screens without ops context. That is the honest trade: you pay for systems shaped around how Indian multi-outlet and service businesses actually run.

What not to buy

Skip these purchases when the pitch is louder than the map:

• A chatbot as a substitute for CRM hygiene.
• A "full AI transformation" with no named workflows.
• A dashboard with no owner for data quality.
• A six-month retainer that never defines a shippable milestone.

Buy the smallest system that removes a real weekly pain. Expand once people use it.

FAQ

How much does custom AI automation cost for an Indian SMB?

At NOW Media, our published range for custom engagements covering audit, architecture, build, and adoption is ₹5 lakh to ₹30 lakh (approximately $6K to $40K USD). Multi-outlet systems with assistants sit toward the top of the NOW band or above if scope expands.

When should a founder build a custom dashboard instead of buying SaaS CRM?

Build when SaaS reporting cannot express how you actually run: multi-outlet roll-ups, Indian FY targets, partner views, mixed revenue and delivery on one screen, or workflows that live in WhatsApp and sheets. Buy SaaS when one clean sales motion fits the product and your team will live there. Hybrid is normal: SaaS for CRM basics, custom for the operating layer SaaS ignores.

What is included in a ₹5-30 lakh AI automation engagement?

At NOW Media, that band usually covers workflow audit, gap mapping, custom architecture, AI-enabled build (dashboards, CRM/lead flows, ops tools, assistants as scoped), and training with post-launch improvement. Exact inclusions track integrations, outlets, roles, and data condition. A ₹5L project is narrower than a ₹30L multi-outlet platform.

How long does it take to ship a founder ops dashboard?

An indicative planning window for a founder dashboard is 6-12 weeks, within our published 4-16 week automation range. Confirm this against the agreed scope. Narrow automations can finish closer to four weeks. Multi-outlet platforms with assistants lean toward twelve to sixteen.

Is AI automation just chatbots?

No. Chatbots are one optional surface. Useful automation includes founder dashboards, lead and CRM flows, operations tools, multi-outlet reporting, approvals, and internal assistants connected to real business data. If the pitch is only a chat widget, you are not buying automation of the business.

Next step

If you are comparing Zoho or HubSpot retainers against a custom build, start with a **Discovery Blueprint**. We map how work moves today, where visibility breaks, and which tier (if any) is worth paying for.

Email hello@nowmedia.in or use /contact. Service detail: /services/ai-automation.