A Shopify build in Bengaluru lands somewhere between ₹60,000 and ₹25 lakh, and that range is useless to you until you know which band you're in. Rough markers from what we quote: a premium theme configured properly, with no custom sections and no migration, runs ₹60,000 to ₹1,50,000. A standard D2C store with eight to twelve bespoke sections, three or four integrations and a clean data import sits at ₹2,00,000 to ₹5,00,000. A replatform off Magento or WooCommerce with a real catalogue behind it, or a Plus build with checkout extensibility and B2B, starts around ₹8,00,000 and goes up from there. The interesting part isn't the spread. It's that two agencies on Hosur Road will quote ₹1.8L and ₹4.6L for what the merchant sincerely believes is the same website.
It usually isn't. Three things account for most of that gap, and you can force both quotes onto the same footing in about twenty minutes of asking.
Custom sections are the line item nobody counts
Every Shopify quote includes "theme setup". That phrase covers everything from uploading Dawn and swapping the logo to writing fifteen schema-driven sections that your marketing person can rearrange without calling anyone.
A custom section is a block of the page with its own settings: a bundle builder, an ingredient strip, a size-chart drawer, a comparison table, a shade selector that filters variants by skin tone. Each one is design, build, schema, mobile behaviour and QA. Six hours is a fair average once you account for the round of changes after the client sees it on their own phone. Some take two. The bundle builder takes twenty.
So ask both quotes for a numbered list of sections. Not "custom homepage". A list. If one says nine and the other says "theme customisation as per design", you are not comparing prices, you're comparing two people's imaginations.
App replacement versus app rental
The second gap is philosophical and it costs real money either way. Some agencies build features into the theme. Others install an app and configure it.
Custom code has no monthly fee and no third-party script slowing your collection pages down. It also has no vendor fixing it when Shopify changes something in 2027. Apps cost $15 to $99 a month each and stack up fast, but somebody else maintains them and the feature ships on Tuesday instead of in three weeks.
Our rule: rent anything where the vendor's whole business is that one problem, build anything that is specific to how you sell. Product filtering and on-site search is the clearest example of the first kind. Building a faceted filter system by hand for a 4,000-SKU catalogue is forty hours of work that will be worse than an app you can install this afternoon, which is partly why we ended up publishing FilterPro rather than rebuilding filters for every project. Subscriptions, reviews, loyalty: rent. A made-to-order flow with fabric, lining and monogram options priced differently: build.
The quote that looks cheap is often the one with eight apps behind it. Add the monthly fees up over 24 months before you decide it was cheaper. Six apps at an average of $35 is $210 a month, about ₹17,500, which is ₹4.2 lakh over two years.
Migration depth, which is where most of the hidden hours live
"We'll migrate your data" means at least four different jobs and the quotes rarely say which ones are included.
- Products with variants, metafields, option sets and images at the right resolution.
- Customer accounts, including the fact that passwords never come across and everyone has to reset.
- Historical orders, which you need for GST records and returns, and which is the slowest part.
- URL redirects from the old structure, blog posts included. Skip this and you hand back rankings you spent two years buying.
A 500-SKU store with clean data is a day's work. A 12,000-SKU store where sizes were stored as separate products and half the descriptions have inline styles from 2019 is two weeks, most of it cleaning rather than moving. We get the effort estimate wrong on the first pass often enough that we now ask for a CSV export before quoting a replatforming project. If an agency quotes migration without looking at your data, the number is decoration.
Two Bangalore quotes, same store, and where the ₹2.8 lakh went
Take a real shape of project: apparel brand, moving off WooCommerce, 12,000 SKUs after variants, 40,000 customers, 9,000 historical orders. Quote A is ₹1,80,000. Quote B is ₹4,60,000. The founder assumes B is padding.
Line the scopes up at a blended ₹1,800 an hour:
- Nine custom sections that A didn't include, at six hours each: 54 hours, ₹97,200.
- Full migration including order history and 1,100 redirects: 40 hours, ₹72,000.
- Razorpay plus a COD flow with pincode serviceability, Shiprocket, and GST-compliant invoicing: 30 hours, ₹54,000.
- Device QA and a real UAT round: 20 hours, ₹36,000.
- Project management and contingency: ₹20,800.
That totals ₹2,80,000, which is the gap almost exactly. Quote A was honest. It was also for a different website: theme setup, product import, Razorpay, go live. Both are correct answers to questions nobody wrote down.
What developers actually charge per hour here
Freelancers in Bengaluru work at roughly ₹800 to ₹2,000 an hour depending on whether they write Liquid or also write the app extension. Agencies bill ₹1,500 to ₹4,000, and at the top of that you're paying for a designer, a developer, QA and someone who answers on Saturday during your sale. Add 18% GST on services, which people forget when they're comparing a freelancer's cash rate to an agency invoice.
Hourly only makes sense for maintenance and small changes. For a build, insist on fixed price against a written scope, with hourly as the mechanism for anything added later. Open-ended hourly on a first build is how a ₹2L project becomes ₹5L with nobody behaving badly. Rates across the country don't vary as much as people expect either; we've written out how pricing compares across Indian cities and the delta between Bengaluru, Mumbai and Delhi is smaller than the delta between two agencies on the same street.
If you're a startup launching your first store
Spend less than you think. A first store's job is to find out whether anyone wants the product, and a ₹4 lakh custom build tells you nothing a ₹90,000 build wouldn't.
Buy a premium theme in the $300 range, have someone set it up properly, wire up Razorpay with UPI and cards, configure COD with pincode restrictions and a partial-prepaid nudge, connect Shiprocket, get GST invoicing sorted, and launch. Keep ₹50,000 back for the fixes you'll want in month two, because you will want them and they'll be better informed than anything you could specify now.
Where we'd argue for spending more on day one: if you sell perishables under FSSAI rules with batch and expiry logic, if you're doing made-to-order with configurable pricing, or if your catalogue is large enough that navigation is the product. Those aren't things you retrofit cheaply.
The cheapest quote in Bengaluru, and what it leaves out
You can find someone to do it for ₹25,000. Sometimes that's fine. If your requirement is genuinely a theme install with fifty products, a ₹25,000 developer and a ₹2,00,000 agency will produce a similar store, and paying the higher number is silly.
What tends to be missing at the bottom of the market: mobile performance, ownership, and the India-specific plumbing. On the first, most cheap builds ship a homepage with eight uncompressed hero images and an LCP over four seconds on a 4G connection in a Tier-2 city, which is where a lot of your traffic is. On the second, ask who holds the theme repository and whether you get the code. On the third, COD is the one that hurts: a store that takes cash orders without address validation, a prepaid discount, or an RTO flag on repeat refusers will bleed 20-30% of those orders back as returns, and no amount of homepage polish fixes that.
Cheap is a legitimate strategy. Cheap without knowing what was excluded is just a deferred invoice.
Where prices are going into 2026
Two forces pulling opposite ways. Theme quality and AI-assisted development have made the simple end of the market cheaper in real terms; work that took a week in 2020 takes two days. At the same time, checkout extensibility, Shopify Functions and app extensions mean the complex end now requires actual JavaScript engineers rather than Liquid generalists, and those people cost more.
Practical read for next year's budget: assume the standard D2C band holds around ₹2L to ₹5L, assume anything involving checkout logic, B2B pricing tiers or custom apps costs more than it did, and assume the ₹50,000 brochure build gets cheaper still. If you're weighing Plus, the licence is contracted per merchant and the honest answer is that most brands under roughly ₹8-10 crore GMV don't get their money back on it. The exceptions are wholesale-plus-retail businesses and anyone who genuinely needs checkout customisation for COD gating or bundling.
The questions that make two quotes comparable
Send these to both agencies in the same email. The answers will be more informative than the prices.
- List every custom section by name, with the hour estimate against each.
- Which paid apps are in the scope, and what do they cost per month, in total, at my order volume?
- Does migration include order history, customer records and a redirect map? Have you seen my data export?
- Who configures COD rules, pincode serviceability and the prepaid incentive, and is that in the price?
- What's the target LCP on a mid-range Android over 4G, and will you show me a Lighthouse run before handover?
- How many rounds of revisions, and what triggers a change request?
- What happens the week before Diwali if something breaks, and what does that support cost?
- Do I get the theme code and the GitHub repository?
Question seven catches more people than the rest combined. A build finished in August with no support arrangement is a build you'll be firefighting alone during your biggest sales week.
If you already have quotes on the table and want a second opinion on what they actually cover, send them over with your store URL and we'll mark up the scope gaps. Our store audit is free and takes about a week, and our Bengaluru team is ten minutes from Kudlu Gate if you'd rather do it over coffee.