Three switches decide it, and only one has to flip. The Indian price band for each answer, the four things a rebuild does not inherit for free — and the two dates, 3 October and 13 May 2027, that quietly decide the rest.
Three switches decide it, and only one has to flip to put you in rebuild territory: (1) the platform changes — a website builder to WordPress, WordPress to custom code, one framework to another; (2) the site has to start remembering something it never stored — logins, orders, uploaded documents, a consent record, a rule about who may see what; (3) the URL structure changes. If all three stay off, you are buying a redesign — on an existing Indian site that is the lower half of the same published bands, roughly ₹25,000–₹50,000 for a brochure site and ₹35,000–₹75,000 on WordPress, because the content, the hosting and the addresses already exist. If any one flips, you are buying a rebuild, and rebuilds start at ₹1,00,000 for anything with logins, because the price is a new build plus a carry-over that almost nobody itemises. Count the switches before you take a quote: a vendor who quotes a rebuild for a purely visual problem, or a redesign for a data problem, has answered a question you did not ask. And the expensive error is the second one — ₹60,000 spent on a redesign that sits on a data model which cannot hold what you actually need, with the rebuild still owed afterwards.
Not sure which switches your site flips? Send us the URL and we will tell you ›
The design hours in a redesign and the design hours in a rebuild are roughly the same. That is not where the money goes. The entire price gap is the cost of carrying four things across: your URLs, your content, your data, and your accounts. A redesign inherits all four for free — nothing moves, so nothing has to be carried. A rebuild inherits none of them for free, and every one of the four is a separate line of real work: an exported URL list and a 301 map, a page-by-page content migration, a mapped database export with a verification pass, and the transfer of the domain, hosting, analytics, Search Console, gateway and repository accounts into your name. The reliable test of a rebuild quote is not the total. It is whether those four lines appear at all. A rebuild quoted without them has been priced as if it were a brand-new site on an empty domain — which is a real quote for a different project, and the difference arrives later as change requests or, worse, as traffic that never comes back.
Declared bias, stated plainly: rebuilds are our higher-value work and redesigns are not. ZoopCoder has a direct financial interest in you concluding “rebuild”, so this page is written to make that conclusion harder to reach, not easier. There is a section below titled four reasons to rebuild that are not good enough, and on the arithmetic that follows, most Indian small businesses reading this page should redesign and keep the difference.
Write down every change you want in plain words — one line each, no technical language — then run each line past these three questions. You are looking for a single yes. One yes anywhere on the list puts that item into rebuild territory; zero yeses across the whole list means the entire project is a redesign, whatever anyone calls it.
| Switch | What it looks like in real life | Verdict |
|---|---|---|
| 1. The platform changes | Wix or a builder → WordPress. WordPress → custom code. A bespoke PHP site → a JS framework. Also: the platform is abandoned, or nobody has the code. | Rebuild. Nothing carries automatically across a platform boundary — not the theme, not the plugins, not the editing workflow. |
| 2. The site must remember something new | Customer logins. Saved orders or a cart. Uploaded documents. A consent record. Roles — staff sees this, customers see that. A booking that has to not double-book. | Rebuild. This is a data-model change, and no amount of design work adds a table to a database that does not have one. |
| 3. The URL structure changes | Page addresses change. Sections merge or split. You move to a new domain, add a language prefix, or drop /index.php-style paths. | Rebuild — and the riskiest of the three. Every unmapped old address is standing thrown away. Needs a written 301 map. |
| All three off | Colours, layout, fonts, photos, copy, page order, a new logo, a rearranged menu, a better contact form, mobile fixes, speed fixes. | Redesign. Do not pay rebuild prices for it. |
The useful consequence of running the test line by line rather than on the project as a whole: quite often eleven items on a twelve-item list are pure redesign and one item — usually “let customers log in” — is the only thing flipping a switch. That is not automatically a rebuild. It is frequently a redesign now, at a fifth of the price, with the one stateful feature carved out as a separate, properly-scoped project later. Any vendor who cannot split a list that way has not read the list.
ZoopCoder does not publish a separate redesign rate card, and neither should anyone else — a redesign is quoted against the same bands as a build, and it lands in the lower half of them because the content, the domain, the hosting and the addresses already exist and do not have to be made or moved. Everything below traces to the bands on the ZoopCoder pricing page. If a quote you are holding sits above the band for your site type, ask which switch the vendor thinks flipped.
| What you actually want | Switch flipped? | Answer | Indian price band |
|---|---|---|---|
| “It looks dated” — new look, same pages | None | Redesign | ₹25,000–₹50,000 brochure · ₹35,000–₹75,000 WordPress |
| “It is slow on mobile” | None | Measure first, then fix | ₹0–₹15,000 in most cases — images, hosting plan, plugins |
| Add pages, add a blog, rewrite copy | None | Redesign / content work | Within the same band as above |
| Move off Wix / GoDaddy builder to WordPress | 1 | Rebuild | ₹35,000–₹75,000 + content carry (builders rarely export cleanly) |
| Sell online — catalogue, cart, payments | 2 | Rebuild | ₹50,000+ — see the e-commerce guide |
| Customer logins, saved records, roles | 2 | Rebuild (custom web app) | ₹1,00,000+ |
| Restructure the site / merge sections / new domain | 3 | Rebuild + redirect map | Band for the site type, plus the carry below |
| Turn it into a product other people pay to use | 1 + 2 + 3 | Not a rebuild — a new product | ₹1,50,000–₹5,00,000 — see the SaaS MVP guide |
This is the table to hold beside a rebuild quote. Each row is a line that should exist in the document. A rebuild quote that names none of them has not been scoped for your site; it has been scoped for an empty domain.
| Inheritance | Redesign | Rebuild | What it costs to carry — and what breaks if you skip it |
|---|---|---|---|
| URLs | Kept, free | Lost by default | A written 301 map, published on day one. Skip it and every old address 404s — rankings, backlinks and any live ads pointing at those pages all stop working at once. The only irreversible one on this list is this and the next-but-one. |
| Content | Kept, free | Re-entered page by page | Real hours, and the line most often silently handed back to the client mid-project. Ask explicitly: who types the content in, you or me? The answer changes the timeline more than any technical decision. |
| Data | Untouched | Needs a mapped migration | Enquiries, orders, accounts, uploaded files. Needs an export, a field mapping, an import and a verification pass against row counts. This is the one where a mistake is permanent — keep the old database read-only for 90 days after cutover. |
| Accounts | Untouched | Must be transferred | Registrar, hosting panel, analytics, Search Console, payment gateway, repository. Not technical work at all — and the single most common reason an Indian rebuild sits stalled for three weeks. Start it in week one. See who owns the code. |
| Analytics history | Continuous | Continuous if you keep the property | Free to keep, impossible to recreate. Reuse the existing GA4 and Search Console properties rather than creating new ones, and annotate the cutover date so you can tell a rebuild effect from a seasonal one. |
This is the part of the decision that is worth real cash and appears on no Indian agency page we could find. The capital-versus-revenue line in Indian income tax runs almost exactly along the redesign/rebuild axis, and it changes what the same rupees do to this year’s tax bill.
Website expenditure has generally been treated as revenue expenditure, deductible in full in the year it is incurred under section 37 of the Income-tax Act, 1961. In CIT v. Indian Visit.com (P.) Ltd. the Delhi High Court held that expenditure on a website produces no accretion to the fixed capital of the business — the purpose is to facilitate and promote the business, not to acquire an asset — and, importantly, that merely producing an enduring benefit does not by itself make an expense capital in nature. A redesign of an existing site sits comfortably inside that reasoning.
A ground-up rebuild that the business genuinely capitalises as new software is a different animal: it is depreciated rather than deducted, and computer software falls under “computers including computer software” in the depreciation schedule at 40% on the written-down value. And then there is the date. The second proviso to section 32(1) restricts depreciation to half the normal rate where an asset is acquired during the year and put to use for less than 180 days in that year. In an ordinary financial year, 3 October is the last day on which an asset can go live and still complete 180 days.
| Treatment of the same ₹1,50,000 | Basis | Year-one deduction | Cash effect at 25.17% |
|---|---|---|---|
| Revenue — typical redesign | s.37(1); Indian Visit.com | ₹1,50,000 (100%) | ₹37,750 |
| Capitalised, live on or before 3 Oct | 40% WDV, full rate | ₹60,000 (40%) | ₹15,100 |
| Capitalised, live on or after 4 Oct | s.32(1) 2nd proviso, half rate | ₹30,000 (20%) | ₹7,550 |
| Spread across the three | Same money, same project | ₹30,000 – ₹1,50,000 | ₹30,200 — about 20% of the project price |
Read that bottom row again: the same ₹1,50,000, spent on the same website, returns anywhere between ₹7,550 and ₹37,750 of year-one tax relief — a ₹30,200 swing decided by an accounting classification and a single calendar date. At the 31.2% effective rate a proprietor or firm at the top slab pays, the swing is larger still, around ₹37,400. The balance is not lost in the capitalised case — it unwinds over later years on the written-down value — but it is not cash you have this year, and rebuilds are precisely the projects that slip past early October.
The obvious bad idea, said out loud so nobody has to ask us: do not instruct a developer to write “redesign” on the invoice for a rebuild. Classification follows what was actually done, not the noun on the bill, and a mislabelled invoice is the first thread that comes loose if the treatment is ever examined. The legitimate version of the same optimisation is the honest one: if the work genuinely is a redesign, say so and claim it; and if you are capitalising a rebuild, put the go-live date in the project plan rather than discovering it in April.
References are to the Income-tax Act, 1961 and the depreciation schedule to the Income-tax Rules, 1962, current as at July 2026, and the 25.17% figure is the effective rate for a domestic company under the section 115BAA regime (22% plus surcharge and cess). Whether a specific spend is revenue or capital is fact-dependent and is contested often enough to be worth an hour of your CA’s time. This is general information about how the classification works, not tax advice on your facts.
Everything above assumes you are choosing. For a large number of Indian websites, switch two is going to flip on a deadline instead. The Digital Personal Data Protection Act, 2023 was operationalised by rules notified in November 2025 with a phased transition, and the substantive obligations on an ordinary Indian data fiduciary are set to apply from 13 May 2027, with penalties under the Act reaching ₹250 crore for a failure of reasonable security safeguards.
The obligations that touch a small business website are unglamorous and structural rather than cosmetic: an itemised notice in clear, plain language, available in English or any language in the Eighth Schedule to the Constitution at the user’s option; consent as a specific affirmative action rather than a pre-ticked box or an implied acceptance; withdrawal of consent made as easy as giving it; and erasure once consent is withdrawn or the purpose is served. Note what those four require of the software: somewhere to write a consent record, something to match a withdrawal against, and a path to delete.
That is the whole redesign/rebuild question again, arriving on a date. A brochure site whose contact form emails an inbox and stores nothing can meet most of this with copy changes and a written policy — a redesign. A site that quietly stores enquiries, accounts or orders in a database with no consent field and no deletion path generally cannot, because there is nowhere to write the consent and nothing to delete against. The practical consequence for anyone planning work in 2026: if you are going to rebuild anyway, build to this standard once, rather than paying for a rebuild now and a compliance retrofit in eighteen months. And if you are redesigning, this is the cheapest moment you will ever get to add the consent checkbox, the retention rule and the deletion route — because the developer is already in the codebase.
DPDP timelines are phased and have moved before; treat 13 May 2027 as the planning date rather than a certainty, and confirm your own classification and obligations with counsel. Nothing here is legal advice.
We sell rebuilds. Read this section knowing that. Each of these is a real reason Indian small businesses are sold a six-figure project, and each of them has a cheaper correct answer.
The number that argues against us, published anyway: a redesign that keeps the platform, the content and the URLs typically costs between a third and a fifth of a rebuild of the same site and carries almost none of its risk. On expected value, most small Indian businesses arriving at this question should redesign, keep the difference, and revisit in two years — and the honest sales answer, when the switch test comes back all-off, is to quote the redesign and say so.
| Project | Elapsed time | What sets the date |
|---|---|---|
| Visual redesign, same platform, same URLs | 1–3 weeks | How fast you approve the design. No migration, no redirects, no re-entry. |
| Rebuild — small brochure site | 2–3 weeks | Same as building it new, plus the content carry. |
| Rebuild — content-managed / WordPress site | 4–6 weeks | Content migration and the redirect map, not the theme. |
| Rebuild — adds logins, roles or payments | 8–14 weeks | It is a custom application now. See the timeline guide. |
| Account transfers, in any of the above | Start in week 1 | Registrar and gateway transfers are other people’s queues. This is what actually stalls Indian rebuilds. |
Run the three switches. Does it need a different platform? Must it remember something it does not store today? Do the URLs change? All three off means redesign, and on an existing Indian site that is ₹25,000–₹50,000 for a brochure site or ₹35,000–₹75,000 on WordPress. Any one on means rebuild, starting at ₹1,00,000 once logins are involved. Run it line by line on your wish list rather than on the project as a whole — it is common for one item out of twelve to be the only thing flipping a switch, and that item can often be carved out and done later.
A redesign is not a separate product with its own rate card — it is quoted against the same bands as a build and lands in the lower half of them, because the content, the hosting, the domain and the URLs already exist. On the published bands that is roughly ₹25,000–₹50,000 to restyle a brochure site and ₹35,000–₹75,000 on WordPress, with a small refresh of a few pages sitting below both. If a quote sits above the band for your site type, ask which switch the vendor thinks flipped — and note that a “redesign” which changes the URL structure is not a redesign whatever it is called.
Four situations, and not before. The site must start holding accounts, orders, uploaded files or role-based permissions and today holds none. The platform is closed or abandoned — a builder that will not export, or a bespoke system with no repository and nobody to call. The consent and personal-data handling has to change in a way the current forms and database cannot express, which is the DPDP situation described above. Or the stack is a genuine security liability where patching costs more than replacing. Everything else — dated, slow, hard to edit, unresponsive developer — is fixable for a fraction of a rebuild.
It can, and the mechanism is nearly always the same: URLs change and the old addresses are not mapped. A redesign that keeps every URL carries close to zero risk, because from a search engine’s point of view nothing moved. A rebuild changes addresses by default, and every old address that 404s instead of permanently redirecting throws away whatever standing that page had. The fix is real work but not complicated: export every URL, decide each destination, publish the 301 map with the new site, keep the sitemap accurate, and watch crawl errors weekly for a month. It is also the line most often missing from an Indian rebuild quote.
Website spend has generally been treated as revenue expenditure, deductible in full in the year under section 37. In CIT v. Indian Visit.com (P.) Ltd. the Delhi High Court held there is no accretion to fixed capital, that the site is a tool to facilitate the business rather than an asset acquired, and that an enduring benefit alone does not make an expense capital. A rebuild the business genuinely capitalises as software is depreciated instead, at 40% WDV — so on ₹1,50,000 that is a full ₹1,50,000 deduction on one treatment against ₹60,000 on the other. Classification follows the facts, not the word on the invoice. Confirm with your CA.
If the spend is capitalised, yes — by a surprising amount for one calendar day. The second proviso to section 32(1) halves depreciation where an asset is put to use for less than 180 days in the year, and in an ordinary financial year 3 October is the last day that still completes 180. Live on or before 3 October: ₹60,000 of depreciation on a ₹1,50,000 rebuild. Live on 4 October: ₹30,000. Nothing changed except the day. Rebuilds are also the projects most likely to slip, which is why the date belongs in the plan. A redesign deducted as revenue expenditure is unaffected.
It turns a design question into an architecture question, on a date. The rules notified in November 2025 phase in obligations for ordinary data fiduciaries from 13 May 2027, with penalties under the Act reaching ₹250 crore for a failure of reasonable security safeguards. What a small site has to produce: an itemised notice in plain language, available in English or any Eighth Schedule language at the user’s option; consent as an affirmative action; withdrawal as easy as giving; and erasure afterwards. A form that emails an inbox and stores nothing can usually get there with copy. A database of enquiries with no consent field and no deletion path cannot — that is switch two, flipping on a deadline rather than a decision.
Four things, and the price gap is mostly the cost of carrying them. URLs — kept free by a redesign, lost by default in a rebuild, needing an exported list and a 301 map. Content — re-entered page by page, and the line most often quietly handed back to you. Data — enquiries, orders, accounts and files needing a mapped export, an import and a verification pass, and the only one of the four where a mistake is permanent. Accounts — registrar, hosting, analytics, Search Console, gateway, repository: not technical work at all, and the usual reason a rebuild sits stalled for three weeks. Ask any rebuild quote to price all four as named lines.
A visual redesign keeping the platform, content and URLs is typically 1–3 weeks, because there is no migration, no redirect map and no re-entry. A rebuild is the same as building new: roughly 2–3 weeks for a small brochure site, 4–6 weeks for a content-managed site, plus the carry. A rebuild that adds logins, roles or payments is a custom application at 8–14 weeks. The variable that moves those dates most is not the developer, it is how fast approvals and content come back — which is why a quote naming your approval turnaround in working days is more honest than one naming only an end date.
Almost certainly not, and this is the most common way Indian small businesses are oversold. Looking dated is what a redesign is for. Slowness is usually one of three cheap things — uncompressed images, oversubscribed shared hosting, or plugin bloat — and none of them is the architecture. Measure first. If the site scores badly purely on images and hosting, a rebuild will not make it fast, it will just make it new. Fix the images and the hosting, re-measure, and only then decide whether what remains is worth a redesign or a rebuild.
Paste the URL and the list of changes you want. We will tell you which of the three switches your list actually flips — including, often, that none of them do and you should be buying a redesign at a fifth of the price. If it is a rebuild, the quote comes itemised with the 301 map, the data migration and the account transfers as named lines.