PECUC

Context

PECUC has protected children's rights in Odisha since 1990 — with UNICEF and the European Union among its partners, and three decades of photographs to prove the work. Their website was built in 2001 and treated all of it as furniture. I redesigned the full 15-page site around the one thing they had that no design trend can fake: evidence.

People's Cultural Centre is a non-profit development organisation working for the rights of children, women, older people, indigenous communities and the rural and urban poor. It began with a children's festival attended by 5,000 children, and has since run more than 178 projects across 30 districts of Odisha and 4 districts of Delhi. It holds UN ECOSOC Special Consultative Status, is FCRA compliant and ISO certified, and the Government of Odisha has adopted some of its community models into state policy.

The problem

An organisation with UNICEF's logo, 178 projects and thirty-five years of photographs was presenting itself online in a way that made a first-time visitor doubt all of it.

The old site opened with a slider running eleven messages. Seventeen links sat above the fold before a single sentence explaining what PECUC does. Photographs of the children this organisation changed were shrunk to thumbnails beside links labelled homedetails.aspx?id=H1.

And the two links that matter most were broken. Donation pointed to an empty anchor. Volunteering downloaded a PDF to print and fill by hand. For a non-profit, those two links are the product — an organisation this credible, presenting itself this badly, loses donations it will never know it lost.

Stage 01 — The audit

I didn't run user research on this project. I didn't need to. I needed to open the site and count.

1. 37

   Navigation destinations

   Nine main items, eight utility links above them, and twenty more inside the dropdowns — "Who we are" held 5, "Theme" held 9. Everything equally important, so nothing was.

2. 09

   Disconnected properties

   A separate Blogspot blog. An online resource centre. A photo gallery and a video gallery on different templates. Press and publications. Annual reports. Board documents served as raw Word files. Each a different visual language, each a dead end.

3. 03

   Accounts of what year it was

   The footer read "All rights Reserved 2001-2023". The news section had entries from March 2025. The homepage said thirty-six years of service. All on one page.

None of this was neglect. PECUC were still publishing to this site in 2025. It was a site structurally unable to hold what the organisation had become.

Sitemap and IA

Twelve content pages and three utility ones, drawn as a tree before a single screen existed. Everything that used to be its own scattered property now hangs off one Resources parent.

Mapping every label to a URL and a page type is unglamorous work that catches naming mistakes here instead of in a build. It caught two:

• the footer links "Annual Report" at a path the tree calls Impact Reports
• Career is a dropdown with nothing designed behind it yet

Both are flagged rather than quietly fixed. A structure with a known gap is more useful to the next person than one pretending to be finished.

Stage 02 — Restructuring

Before any visual work, I rebuilt the map. A visitor arrives doing one of four things: understanding the work, verifying the organisation, giving money, or offering time. The old navigation served none of those — it served the org chart.

Thirty-seven destinations became seven, and the nine scattered properties were consolidated into a single filterable Resources hub with one visual language and a path from any piece of content to any other.

37 destinations

7 destinations

Stage 03 — Building the visual language

Stage 04 — Iteration

My first layout didn't survive client review, and it's still in the file because it earned its place in this story.

The client liked the type. Liked the colour. Didn't like the layout. So the visual language I'd built was right and the arrangement wasn't — a far better failure than the reverse, because a system that survives its first layout being thrown out is a system, not a lucky composition.

I redesigned the structure, kept the identity, and version two was approved. A round of small fixes followed, and the file went to development.

Version 1 — rejected

Version 2 — approved

Homepage

The approved layout carried through to the built homepage — photography-first, one heading, one clear next step, and the hierarchy the old slider never had.

Stage 05 — Designing the two broken journeys

There was no funnel to improve here. Donation and volunteering both had to be designed from a blank page.

Donation became a tiered calculator — frequency first, then named tiers tied to real programmes: Sradha Supporter, Grandparent Guardian, AADI Champion, Community Builder, Transformation Partner. Trust signals — FCRA, ISO, financial transparency — sit at the moment of commitment.

Volunteering became a staged signup, replacing a printable PDF.

1. 01

   Who you are

   Name, contact, location — everything a volunteer can answer without thinking.
2. 02

   What you're drawn to

   Which programmes and causes they want to work on.
3. 03

   Availability

   When and how often, so the match is real before anyone follows up.

Donation

Volunteering

The decision log

Every significant call, what drove it, and what it cost.

Before and after

Old-site capture on the left, new design on the right.

The hero — before

The hero — after

Programmes. Nine themes sat inside a dropdown, which meant the work itself was something you had to go looking for. The rebuild puts them on the page as a section you scroll past rather than a menu you open.

Programmes — before

Programmes — after

Two more pages were doing real organisational work behind a presentation that undid it.

Careers. The old page read like a noticeboard — roles as undifferentiated links, nothing about what it's like to work there. The rebuild leads with the work and the people, then the openings.

Careers — before

Careers — after

Press and media. Thirty-five years of regional coverage — the most credible proof an NGO has — was sitting in an undated list of raw links.

Press — before

Press — after

Outcome

Scope

15 pages — homepage through 404, privacy and terms.

Identity

A visual design guide: Anton/Inter type system, a blue-yellow palette drawn from their logo, and photography-first layout rules.

Structure

Navigation 37 → 7, clicks to content 4 → 2, nine sub-sites consolidated into one filterable hub.

Journeys

A tiered donation calculator and a staged volunteer signup, both designed from scratch.

System

A card-based component library built to survive non-designers adding content.

Status

Approved, handed off, in development.

Client's words

What I'd do differently

1. Document as I go

   I reconstructed this reasoning afterwards. The reasoning held; the record didn't.

2. Test the tiers with a real donor

   I set the amounts and programme mappings without talking to anyone who had given to PECUC before.

3. Design for the content editor

   The card system works when filled correctly. I didn't fully solve for a staff member pasting a headline twice as long as the template expects — which is exactly how the old site decayed.

What it taught me

1. Visual design is an argument, not a coat of paint

   Every choice here exists to make one claim: this organisation is real, and the work is real. Once I knew that was the argument, every decision got easier.

2. The best material is usually already there

   I didn't create PECUC's credibility — 35 years of photographs and UNICEF's logo did. My job was to stop the design from hiding it.