Case study / Direct to consumer, anonymous

Four Product Lines That Read as Four Companies

A Mumbai based direct to consumer brand had grown from one product line to four. Their own repeat customers did not know the other three existed. They came to us asking for four logos.

At a glance

Client: Not named. The client did not give permission, so this page describes the work and not the company.
Sector: Direct to consumer retail, sold online and through marketplaces
Base: Mumbai, India
The ask: Four new logos, one per product line
What we did: Brand architecture, parent identity, naming system, guidelines, a staged migration order

4

Product lines that had grown up separately, each with its own name and its own look.

1

Parent brand identity we built, with the four lines expressed as descriptors under it.

0

New logos delivered, against a brief that asked for four. That was the main recommendation.

3 of 4

Lines renamed. One kept its own name because it had genuinely earned equity of its own.

This client cannot be named, which means you cannot verify these counts the way you can on our GREVESGROUP page. Judge this one on whether the reasoning holds up, not on the numbers. The section on what we did not measure is below, and it includes a figure the client gave us that we have chosen not to use.

How a brand ends up looking like four brands

The company started with one product, did well, and added a second line. Then a third, then a fourth. Each launch happened on its own schedule, under its own pressure, and the branding for each was decided at the time by whoever was running that launch. A freelancer for one, an agency for another, the founder and a template for a third.

Every individual decision was reasonable. Nobody ever sat down and decided how the four lines related to each other, because at no point was that the urgent question. The urgent question was always shipping the next line.

The symptom that brought them to us was specific and measurable from their own side: customers who had bought repeatedly from one line were not buying from the others. Not because they had tried them and preferred something else. Because they did not know the other lines were the same company.

The brief they sent was for four new logos. The thinking was that the lines looked inconsistent, so better logos would fix it. That is a reasonable diagnosis if you are inside the business and looking at four logos side by side on a slide.

What the problem actually was

Four logos would have made the inconsistency prettier without touching the cause. The cause was that no decision had ever been taken about the relationship between the lines, which meant the customer had to work it out, and the customer was not going to.

There was no stated relationship between the lines

Brand architecture is the answer to one question: how do these things relate to each other. There are only a handful of workable answers, and the business had never picked one. Four separate identities is itself an answer, and it was the one they were accidentally giving.

The parent company was invisible at the point of sale

On the packaging, on the marketplace listings and in the ads, nothing told a buyer of one line that the others existed or were related. The company existed on the incorporation documents and nowhere a customer would look.

The naming had no rule, so line five would repeat the problem

Each line had been named on instinct. With no convention, the fifth line would be named on instinct too, and the architecture would come apart again as soon as it was fixed.

What we could not simply change

This is where brand architecture gets difficult. The theory is simple. The constraints are what make it a real project.

▶Printed packaging already paid forThere was stock on hand and more on order. A change that required everything to be reprinted at once was not going to happen, and recommending it would have meant the whole project stalled.
▶Marketplace listings carry historyOn the large marketplaces, a listing accumulates reviews and ranking history. Changing the brand name on a listing can put that at risk. Any renaming had to account for which listings could absorb the change and which could not.
▶One line had real equityOne of the four was genuinely known by name. Folding it into a parent naming convention would have thrown away the one piece of recognition the company had actually built.
▶No appetite for a single big switchThe business could not go dark for a rebrand, and did not want to confuse existing customers mid season. Whatever we designed had to survive being rolled out in pieces, over time, with old and new in market together.
▶A small internal teamThere was nobody whose job was brand. Whatever we handed over had to be usable by a small team with other priorities, which rules out a system that needs a designer to interpret it.

What we delivered

The deliverable list is shorter than the brief asked for, and the thinking behind it is longer. That is the right shape for this kind of problem.

A brand architecture decision, written down

The relationship between the parent and the four lines, stated explicitly, with the options we rejected and why. This is the document the business actually needed, and it is the one nobody had.

A parent brand identity

One identity for the company itself, designed to sit above the lines rather than compete with them, and to work small on packaging where it would mostly appear.

A naming system with a rule for line five

A convention that says how a new line gets named and what it inherits from the parent, so the next launch does not reopen the question.

Brand guidelines built to be used

Written for a small team with no designer, which meant fewer rules, stated more plainly, with the common mistakes shown rather than described.

A staged migration order

Which touchpoint changes first, which can wait for the next print run, and which marketplace listings should be left alone until the rest has settled. Sequenced by risk rather than by visibility.

A visual system that differentiates without separating

Each line stays distinguishable on shelf and in search results while reading as part of one family, which is the actual design problem underneath the four logo request.

What we decided not to do

The brief was four logos. Four logos is more design work, more revisions and a bigger invoice than what we recommended instead. These are the choices that cost us money.

✕We did not design the four logos we were asked forIt was the brief, it was billable, and it would have left the business exactly where it started with nicer artwork. We said so before quoting, which meant risking the job to make the point.
✕We did not rename the line that had equityConsistency would have been tidier and the system would have looked better in a presentation. That line was the only name any customer actually recognised, so it kept its name and the architecture bent around it.
✕We did not recommend a single launch momentA coordinated relaunch is the version that makes a good case study. It also needed money for packaging that did not exist and risked confusing customers mid season. Staged was less impressive and more likely to actually happen.
✕We did not touch the marketplace listings firstThey are the most visible surface, so they are the tempting place to start. They also carry the review and ranking history, which made them the highest risk and therefore the last thing to change.
✕We did not build a system that needed usIt would have been straightforward to design something that required a designer for every new asset, which produces ongoing work for us. The guidelines were deliberately written so the internal team does not have to call us.

What changed

Structural changes, stated as things that are now true about the business and were not before. Commercial effects are dealt with in the next section.

✓The relationship between the lines is now a decisionIt is written down, with the rejected options recorded. Anyone joining the business can read what the architecture is instead of inferring it.
✓The parent company exists where customers lookIt appears on the touchpoints a buyer actually sees, so a customer of one line has some way of knowing the others are related.
✓Naming the next line is no longer a debateThere is a rule. The fifth line inherits a structure instead of starting another round of instinct.
✓The team can produce assets without a designerThe guidelines were written for the people who actually use them, which was the point of writing fewer rules more plainly.
✓The rollout can happen in piecesNothing in the system requires everything to change at once, so existing packaging stock can run down instead of being written off.

What we have not measured, and will not claim

This is the section that matters most on an anonymous case study, because you have no way to check anything else. Here is what we do not know and what we have deliberately left out.

⚠The client reported a brand recognition figure of about 35 percent from a survey they ran after the work. We are not using it as a result on this page. We did not design that survey, we did not see the sample or the question wording, and we did not audit the method. It may well be accurate. It is just not evidence we can stand behind, and a number we cannot defend is worth less to you than an honest gap.
⚠We have not measured the cross sell effect, which is the thing the project was actually meant to fix. That would need their order data over a meaningful period, attributed properly, and we do not have it.
⚠No revenue, order value or margin figures appear here. Those are the client's numbers and we have not asked to publish them.
⚠The migration was staged over time, which makes any before and after measurement harder rather than easier. Anyone telling you they cleanly measured the effect of a staged rebrand should be asked how they separated it from everything else happening in the business.
⚠Because the client is not named, you cannot independently verify any of this. That is a real limitation of this page and we would rather say it than let the format imply otherwise.

How to tell whether you have this problem

You cannot check this case study, because the client is anonymous. You can check whether the same thing is true of your own business, which is more useful anyway. Four tests you can run this week.

01 Ask a repeat customer to name your other lines

Not a survey. Three or four actual customers, on the phone. If they cannot name the other lines, or are surprised they are yours, you have the problem this page describes, whatever your logos look like.

02 Look at your packaging from arm's length

Put the lines next to each other and step back until you cannot read the small type. If nothing at that distance says these are related, then nothing says it to a shopper either.

03 Try to write down the rule for naming your next line

If you can write it in one sentence, you have an architecture. If you find yourself explaining the history of each existing name instead, you do not, and the next launch will make it worse.

04 Search your own brand name and see what comes back

Search the parent company name and look at what a buyer would find. If the lines do not appear, or appear without any visible connection to each other, that gap is the same problem showing up in search.

What we would do differently

We should have asked for the order data at the start, before doing any design work, and set up a way to look at cross line purchasing over time. We did not, which is why this page has no result for the thing the project was for. That is our mistake rather than the client’s, and it is the reason we now ask for baseline data before quoting on architecture work.

We accepted the client’s recognition survey at face value when they shared it, and used it in conversation for a while before deciding it was not solid enough to publish. We should have asked how it was run the day we received it. Checking a number is a five minute conversation, and doing it late meant we had already repeated it.

The staged migration order was right, but we delivered it as a document rather than as dates. A sequence without dates tends to stall after the first easy step. Next time the order comes with a suggested calendar attached, even a rough one, because that is what makes it get done.

We also under invested in the handover. The guidelines were written to be usable without a designer, which was correct, but we did not sit with the team and watch them try to use it. An hour of that would have found the parts that were clear to us and not to them.

The services this engagement drew on

If this sounds like your situation, these pages set out the method and what it costs.

Other case studies

A named client, a similar diagnosis

A corporate investigation firm asked for a logo. The problem was that the brand contradicted the pitch.

A practice nobody could find

Identity, website and company profile for a retreat leader whose only channel was word of mouth.

A wildlife trust with nothing to show

Brand, website and an annual report for a conservation trust whose work happened entirely in the field.

Questions people ask about this one

Why is this client not named?

Because they did not give permission. We name a client only when they have agreed to it, which is why our GREVESGROUP case study carries a name and this one does not. If you work with us, your name does not appear anywhere without you saying yes, including in a pitch to someone else.

What is brand architecture, in plain terms?

It is the decision about how the things you sell relate to each other and to the company that makes them. Whether the company name leads and the products sit underneath it, whether each product stands alone, or something in between. Every business has an architecture whether or not anyone chose it, which is the whole problem in this case.

You turned down the work you were asked to do. Does that happen often?

We turned down part of it. We said the four logos would not fix the problem and proposed architecture instead, which was a smaller design job. It happens when the brief and the problem do not match, which is more common with branding than with websites, because the symptom is usually visual and the cause usually is not.

How do you decide which product lines keep their own name?

By whether the name carries recognition that would be destroyed by changing it. That is a question about what customers already know, not about what fits the system. In this case one line out of four passed that test, so it kept its name even though the system would have looked neater without the exception.

Is a staged rollout not just a worse rebrand?

It is a slower one, and it means old and new sit in market together for a while, which is genuinely untidy. It is better than the alternative when the alternative is writing off packaging stock or not doing it at all. The judgement is about what the business can actually absorb, not about what looks best in a deck.

We have two lines, not four. Is it too early to think about this?

Two is the right time, and it is much cheaper than four. The decision costs almost nothing when you have two lines and no printed stock to worry about. It gets expensive in proportion to how much material already exists carrying the old assumption.

Can you show us the actual guidelines document?

Not this client’s, because it carries their name and their products throughout. We can show you the brand guidelines we produced for PIXELS TRAIL itself, which is a real document built the same way, and we can walk through the structure of what a client receives.

What does brand architecture work cost?

The architecture decision itself is a relatively small piece of work, usually a few weeks, and the cost sits well below a full identity project. What moves the number is how much has to be redesigned afterwards and how many touchpoints are in scope. The branding service pages linked above carry indicative pricing in INR with USD equivalents.

How long until it is visible to customers?

With a staged rollout, the first visible changes come quickly because they are the digital ones, and the physical ones follow print runs. That means months rather than weeks for the full picture, which is one of the honest costs of not doing a single big launch.

Who did the work?

Swaranjit Singh, who runs PIXELS TRAIL. Branding engagements here are run by the person who does the thinking, not handed from a strategist to a designer to an account manager.

More than one product line and a nagging feeling they do not add up?

Run the four tests above first. If any of them worries you, send us what you have, the lines, the packaging and the names, and we will tell you whether this is an architecture problem or something simpler. If it is simpler we will say so.