Brand Identity and Graphic Design Case Studies
Creative5 Case StudiesUpdated Sep 2026
These five cases come from the working life of brand and design teams that have to make hard calls with limited money, mixed evidence, and public attention. Each one opens with a situation that looks ordinary on the surface, then tests how a team reads audience intent, judges channel performance, protects brand consistency, and proves whether conversion actually moved. The organizations are fictional, but the decisions are the sort that come up every few months in agencies and in-house studios alike. Read them as working notes rather than templates. The figures belong to each organization, and the reasoning matters more than any single number.
Case Study 1
Launching a regional campaign on a fixed budget
Tallow and Pine, a family owned candle and home fragrance brand selling through twelve regional stores and a small web shop
The situation
Tallow and Pine had a fixed launch budget of 42,000 dollars for its new winter collection, the largest the company had attempted. The brand already had a clear visual identity: warm amber tones, a hand-set serif wordmark, and photography shot in the owner's own kitchen. The question was where to spend. The marketing lead, Dana, had three candidates: paid social in two metro areas, a sponsored placement in a home decor newsletter with a modest readership, and in-store sampling at the twelve locations. Each promised something different. Social could reach a broad audience quickly, the newsletter offered a more engaged reader, and sampling could put product directly into hands. Leadership's brief was simple: show a measurable return by the end of the quarter.
The challenge
The budget could not cover all three. Paid social alone would consume most of it if the team wanted enough impressions to read anything from the data. The newsletter was cheap, but its readers overlapped with the existing email list, so it might only reach people who already knew the name. Sampling was expensive per contact, yet it was the only option that put the scent in front of someone before they chose. Dana needed a plan that compared reach, cost per conversion, and a measurement method that would survive a skeptical finance review.
Analysis
Start with intent before channel. Shoppers searching for a winter gift or scrolling for home updates are at different points, and a channel that reaches many people in a passive mood will look cheap on cost per click while converting poorly. The newsletter audience was closer to purchase intent, since they had opted in to a decor publication, but its size capped what the budget could return. Sampling carried the highest cost per contact and the strongest evidence of a real purchase, because the staff could count every sale made within the same visit.
Teams grounded in Brand Identity and Graphic Design tend to insist that every channel carry the same visual system, because a sample card or a newsletter header that looks unlike the storefront teaches the customer nothing. Tallow and Pine's amber palette and serif wordmark had to appear identically across all three. Consistency also gave the team a way to compare results: when the creative is held constant, differences in response point to the channel rather than the artwork. The measurement plan rested on three things: a tagged link for each channel, a discount code printed only on sampling cards, and weekly store sales compared against control regions.
Actions taken
- Cut paid social to a two-week test in one metro area, with a fixed daily cap and the same amber creative used in stores so the ad reads as the same brand.
- Move the newsletter sponsorship to a single placement with a unique tracking link, and ask the publisher for a click report broken out by reader segment before committing the full fee.
- Run sampling at five of the twelve stores chosen to match the two test regions, with staff logging every sale that follows a sample within the same visit.
- Hold back a share of the budget until the midpoint review, so the strongest channel can receive the remainder rather than the spend being split evenly by default.
Results
At the midpoint, paid social drove plenty of visits but few orders, and its cost per order sat well above the others. The newsletter sent fewer clicks than hoped, yet a larger share of them ended in a purchase. Sampling gave the clearest evidence: test stores outsold their controls, and staff logs tied many of those sales to a sample taken in the same visit. The held-back funds moved into sampling and a second newsletter placement. Finance accepted the result because control stores, not platform reporting, supplied the comparison.
Questions for discussion
- Which of the three channels would you have cut first, and what evidence would have changed that choice?
- How would the measurement plan change if the brand had no control stores and could only rely on tagged links?
- What would a fixed budget do to the decision if the newsletter had a larger audience with little overlap with the email list?
Case Study 2
Repositioning a nursery seen as a hardware store
Greenwell Supply Co., a regional garden center chain with six stores and a small online catalog
The situation
Greenwell Supply had spent several years describing itself as the place serious gardeners go for expert plant advice and thoughtful landscape planning. The logo, a dark green mark on warm kraft tones, and the store copy reflected that ambition. A customer survey of 1,240 past buyers told a different story. Most people described the stores as the place for cheap soil, cheap tools, and quick weekend fixes. Plant knowledge and design help scored far below price, and many shoppers had never asked staff a question. The gap was a message problem, not an identity problem.
The challenge
The leadership team wanted a proposition that would move the stores upmarket without alienating the weekend crowd who came for soil and mulch. A price-led story would be easy to write, but it would confirm what customers already believed. A pure expertise story risked sounding like a boutique with no reason to shop there on a Saturday. The team had about one season to test two or three positions, and any result had to show whether shoppers would book help, not just say they liked the idea.
Analysis
Intent is the first thing the survey exposed. The weekend shopper arrives with a job to finish, a raised bed to fill or a fence line to fix, and wants the bag of soil in the cart quickly. A design consultation asks for a different kind of visit, one planned and tied to a larger project. The store signage and the weekly email drew steady traffic for promotions, while the online catalog's plant pages got attention but few bookings. Greenwell had been measuring success by transactions per visit, which rewarded the price-led behavior the repositioning was meant to change.
Brand consistency was the quiet problem. The logo and the color palette promised care and expertise, but the shelf tags were printed in aggressive red sale style, and the staff uniforms gave no clue to who could answer a plant question. A proposition is only as credible as the cues around it. Conversion evidence had to come from a controlled test, not opinion. The team built three landing pages, each carrying the same visual system but a different promise: plant guidance, design planning, and seasonal value. They tracked booking rates for a free twenty-minute consultation, which measured whether the expertise claim was believed.
Actions taken
- Rewrite the store signage around plant guidance first, keeping price visible but no longer the headline on the entrance wall during the test period.
- Launch three landing pages with the same visual system and a different promise, and send each to a matched slice of the email list.
- Put a staff member with plant training in a visible role at the garden desk, and have the uniform carry the same green mark used on the website.
- Replace the red sale tags on expertise categories with neutral tags that show a short care note, and track whether shoppers who read the note add more items.
Results
The plant guidance page converted best among people who had already browsed the catalog, and the design page drew few visits but the highest booking rate per visit. The value page brought the most traffic and the least interest in consultations. Shoppers who booked a session spent more on their appointment day than the store average. The team did not claim a full repositioning. The narrower result was useful: plant guidance was believable when staff and signage supported it, and design needed a longer runway.
Questions for discussion
- Which of the three promises would your own team have been most tempted to lead with, and why?
- How should a regional chain weigh a higher booking rate from a small group against the traffic a price message still brings?
- What evidence would convince leadership that a design story is worth a longer test?
Case Study 3
Diagnosing a falling conversion rate with steady traffic
Tidewater Lantern Co., an online lighting retailer shipping from a single warehouse and two partner showrooms
The situation
Tidewater Lantern sells outdoor and entry lighting through a web shop that had enjoyed a steady run of organic search and paid campaigns. In the third week of the quarter, the conversion rate began to slide while sessions stayed almost flat. Each day the store took about 4,800 visits, and daily orders fell from roughly 140 to about 95 over the next fortnight. The brand team had recently refreshed the product photography and rewritten product descriptions to sound warmer and less technical. Leadership's first instinct was to blame the new copy, which had gone live a few days before the drop began.
The challenge
The copy was an easy suspect, but it was not the only change. Ad spend had been reallocated in the same fortnight, a checkout update had shipped, and the mobile share of traffic had risen. Each of these could move the number on its own. The team needed to inspect the journey from first visit to confirmed order, decide which attribution model to trust, and rank the candidate fixes before anyone spent a week rewriting pages again. Experiments were expensive in staff time, so the order mattered.
Analysis
Intent comes first here, since a falling rate with steady traffic often means the people arriving have changed, not only the pages they see. The team split sessions by source and found that paid social had grown its share of visits in the same window, and those visitors browsed several products but rarely searched for a specific fitting. Organic visitors, by contrast, still arrived with a clear product in mind and converted at a steady rate. The first-click and last-click reports disagreed about which channel deserved credit, so the team stopped relying on either report alone and compared source cohorts across the same dates.
Brand consistency became relevant once the source data was clear. The new descriptions sounded warm, but the spec tables still used the old technical labels, so a shopper comparing two fixtures could not match the wattage or the mounting height to the copy. Mobile visitors, who had grown to a large share of sessions, were landing on pages where the warm headline pushed the price below the fold. Conversion evidence came from session replays and a funnel by device. The funnel showed the largest drop between product page and cart, and the replays showed mobile users tapping the spec tab repeatedly before giving up.
- 1Source mix shiftPaid social grew its share of sessions in the same fortnight.
- 2Product page clarityWarmer copy removed the fixture specs that comparison shoppers needed.
- 3Mobile spec tabRepeated taps on a collapsed spec panel suggest missing answers.
- 4Cart and address stepA new postcode check added a field that some mobile users abandoned.
- 5Attribution reviewCompare source cohorts before crediting or cutting any channel.
Actions taken
- Restore the technical spec labels on every product page, keeping the warmer headline above them so the copy tone survives without hiding the facts.
- Move the spec panel out of its collapsed tab on mobile and test that change against the current layout on a random half of mobile sessions.
- Remove the new postcode field from the cart, collect the delivery area on the confirmation page, then check whether abandonment at that step falls within two weeks.
- Hold paid social spend at its current level for one fortnight while the cohort comparison runs, so the team can judge the channel on conversion rather than on the reports alone.
Results
The spec restoration and the mobile panel change shipped first, since both were cheap and easy to reverse. Mobile conversion recovered most of the gap within a fortnight, while desktop held steady. Moving the postcode field lowered abandonment at the cart step for every visitor. Paid social was held flat, and the cohort comparison showed its visitors converted below organic, though by less than the first-click report had suggested. Daily orders climbed back to about 130, still short of the earlier peak. The team concluded that the copy had contributed to the drop, but it was not the cause.
Questions for discussion
- If the copy change had been the only recent edit, what would you have checked first to confirm or rule it out?
- How would you decide whether a channel that reports well on first click deserves more budget?
- Which of these fixes would you have run as an experiment rather than shipping outright, and what would the test need to measure?
Case Study 4
A product safety rumor spreading through social channels
Marlow and Finch, a family owned kitchen appliance maker selling through independent retailers and its own website
The situation
Marlow and Finch makes countertop kitchen appliances and sells them through independent retailers and its own site. In the first hour after a short video appeared online, a viewer claimed that a blender in the current line overheated and cracked its base. The clip was posted by a small account, but within a day two consumer groups and a cooking blogger had shared it. The social team's first reply, posted on the main account, called the claim misleading and urged viewers to trust the brand. The reply offered no facts, and within hours the comments held a growing list of similar stories, some from people who did not own the model.
The challenge
The company had no internal evidence yet, only a return log showing a handful of warranty claims for the model over the past few months. Retail partners were asking whether to pull stock, and the social team needed to stop answering rumors one comment at a time. The brand's reputation rested on plainspoken reliability, which made defensiveness costly. The team had to establish what was known, decide who needed to hear it first, and set indicators that would show when the conversation had actually calmed rather than just gone quiet.
Analysis
Audience intent in a crisis splits quickly. Some people are worried owners looking for a safety answer, some are shoppers deciding whether to buy, and some are simply hunting for a story to share. A single public reply reaches all three and satisfies none. Channel performance showed the problem: the main account had the largest reach but the least credibility on the claim, while the retailer's product pages drew direct questions from people already in the purchase stage. The team mapped each group to the channel where it actually looked for answers, which turned out to be the help desk and warranty inbox, not the feed.
Brand consistency mattered because the usual voice, warm and a little playful, sounded evasive when a safety question was in the air. The team kept the friendly tone but dropped the jokes, used the same plain headline on every channel, and signed each statement with the name of the engineering lead. Conversion evidence came from the retail side. Sales on the affected model dipped in the first two days, but the sell-through on the sister models held steady, which suggested buyers were reacting to a specific claim rather than to the brand. That distinction shaped what the team would promise in the recovery plan.
- First hourClip posted, first reply drew criticism for offering no facts.
- Within the first dayTwo consumer groups shared the clip, and retailers asked whether to pause stock.
- Second dayEngineering tested the model, and a verified report was published with the test method.
- End of week oneReturns were compared with sister models to test the pattern.
- Two weeks laterRecovery indicators reviewed: complaint volume, retailer orders, and share of voice.
Actions taken
- Publish a single factual statement on every channel that names the tested model, the test method, and the engineer who signed it off.
- Send retail partners a short briefing with the same facts, a stock recommendation based on the test results, and a named contact for questions.
- Route warranty inbox and help desk replies through one template so that owners hear the same answer whether they write on social or by email.
- Set four recovery indicators before the next post goes out: complaint volume, retailer order trend, warranty claim rate, and share of positive mentions.
Results
The verified test report went up on the second day and was shared more widely than the original clip, partly because retailers linked to it. Complaint volume peaked on day two and fell steadily afterward. Retailer orders dipped, then recovered once partners had the briefing, and warranty claims stayed within the usual range. The team stopped posting the defensive line and replaced it with the report. Two weeks later, the indicators showed a conversation that had cooled, not just gone quiet.
Questions for discussion
- What is the smallest set of facts you would publish in the first hour, and what would you hold back until testing finished?
- How would you decide which stakeholder group to brief first when retailers, owners, and the media all want answers at once?
- Which recovery indicator would you trust least, and what second measure would you add to check it?
Case Study 5
Auditing uneven content before the next refresh cycle
Harbor Street Clinics, a group of four community clinics that publishes patient guides and articles online
The situation
Harbor Street Clinics publishes about 340 patient guides and articles across its website, email newsletter, and waiting room handouts. The library grew over several years through contributions from clinicians, front desk staff, and an outside agency, and the results were uneven. A small set of guides on common conditions drew most of the traffic, while many articles had fewer than a dozen visits a month. The writing ranged from dense and clinical to chatty, and the visual templates had been applied inconsistently. Some pages had no named reviewer, and a few carried advice the clinical leads had since changed.
The challenge
The next refresh cycle had a budget for 60 updates, which covered less than a fifth of the library. The editorial lead, Renee, needed a way to decide which pages to refresh, merge, or retire without relying on opinion or on whoever complained last. Approval was slow, since each change needed a clinical reviewer, a brand check, and a sign off from the operations director. The audit had to cover topics, formats, search intent, and approvals, and produce a ranked list the team could act on.
Analysis
Search intent is the first filter. A page titled as a general condition overview draws people still researching, while a page on a specific medication question draws people who need an answer before a visit. The audit compared each article's title and opening section against the questions that actually appeared in site search and in the clinic's patient messages. Channel performance then showed where the content was working. Guides linked from the appointment reminder email drew most of the repeat visits, while articles shared only on social media spiked briefly and vanished. A high page view count did not always mean the page helped anyone.
Consistency and conversion evidence sat together in this audit. The team looked at whether each page carried the current template, a named reviewer, and a visible date of last review, and it checked whether the page led to the next step, such as booking an appointment or calling the front desk. Pages with a clear next step and a recent review produced most of the booked appointments that could be traced to content. Pages without a next step produced reading time but almost no action. The audit scored each page on those signals, which kept the discussion anchored to evidence rather than to who had written the piece.
Actions taken
- Assign every article a named clinical reviewer and a review date, and set a rule that pages without both are hidden from search until they have one.
- Score each page on search demand, booked appointments traced to it, and template compliance, then sort the library into refresh, merge, and retire groups.
- Spend the budget for 60 updates first on high-traffic guides that link to a booking step, since those pages return the most value for each hour of clinical review.
- Create a shared approval path with a two-day clinical turnaround and a brand check that uses the existing template checklist, so the operations director signs off only on exceptions.
Results
The audit ranked 340 pages into three groups. About 60 were refresh candidates, 90 were merged into stronger guides, and the rest were retired or left alone after review. The first refresh wave went to guides linked from appointment reminders, where booked appointments traced to content rose once the next step was clearer. Merged pages carried traffic over to their replacements within a few weeks. Review turnaround shortened, though clinician time remained the real limit. Voice across the library still varied, but it now varied inside a shared template.
Questions for discussion
- How would you weigh a page with high traffic and no next step against a quieter page that reliably leads to a booking?
- Who should hold the authority to retire a clinical article, and what evidence would they need before doing so?
- What would you change in the approval path to shorten turnaround without weakening clinical review?
Learning Resources
Creative CASE STUDIES
China, India, United States, Indonesia, Pakistan, Nigeria, Brazil, Bangladesh, Russia, Japan, Mexico, Philippines, Vietnam, Egypt, Turkey, Germany, United Kingdom, France, Italy, Canada, South Korea, Australia, Spain.