Advanced Capstone: A Complete Campaign Kit
Bring the Advanced level together: audit a four-asset campaign kit for contrast, margins, safe zones, phone size, alt text, file names and print size, fix every problem and check again.
In this lesson you will
- Plan a campaign kit from a written brief
- Audit several formats with one set of rules
- Fix problems and re-check
- Combine measurement with human review
About 30 minutes
One campaign, five outputs, one standard
The library festival is in three weeks, and Meera wants a complete kit: a post, a story, a video thumbnail and a flyer, all matching, all correct. Four volunteers have made drafts. On the day before the deadline she asks Priya, now the club’s design lead, to check everything. The old way would be to open each file and stare. Priya has a better way. She has a brand system, an accessibility habit, print knowledge, a file-name pattern and a checker. This capstone puts the Advanced level together, in the same way a professional’s last check before a launch would.
The brief
- A post at 1080 by 1350 pixels, a story at 1080 by 1920, a video thumbnail at 1280 by 720 and an A5 flyer with 3 millimetres of bleed, which at 300 dots per inch is 1819 by 2551 pixels (lesson 24).
- The brand system from lesson 21: ink text on white, purple for headings, and pink used only where ink text sits on it, never as text on white.
- Every text and background pair at a contrast of 4.5 or better (lessons 6, 11 and 22).
- Text inside the margin, and clear of the top and bottom 250 pixels on the story (lessons 8 and 16).
- Text of at least about 12 pixels when shown on a phone (lesson 5).
- Alt text for every image, and file names that match the pattern (lessons 22 and 27).
The audit
The program below holds the four drafts as data. Each has a kind, a file name, alt text, a canvas size and a list of text parts with a size, colour and position. A single audit function applies all the rules, and reports every problem with a category. Then the program applies the fixes and audits again. Finally it draws the corrected assets as real picture files and reopens them to confirm their sizes. The drafts contain deliberate faults of the kinds you have met across the course.
from PIL import Image, ImageDraw, ImageFont
import copy, os, re
T = {"bg": "#FFFFFF", "text": "#14112E", "primary": "#5A28C8", "accent": "#F06CC2"}
FONT = next(p for p in ("/usr/share/fonts/truetype/google-fonts/Poppins-Bold.ttf",
"/usr/share/fonts/truetype/dejavu/DejaVuSans-Bold.ttf") if os.path.exists(p))
# kind: (width, height, margin, unsafe band at top and bottom)
FORMATS = {"post": (1080, 1350, 90, 0), "story": (1080, 1920, 90, 250),
"thumb": (1280, 720, 60, 0), "flyer": (1819, 2551, 150, 0)} # flyer = A5 + 3 mm bleed at 300 dpi
NAME = re.compile(r"^[a-z]+-[a-z0-9]+-v\d+\.png$")
def lum(h):
f = lambda c: c / 12.92 if c <= 0.03928 else ((c + 0.055) / 1.055) ** 2.4
r, g, b = (f(int(h[i:i + 2], 16) / 255) for i in (1, 3, 5))
return 0.2126 * r + 0.7152 * g + 0.0722 * b
def contrast(a, b):
hi, lo = sorted((lum(a), lum(b)), reverse=True)
return (hi + 0.05) / (lo + 0.05)
def audit(a):
kind = a["kind"]
W, H, margin, unsafe = FORMATS[kind]
found = []
for name, text, size, colour, y in a["parts"]:
f = ImageFont.truetype(FONT, size)
if contrast(colour, T["bg"]) < 4.5: found.append(("contrast", kind + " " + name))
if margin + f.getlength(text) > W - margin: found.append(("margin", kind + " " + name))
if y < unsafe or y + size > H - unsafe: found.append(("safe zone", kind + " " + name))
if size * 390 / W < 12: found.append(("phone size", kind + " " + name))
if not a["alt"]: found.append(("alt text", kind))
if not NAME.match(a["file"]): found.append(("file name", kind))
if a["size"] != (W, H): found.append(("canvas size", kind))
return found
drafts = [
{"kind": "post", "file": "post-booksale-v1.png", "alt": "", "size": (1080, 1350), "parts": [
("title", "Book Sale Saturday", 150, T["primary"], 200), ("cta", "Come along", 60, T["accent"], 1100)]},
{"kind": "story", "file": "Story Final.png", "alt": "", "size": (1080, 1920), "parts": [
("title", "Book Sale", 150, T["primary"], 600), ("cta", "Come along", 60, T["text"], 1800)]},
{"kind": "thumb", "file": "thumb-booksale-v1.png", "alt": "", "size": (1280, 720), "parts": [
("title", "Book Sale on Saturday at the Hall", 110, T["primary"], 200), ("cta", "Free", 30, T["text"], 400)]},
{"kind": "flyer", "file": "flyer-booksale-v1.png", "alt": "Flyer", "size": (1748, 2480), "parts": [
("title", "Book Sale", 260, T["primary"], 400), ("cta", "Parkside Village Hall", 100, T["accent"], 1900)]},
]
before = [x for a in drafts for x in audit(a)]
kinds = {}
for k, _ in before: kinds[k] = kinds.get(k, 0) + 1
print("problems found:", len(before))
for k in sorted(kinds): print(f" {k}: {kinds[k]}")
final = copy.deepcopy(drafts)
final[0]["alt"] = "Autumn book sale poster"
final[0]["parts"][0] = ("title", "Book Sale", 150, T["primary"], 200)
final[0]["parts"][1] = ("cta", "Come along", 60, T["text"], 1100)
final[1].update(file="story-booksale-v1.png", alt="Book sale story")
final[1]["parts"][1] = ("cta", "Come along", 60, T["text"], 1500)
final[2]["alt"] = "Book sale thumbnail"
final[2]["parts"][0] = ("title", "Book Sale", 110, T["primary"], 200)
final[2]["parts"][1] = ("cta", "Free", 60, T["text"], 400)
final[3]["size"] = (1819, 2551)
final[3]["parts"][1] = ("cta", "Parkside Village Hall", 100, T["text"], 1900)
after = [x for a in final for x in audit(a)]
print()
print("problems after fixes:", len(after))
for a in final:
img = Image.new("RGB", a["size"], T["bg"]); d = ImageDraw.Draw(img)
for name, text, size, colour, y in a["parts"]:
d.text((FORMATS[a["kind"]][2], y), text, font=ImageFont.truetype(FONT, size), fill=colour)
img.save(a["file"])
print("files:", ", ".join(a["kind"] + " " + "x".join(map(str, Image.open(a["file"]).size)) for a in final))
problems found: 11
alt text: 3
canvas size: 1
contrast: 2
file name: 1
margin: 2
phone size: 1
safe zone: 1
problems after fixes: 0
files: post 1080x1350, story 1080x1920, thumb 1280x720, flyer 1819x2551
Read the results. The audit found 11 problems in the four drafts, across these categories: alt text (3), canvas size (1), contrast (2), file name (1), margin (2), phone size (1), safe zone (1). Look at what each means. The two contrast problems are pink text on white, which scores 2.75, the same failure found in lesson 11. The two margin problems are headlines that run into the right margin, the lesson 10 fault again. The safe-zone problem is the story’s call to action sitting at the bottom where the app covers it. The phone-size problem is a 30 pixel line on a wide thumbnail, which shows at about nine pixels on a phone. The three missing alt texts, the badly named story file and the flyer made without bleed are the kind of thing nobody notices until the printer or the scheduler rejects the file. None of these was caught by looking at the designs on a laptop.
Then the fixes: ink instead of pink for the two calls to action, shorter headlines, a call to action moved up, a bigger line on the thumbnail, alt text added, the file renamed and the flyer remade at the bleed size. The second audit reports 0 problems. The program saved the finished files at these sizes: post 1080×1350, story 1080×1920, thumb 1280×720, flyer 1819×2551. The checker is also a record: it shows exactly what was wrong and what was changed, which is useful for the project summary and for the next campaign.
What the checker cannot do
The checker found 11 problems, but it cannot say whether the festival poster is persuasive, whether the photograph is lovely or whether the headline is funny. Those need people. Show the kit to someone who has not seen it. Look at it at phone size, and in greyscale. Print a test of the flyer and hold it at arm’s length. Ask a person who uses a screen reader to try the files if you can. A kit that passes the checks and fails the human test is not finished.
Your capstone project
- Choose a real message of your own: an event, a product, a club or a cause.
- Write a one-page brief with sizes, brand tokens, rules and file names.
- Make a master design, and then the outputs, using templates and the brand kit.
- Run your own checks and keep a list of the problems and fixes.
- Add alt text, test colour vision in greyscale, and preview on a phone.
- Prepare print files with bleed, and export the right file type for each use.
- Write a short summary for the client, with files, dates and what to do next.
- Review: what would you do differently next time?
Treating a pass as the end. A checker finds known problems. A new campaign brings new ones. Whenever a mistake slips through, add a rule for it, so the same mistake never costs you twice.
You have now finished the whole Canva course: from choosing a size to running a campaign. The most valuable habit across all thirty lessons is the same one: measure what you can, and use your eyes and other people for the rest.
Practice for this lesson
Check what you learned
Five quick questions. Pick an answer and you will see straight away whether it is right and why.
Why audit all outputs in one program?
- The same rules apply across every format
- Programs are faster than people at taste
- It removes the need for design
- It creates the brand
One set of rules applied to every asset finds the same kinds of problem across the whole kit.
Why does the flyer have a canvas of 1819 by 2551 pixels?
- It matches Instagram
- A5 with 3 mm bleed at 300 dpi
- It is a square
- It is the thumbnail size
A5 plus 3 mm on each side, at 300 dots per inch, is 1819 by 2551 pixels.
What was wrong with the story’s call to action?
- It had too much contrast
- It was too large
- It was in ink
- It sat where the app covers the screen
It was placed in the bottom band that apps cover with buttons and names.
What can the checker not judge?
- Contrast ratio
- Whether the poster is persuasive
- Margins
- File names
Persuasion, beauty and humour need human judgement.
What should you do when a new kind of mistake slips through?
- Ignore it
- Remove the checker
- Blame the tool
- Add a rule for it
Adding a rule means the same mistake is caught next time.
Summary
- One set of rules applied to every format finds consistent problems.
- A checker finds the measurable faults, and people judge the rest.
- Record problems and fixes, and add a rule for each new mistake.
- The Canva course is complete: from a first design to a checked campaign kit.
Questions
Do I need code to run a capstone like this?
No. You can follow the same rules by hand with a checklist. The code shows what a careful review measures and saves time with many files.
How do I know my kit is good enough?
When it passes your measurable checks and a person who has not seen it can read, understand and use it on a phone and on paper.
What should I do after the course?
Make real work for real people, keep your design system and checklist, and build a small portfolio of finished projects.
