David Wood's Seven Pillars of YouTube

The teacher

David Wood and his channel

Acts17 Apologetics was David's original channel. His channel today is Apologetics Roadshow. YouTube began for him, together with Nabeel Qureshi, as a place to store videos of speaking engagements, debates and the occasional announcement.

Years ago, Quran burnings began in the United States. David looked at the numbers on a video he had made about them and saw something he had not expected. He was reaching people in countries that are usually closed to this kind of message.

We just entered a new age of humanity where we can reach anyone in those countries with an internet connection about alternatives to Islam.

So David made YouTube his main work, and speaking and debating second. Some of his videos now do best outside the West.

Christians for 14 centuries couldn't have even dreamed of having these opportunities to reach people around the world... I think we need to pay attention to the times we're living in.

His style is his own: bold, sarcastic, full of cultural references, built from years of talking with Muslims and learning what lands. Above all, he found that good content is what makes people watch. The pillars that follow are about the other half, how that content gets in front of people.

A · Before the pillars

Have a plan

Putting a video on YouTube is not the same as being seen.

B · Before the pillars

The YouTube algorithm

Every thumbnail you see on YouTube is there because the algorithm chose to show it.

Pillar 1 of 7

Video title

A title is the first promise you make, to a person and to the algorithm.

Pillar 2 of 7

Description

The description is how you tell YouTube what your video is about, since it cannot watch it.

Pillar 3 of 7

Keywords

Know the words people use, then use them where people will see them.

Pillar 4 of 7

Subtitles

Many people will watch your video with the sound off.

Pillar 5 of 7 · part one

Thumbnails

Your thumbnail is the one tactic aimed at a person's curiosity, not the algorithm's.

Pillar 5 · part two

Stand out in both

In search, the viewer is looking for an answer, and your title, thumbnail and description should all say you have it. In browsing, the viewer was watching something else and never looked for you. In both, David says the title and the thumbnail have to be clear and compelling, in a very small space and a very short time.

Four questions Echo suggests you ask of any thumbnail:

Pillar 5 · taken apart

The sticker cut-out

David's signature: a face cut out with a thick white outline on the right, huge words on the left, and often one red arrow. Echo calls it the sticker cut-out.

• 1 Contrast. White and red words on a deep blue ground: nothing is gray.
• 2 A promise kept. The verse is shown in the frame, so the promise is specific and the video has to deliver it.
• 3 Words, big and left. The largest word is the question. The words sit on the left, away from the face.
• 4 One red arrow. It shows exactly where to look.
• 5 The white outline. A thick white edge lifts him off the background, like a sticker.
• 6 A face you can read. Eyes to the camera, one clear expression, placed on the right.

Pillar 5 · the sticker cut-out

Thumbnails that work

Eight thumbnails from David's channel, and what each one does in a glance.

Pillar 5 · the thumbnail studio

Two tools, one thumbnail

Make the picture in one tool. Add the words in another. Each does what it does best.

Higgsfield · Soul Cinema

Make the picture

Soul Cinema makes cinematic, photographic people and places. Ask for 16:9, one clear subject, and one side of the frame left empty for the words.

ChatGPT · GPT Image 2

Add the words

Upload the picture and ask GPT Image 2 to add them. It is strong at lettering and at editing a picture.

Each tool does what it does best.

The photo stays clean, with no lettering baked into it.

One photo, many thumbnails: reuse it with new words.

The three walkthroughs, picture to thumbnail

1 · The question thumbnail
(walkthrough 1)

2 · The sticker cut-out
(walkthrough 2)

3 · The close-up
(walkthrough 3)

Use your own face, or the face of someone who agreed, and never imitate a real person. Check the spelling of every word before you publish.

Thumbnail studio · walkthrough 1 of 3 · step 1

The question thumbnail

One face looking straight at you, one honest question, and a dark empty side where the words can breathe.

Higgsfield · Soul Cinema · 16:9 · Before

Paste this into Higgsfield, Soul Cinema, 16:9:

Candid documentary still, 16:9. A woman in her early thirties with dark curly hair sits at a small cafe table beside a rain-streaked window at dusk, looking straight into the camera with a quiet, searching expression and one eyebrow slightly raised. An open paperback Bible and a half-finished coffee on the table. She sits in the right third of the frame; the left half of the frame is soft, dark, out-of-focus cafe interior with warm bokeh lights, kept empty and uncluttered. Shot on an 85mm lens, shallow depth of field, soft key light from the window on her face, natural skin texture, slight film grain, muted teal and warm amber grade. No text anywhere in the image.

Thumbnail studio · walkthrough 1 of 3 · step 2

The same photograph, with the words added and nothing else changed.

Before

ChatGPT · GPT Image 2 · After

Upload the picture, then paste:

Edit this photograph into a YouTube thumbnail. Keep the photograph exactly as it is: the same woman, the same face, lighting and framing. Add only text, on the empty dark left side, in two stacked lines: "WHAT IF" and "IT'S TRUE?". Use a tall, heavy, condensed sans-serif (like Impact or Anton), pure white, tightly spaced, with a thin soft black shadow so it separates from the background. Make the word "TRUE?" bright yellow (#FFD400). The text block fills about 40 percent of the image width and stays readable when the thumbnail is shown 120 pixels wide. Keep a 5 percent margin from every edge. Do not add anything else: no logos, no extra words, no borders, no arrows.

Thumbnail studio · walkthrough 2 of 3 · step 1

The sticker cut-out

David's signature starts with an ordinary portrait: a clear face on a plain backdrop, with room on the left.

Higgsfield · Soul Cinema · 16:9 · Before

Paste this into Higgsfield, Soul Cinema, 16:9:

Studio portrait photograph for a magazine, 16:9 frame. A man in his forties with a short salt-and-pepper beard, no glasses, wearing a dark green henley, looks directly into the camera with an open, curious expression and eyebrows slightly raised, a hint of a smile. Head and shoulders, positioned in the right half of the frame, facing the camera. Plain light-grey seamless paper backdrop, bright even beauty-dish lighting with soft shadows, 85mm lens, crisp focus on the eyes, natural skin texture. The left half of the frame is empty backdrop. No text.

Thumbnail studio · walkthrough 2 of 3 · step 2

A face cut out with a thick white outline, a bold ground, the words left and one red arrow.

Before

ChatGPT · GPT Image 2 · After

Upload the picture, then paste:

Turn this portrait into a YouTube thumbnail in the "sticker cut-out" style. Cut the man out of the photo and give him a thick white outline (about 2.5 percent of the image height) with a soft drop shadow, like a sticker. Keep his face, hair, beard and clothes exactly as in the photo, and keep him large on the right side. Replace the background with a flat, saturated gradient from royal blue (#1B3FD8) at the top left to deep violet (#5B21B6) at the bottom right. On the left, add the words "HE CHANGED" and "HIS MIND" in two stacked lines in a heavy condensed sans-serif, white fill with a thick dark-navy outline and a subtle shadow, with "MIND" in bright yellow (#FFD400). Add one flat red arrow (#E11D2E) between the words and the man, pointing at his face. Nothing else: no logos, no extra words.

Thumbnail studio · walkthrough 3 of 3 · step 1

The close-up

One detail, close up, with a calm bench on the left where the words will go.

Higgsfield · Soul Cinema · 16:9 · Before

Paste this into Higgsfield, Soul Cinema, 16:9:

Macro documentary photograph, 16:9. A fragment of ancient papyrus with faded brown ink lines lies on a white archival table; a conservator's white-gloved hand holds a small brass magnifying glass over one corner of it. Soft north-window light, extremely shallow depth of field, warm neutral tones, realistic fibres and age, the ink marks abstract and unreadable. The papyrus and magnifier sit in the right two-thirds of the frame; the left third is a calm, out-of-focus bench. Shot on a 100mm macro lens, slight film grain. No text.

Thumbnail studio · walkthrough 3 of 3 · step 2

The same close-up, with the words, a red circle and one arrow added.

Before

ChatGPT · GPT Image 2 · After

Upload the picture, then paste:

Edit this photograph into a YouTube thumbnail. Keep the photograph exactly as it is. Add: (1) the words "LOOK" and "CLOSER" stacked on the left side in a heavy condensed sans-serif, white with a thin dark shadow, with "CLOSER" in bright yellow (#FFD400); (2) a hand-drawn red circle (#E11D2E, thick, slightly imperfect) around the brass magnifying glass and the ink marks beneath it; (3) one short flat red arrow pointing from the words toward the circle. Keep a 5 percent margin from the edges. No other text, logos or graphics.

Pillar 5 · the thumbnail studio

Your own prompt

Two templates, one for each step. Fill in the brackets and keep everything else.

[STYLE: Candid documentary still / Studio portrait photograph / Macro documentary photograph], 16:9. [WHO OR WHAT, with one clear emotion or action]. [WHERE, and the time of day]. [WHERE THE SUBJECT SITS IN THE FRAME]; the [LEFT / RIGHT] [HALF / THIRD] of the frame is [soft, dark and empty] so words can go there. Shot on a [85mm / 50mm / 35mm] lens, shallow depth of field, [LIGHT], natural skin texture, slight film grain, [COLOUR GRADE]. No text anywhere in the image.

Edit this photograph into a YouTube thumbnail. Keep the photograph exactly as it is. Add only text, [WHERE], in [one / two] stacked line[s]: "[WORDS]". Use a tall, heavy, condensed sans-serif, pure white, with a thin soft black shadow so it separates from the background. Make the word "[KEY WORD]" bright yellow (#FFD400). The text block fills about 40 percent of the image width and stays readable when the thumbnail is shown 120 pixels wide. Keep a 5 percent margin from every edge. Full-bleed image, no border or frame. Do not add anything else: no logos, no extra words.

Filled in for the first example (walkthrough 1)

Pillar 5 · the thumbnail studio · more to try, 1 of 3

Six more to try

Two choices for a comparison: two paths, two answers

Documentary photograph, 16:9. A hiking trail forks in two directions through misty pine forest at first light. The left path is dark and overgrown; the right path is lit by a shaft of low golden sun. A lone hiker with a backpack, seen from behind, stands small at the fork. Soft fog, 35mm lens, natural colours, subtle film grain. Keep the upper third of the frame calm. No text.

Behind the scenes at an event for an outreach day or a conference table

Candid documentary photograph, 16:9. Two volunteers in casual jackets stand at a plain folding table at a street outreach event in a city plaza, laughing with a young man who stopped to talk; on the table, stacks of paperback Bibles and paper cups of coffee. Late-afternoon golden-hour light, shallow depth of field, 50mm lens, natural colours, subtle grain. The left third of the frame is calm, out-of-focus plaza. No banners, no signs, no writing anywhere in the image.

Pillar 5 · the thumbnail studio · more to try, 2 of 3

The world is writing a map or globe, for people far away

Photograph, 16:9. A vintage brass-and-wood world globe on a dark wooden desk at night beside a smartphone whose screen glows with a message thread (blurred, unreadable). A single warm desk lamp lights the scene from the left; the background is deep and dark. Shallow depth of field, gentle film grain, rich but natural colour. The right half of the frame is calm dark space. No readable text.

The message at night a story thumbnail, for one person's question

Cinematic close-up still, 16:9. A young woman in her twenties in a dark bedroom at night, her face lit softly by the cool glow of a phone she holds near her chin, reading a message with wide, uncertain, hopeful eyes. Her face and hands fill the right half of the frame, looking slightly off-camera toward the phone. A warm lamp glows far behind her, out of focus. Shot on an 85mm lens, shallow depth of field, natural skin texture, subtle film grain, deep navy shadows with warm amber accents. The left half of the frame is dark and empty. The phone screen is blurred and unreadable. No text.

Pillar 5 · the thumbnail studio · more to try, 3 of 3

The reaction the sticker cut-out, with a portrait of your own

Bright editorial portrait photograph, 16:9. A young woman in her twenties with a short bob haircut and a yellow cardigan reacts with delighted surprise, mouth open, one hand raised beside her cheek, looking directly into the camera. Head and shoulders in the right half of the frame against a clean pale warm-white studio backdrop, bright even daylight lighting, 85mm lens, natural skin texture, crisp focus. The left half is empty backdrop. No text.

A story that starts at a door for the story of a first visit

Cinematic documentary still, 16:9. A person in a long coat, seen from behind, stands at the open wooden doors of a small stone church at dusk; warm light spills out of the doorway across wet cobblestones, cool blue evening sky above. Wide shot, 35mm lens, shallow haze, natural colours, subtle film grain. The person is small in the right third of the frame; the left two-thirds is calm stone wall and street. No readable text.

Pillar 5 · the thumbnail studio

Background and words only

No photo at all: a color, the words, and an empty space for your own cut-out.

A YouTube thumbnail background in 16:9 with no people and no photographs. A flat, saturated gradient from deep teal (#0B7285) on the left to royal blue (#1B3FD8) on the right, with a very subtle paper-grain texture. On the left half, add the words "ASK ME" and "ANYTHING" in two stacked lines, in a heavy condensed sans-serif, white fill with a thick dark-navy outline and a soft shadow, with "ANYTHING" in bright yellow (#FFD400). Add one flat red arrow (#E11D2E) pointing to the right, toward the empty space. Leave the right 45 percent of the image completely empty so a cut-out photo can be placed there later. Nothing else: no logos, no extra words, no people.

Pillar 5 · the thumbnail studio

Make it better in ChatGPT

Treat the first result as a draft. Ask for the next one in plain words.

• 1 Ask for three variations of the same thumbnail, then keep the clearest.
• 2 Make the words bigger, or use fewer of them.
• 3 Ask for a brighter background when the picture feels dull.
• 4 Upload your own photo as the reference, and say what to keep exactly as it is.
• 5 Check the spelling of every word on the picture.

Check it small

These four are shown at 120 pixels wide. Shrink every thumbnail this far before you choose. If you can still read the words and see the one idea, it works.

120 px

120 px

120 px

120 px

Worksheet

Your thumbnail

Draw or paste your thumbnail in the large frame, and again in the small one.

A face you can read

Three words or fewer

One idea

Contrast

A promise kept

Readable small

Pillar 6 of 7

Call to action

You worked to earn the viewer. Do not stop before you ask for the next step.

Pillar 7 of 7

Optimize your content

Go back. Your old videos can still be found.

From Echo

After the view

A video is a doorway. Someone who walks through it will often write back.

1. A comment arrives
   YouTube comments are a channel Echo answers on. Your responders read and answer the comments under your videos from inside Echo [E2]. They are comments, not private messages, and a comment does not expire, so there is no reply window running out.
2. It becomes a conversation
   A comment asking a real question stops being something one responder notices by accident [E2]. It sits in the shared inbox with the seeker's other messages, and a responder claims it, so two people never answer the same seeker at once.
3. A person answers
   If a seeker moves to WhatsApp, Messenger or text, the conversation lands in the same inbox. Other channels have clocks: a WhatsApp conversation closes after 24 hours without the seeker writing, and Messenger and Instagram after 7 days [E2].

Do this week · from Echo

Find the oldest unanswered comment under your most-watched video, and answer it as a person.

From Echo

What Echo reads afterwards

Views are a stand-in. What happened in the conversation is the thing itself.

A person answers every seeker. Afterwards, EchoAnalyst reads the closed conversation [E1]. It does not talk with seekers and it does not write a responder's replies.

A score on eleven skills, each 1 to 5, with a one-sentence coaching note: among them listening, clarity, depth, closing and follow-up, and grace-filled truth [E1, E3].

Seven kinds of seeker barrier, and whether the responder engaged each one. The Seeker Barriers report then shows, with your own numbers, how often seekers move toward God when their real barrier is addressed [E1, E3].

Fourteen reports, from one responder's review to the Ministry Health Report a board reads [E1]. EchoAnalyst is a paid add-on, and a ministry that has not signed up gets ten free analyses a month to try it [E1].

For a video ministry the useful part is plain: the barriers that show up in conversations are the ones your next video can speak to.

From Echo

Before you publish

• Plan Who it is for, the one question it answers, three places you will share it.
• Title The words a seeker would type are in it, near the front. It promises only what the video gives.
• Description The first two lines carry the point. One link to a place where a person will answer. Chapters if it is long: the first at 00:00, at least three [7]
• Hashtags and tags Up to three hashtags will show above the title [10]. Tags only for words people misspell [2]
• Captions Watched once on mute. Every name and verse corrected [3]
• Thumbnail Custom, 16:9, passes the six checks on the worksheet, tested if you can [4, 5]
• End screen One clear next step, said aloud and shown on screen [8]
• After you publish Someone is ready to answer comments in Echo, and the first comment gets a person within the day.

Evidence

Sources

• [E0] Apologetics Roadshow, Thumbnails and channel. youtube.com/@apologeticsroadshow · the thumbnails shown in this guide
• [1] YouTube Help, Upload YouTube videos. support.google.com/youtube/answer/57407 · titles up to 100 characters, descriptions up to 5,000
• [2] YouTube Help, Add tags to videos. support.google.com/youtube/answer/146402 · tags play a minimal role; tag stuffing is spam
• [3] YouTube Help, Use automatic captioning. support.google.com/youtube/answer/6373554 · automatic captions can be wrong; review in Studio
• [4] YouTube Help, Add custom thumbnails on YouTube. support.google.com/youtube/answer/72431 · verified account, 16:9, 640 px minimum width, JPG or PNG
• [5] YouTube Help, Test & compare thumbnails. support.google.com/youtube/answer/13861714 · up to three thumbnails, test starts at publish
• [6] YouTube Help, A/B test titles & thumbnails. support.google.com/youtube/answer/16391400 · up to three; highest watch time shown to all
• [7] YouTube Help, Video Chapters. support.google.com/youtube/answer/9884579 · first timestamp 00:00, at least three in order
• [8] YouTube Help, Add end screens to videos. support.google.com/youtube/answer/6388789 · 5 to 20 seconds, up to four elements, video at least 25 seconds
• [9] YouTube Help, How YouTube recommendations work. support.google.com/youtube/answer/16089387 · satisfaction surveys, likes, dislikes
• [10] YouTube Help, Hashtags article. support.google.com/youtube/answer/6390658 · three shown above the title; over 60 ignored
• [11] YouTube Help, Understand three-minute YouTube Shorts. support.google.com/youtube/answer/15424877 · square or vertical, up to three minutes
• [E1] EchoGlobal, EchoAnalyst. echoglobal.org/product/echoanalyst · eleven skills, barriers, fourteen reports, what it does not do
• [E2] EchoGlobal, Integrations and channels. echoglobal.org/product/integrations · comments channel, reply windows
• [E3] EchoGlobal, EchoAnalyst white paper. echoglobal.org/resources/echoanalyst-white-paper · criteria, seven barrier categories

Credits

About this guide

Based on a conversation between David Wood and Andre of Echo. One of Echo's free guides.

Find David

David's channel, Apologetics Roadshow
youtube.com/@apologeticsroadshow

Thumbnails

Thumbnails from Apologetics Roadshow.

Example pictures

Example pictures made with Higgsfield Soul Cinema and GPT Image 2.

About Echo

Echo is the conversation platform for Christian ministries. A person answers every seeker, on the app they chose, and the channels come into one shared inbox.