How-To Guide
How should a founder choose between a tutorial post and a product promotion?
A per-post rule for founders: decide by slot and account state, hold a starting ratio for 30 days, promote beside proof, and read the counterexample.
By the Superdirector editorial team. Written November 14, 2026. Every ratio and survey figure below is attributed to the page that published it; none is our measurement, and the ratios are starting lines to test, not results to expect.
At a glance
How to choose between a tutorial post and a product promotion
Decide by the slot the next post has to serve, then by the account's state. Teach with a tutorial post when the job is helping your buyer do one thing today, the audience is still deciding whether you are worth following, and the metric you need is saves and shares. Promote when the ask is earned, proof sits beside the offer, and the metric you need is clicks and conversations. The published ratios disagree with each other on purpose: they are starting lines, so hold one for about 30 days, audit the feed, and let your own numbers move it.
- Slot first: teach for the audience's job, promote for the account's earned ask.
- Hold one starting ratio (80/20, 50/25/25 or 40/30/20/10) for 30 days before judging it.
- Pair every promotional ask with proof, and never stack two promotions back to back.
- Test the counterexample: if you ship steadily, your product updates may be your best content.
Three published starting lines, each held about 30 days before you audit. Original diagram, drawn from the splits TanStack Ship, BioGPT and Elscedres publish.
Make it work for your brand
Find it. Remake it. Post it. Superdirector turns the viral formats in your niche into your brand's carousels, X posts and video scripts, and keeps the calendar full.
How should a founder choose between a tutorial post and a product promotion?
Treat the choice as a per-slot decision, not a verdict on what your account is allowed to be. A tutorial post teaches something the buyer can act on today and earns saves, shares and the trust that makes a later ask work. A product promotion asks for something and spends the attention the other posts have banked. Both belong on one account, in a proportion the account earns, and the fastest way to hold that proportion is to name the job of the next slot before writing anything.
The sources that answer this question agree on the shape and disagree on the numbers. TanStack Ship's founder guide sets an 80/20 rule with value content named as tutorials, insights, tips and threads, and promotional content named as product updates, launches and offers. BioGPT's balance guide opens with a different starting line: 50 percent educational, 25 percent engagement, 25 percent promotional, held for 30 days. Elscedres' mix formula splits four pillars at 40 percent educate, 30 percent entertain, 20 percent inspire, 10 percent promote. The disagreement is the finding: no page holds a law, so the work is picking a starting line, reading your own counts and adjusting.
Pick by the slot, not by identity
The clearest choose-between pages do not argue over the taxonomy; they hand the decision to the job the next post must do. Userorbit, comparing three demo formats, puts a quick-decision table at the top and closes with the rule that the format is a distribution decision, not a permanent product decision. The same frame decides posting types: the account is not a tutorial account or a promotional account, it is an account whose next slot has a job.
| If the job is... | Post the... | Why |
|---|---|---|
| Teaching one narrow task the buyer could repeat today | Tutorial post | Saves and shares are the proof; the page earns later asks |
| Announcing something new that changes their work | Product promotion | The ask is specific and answers a change the reader cares about |
| A launch week | Promotion-heavy stretch, planned | BioGPT's guide flips to roughly half promotional, then returns to base |
| Rebuilding attention after a quiet period | Tutorial post | Value posts earn the audience back before any ask |
| Unsure which side dominates your feed | Neither: audit first | Tag the last 30 posts and count; see the audit below |
One more frame from the same page transfers directly: a demo format chosen for launch can be reused elsewhere when each version has a clear job. A launch-week promotion, cut into a teaching angle afterwards, stops being a spent slot and becomes the tutorial slot's material. The decision is per post and per week, never a permanent identity.
Start from a ratio, then let your numbers move it
BioGPT's short answer is the most operational of the three: start at 50 percent educational, 25 percent engagement, 25 percent promotional, hold it for 30 days, track saves, comments and link clicks, then shift about 10 points toward whichever type your audience rewards. The page also collects the other published lines in one table, where the 80/20 rule is the safe default for small brands, the rule of thirds fits personal brands, the 4-1-1 rule (attributed to marketer Andrew Davis) suits curation-heavy B2B feeds, and the 5-3-2 rule helps brand-new accounts that cannot yet produce five original posts a week.
| Starting line | Splits it as | Published by | Fits |
|---|---|---|---|
| 80/20 | 80% value, 20% promotional | TanStack Ship | The safe default most of the sources accept |
| 50/25/25 | Educational, engagement, promotional | BioGPT | Feeds that need replies and reach, not only saves |
| 40/30/20/10 | Educate, entertain, inspire, promote | Elscedres | Accounts whose promotional posts must stay rare |
The 30-day rule exists because the mix reads differently each week. Before changing anything, run the audit BioGPT prescribes: pull the last 30 posts, tag each as educational, promotional or engagement honestly (a tip ending in buy now is a promotional post wearing a hat), turn the counts into percentages, list the top five by saves, comments and clicks, and shift 10 to 15 points toward the target. BioGPT's author's own first audit came back 61 percent promotional, which is exactly why the audit replaces the feeling.
What each post type has to earn
BioGPT's per-type table is the piece worth copying into your own planning note, because it names the metric that proves each type worked and the formats that carry it. A tutorial that nobody saves was a lecture, not a lesson; a promotion with no proof beside it reads as noise.
| Post type | Its job | Formats that work | Metric to watch |
|---|---|---|---|
| Tutorial (education) | Teach one actable thing | Step-by-step carousels, short tutorials, teardowns, myth-busting | Saves and shares first |
| Product promotion | Ask, with the ask earned | Launch announcements, product demos, customer results, offers | Link clicks and direct conversations |
| Engagement | Start a conversation the platform can see | Questions, polls, defensible takes, behind-the-scenes | Comments relative to reach |
The pairing rule is the same across the sources: every ask sits next to proof. Elscedres makes the contrast argument that when most of the feed serves the reader, a clear promotional post stands out as earned rather than pushy. TanStack Ship states the same pairing from the other side: value content builds trust and authority, and promotional content converts that trust into action. Neither page treats the promotion as a separate account discipline; both treat it as the payoff slot of a teach-first feed.
When your product is the better subject
Every ratio above would fail one honest counterexample, so carry it before you commit a calendar to it. SaaS Club's playbook on Adam Fard, founder of UX Pilot, records the opposite result from the value-first orthodoxy: his educational UX content worked fine for the agency, but on the new product, feature announcements and product updates outperformed education, drawing more replies, shares and engagement than the carefully crafted teaching posts. The principle the playbook draws: educational content positions you as a teacher, product updates position you as a builder, and founders connect more with builders.
The playbook states its own conditions, which is what makes it evidence rather than a dare: this works when you are actively shipping real improvements and talking to an audience of potential users, and it has a skip condition, because talking about the same features every week gets stale fast. His cadence was three to four LinkedIn posts a week, with updates shared as stories about the problem they solve, not as sales pitches. For a founder deciding between the two post types, the practical reading is narrow: the closer you sit to shipping something users are waiting on, the larger your promotional share can fairly be, and the changelog itself is your best content strategy.
What the counterexample does not establish is a number. It is one founder's observed engagement pattern, reported in a podcast playbook, with no audience-size denominator and no claim that the same flip transfers to every account. Use it to justify a bigger promotional share, not to abandon the ratio and audit.
What the platform actually ranks
The platforms publish one piece of this picture themselves. Meta's Instagram Feed ranking card says the Feed ranking system predicts how likely you are to comment on a post, like and save it, or tap on a profile photo, and that the more likely you are to take an action, and the more heavily it weighs that action, the higher the post ranks. That is the official grounding for the metric table above: saves and shares are ranked actions, which is why a tutorial's saves are more than vanity, and why a promotion that earns none of the ranked signals has to win on its own click instead.
Two more official lines set boundaries. The collection step gathers potential posts excluding advertisements, and a variety rule limits the feed to no more than three posts in a row from the same account, which is the platform-side reason the placement rules below exist. What the card does not publish is almost everything else: the exact weighting between saves, likes and shares, and any threshold where a promotion loses reach. Weighting multipliers quoted in coverage of platform remarks are not in the card, so this page states no multiplier.
Sequence rules: never back to back, ramp before a launch
Placement is where most founders fail the mix rather than the ratio. BioGPT's held calendar placed the two promotional slots after value days and the author states the rule plainly: never schedule two promotional posts back to back, because the audience needs a reason to be there before the ask. The Sunday slot in that calendar carried the soft version: product shown in real use, no hard ask.
Launch weeks change the shape only if the ramp is real. Elscedres warns the mistake that kills the formula: three weeks of solid mixed content followed by two weeks of product posts, where the spike arrives without a trust base and tanks reach at the moment you need it. The prescribed fix runs the ramp backwards: if a launch is coming in four weeks, increase the educational and inspiring content now so the audience is warm when the promotion lands. BioGPT's launch rule then takes over: roughly half promotional during the launch week, then straight back to the base mix.
One legal boundary sits on the promotion side. If a promotional post involves an endorsement by someone with a material connection people would not expect, the FTC's Endorsement Guides say the connection should be disclosed clearly and conspicuously, and an endorsement must reflect the endorser's honest opinion. On your own account promoting your own product there is no endorser to disclose; the guides matter the moment a creator, partner or employee's voice carries the promotion.
Teaching ramps for four weeks, launch week runs about half promotional, then the feed returns to base. Original diagram, drawn from the ramp warning and launch rule cited in this section.
Hybrid formats: teach and sell in one post
Sometimes the slot has both jobs. ContentsPilot's hybrid-format catalogue opens with the exact hybrid for this decision: a tutorial carousel whose first slides teach the step-by-step and whose last slide brings a free template or an invitation to test the product. The teaching slides protect the saves; the last slide picks up the clicks.
The same guide sets the guardrails that keep a hybrid from becoming an advertisement: define the hybrid as content that solves a real pain, shows your method or product in action, and offers a soft next step; choose one objective per post (download, message, test, waitlist); and cap CTAs at one main and at most one secondary engagement ask. Its measurement rule closes the loop: track saves, CTA clicks and conversations, and compare the hybrid against a pure tutorial post so the lift is your own number rather than the catalog's promise.
Five teaching slides earn the saves; one closing slide carries a single ask. Original diagram, drawn from the hybrid format described in this section.
Where Superdirector fits
Superdirector, which publishes this guide, carries the calendar side of this decision. The weekly slots, the starting ratio, the audit list and the ramp plan live in the running calendar, and the drafts for both post types open with the decision already made. Growth at $199 a month is one AI-run operated account: scheduled research, routines, and, once posting is turned on for that account, publishing of eligible Instagram carousels and single link-free X text posts under its standing posting permission, subject to account readiness and plan entitlement. Starter at $29 a month is the hands-on entry: MCP access from Claude or ChatGPT for people driving the work themselves. See what runs today.
The decision itself stays yours, and the sources above keep it that way: the ratio that works is the one your own audit reads, and the counterexample means the ratio can bend toward promotion when the account has a story worth telling. Start at a line, hold it 30 days, count honestly, and let both post types keep their place.