Repurposing plan: "How we cut customer onboarding from 14 days to 5"
Assumptions to confirm: you have the rights to reuse Dana Whitfield's article. You didn't give a link, so drafts use [ARTICLE LINK]. I assumed the posts go out as Brightloop or Dana, in first-person "we". Dates run Thu 8 Oct to Wed 14 Oct 2026.
1. Source summary and extraction
Thesis: Brightloop's onboarding took 14 days mostly because of waiting, not the software. Removing the waits cut it to 5 days.
Key points
- The average was 14 days to get a first team scheduled, timed across 112 new accounts.
- The delay was mostly waiting on three things: the customer's admin finding an export file, IT approving a calendar connection, and Brightloop replying to questions.
- Change 1: a 40-page setup guide became a 6-step in-product checklist.
- Change 2: the calendar connection request starts on day zero, before the kickoff call, so IT approval runs in parallel.
- Change 3: "reply within one business day" became a shared channel with same-day answers.
- After six months, the average fell from 14 days to 5, and first-month support tickets fell by about a third.
- The surprise: the checklist saved only about two days. Most of the gain came from parallel IT approval and same-day replies.
- Advice: time where the waiting happens first.
Exact quote: "The fastest way to onboard a customer is to stop making them wait on us," (Marcus Lee, CEO, in the retro).
What the source does not say (so the posts don't claim it):
- It doesn't report revenue impact.
- It doesn't say faster onboarding caused the drop in churn. Churn fell that quarter "for other reasons" too.
- It gives no per-change breakdown beyond the checklist's "about two days".
- It doesn't say the 112 accounts were the same group measured after the changes.
2. Format plan
| Platform | Formats used | Why |
|---|---|---|
| Text post (thesis), text post (recap with caveat) | Professional audience, longer text works | |
| X | Short thread, single post with one number | Thread for the steps, one punchy finding |
| Carousel, quote card | Visual, with one idea per slide |
Confirm current character limits, carousel slide counts and link handling before posting. Instagram captions don't make links clickable, so use link-in-bio or a link sticker.
3. Drafts with source map
Post 1: LinkedIn text, Thu 8 Oct
Last year, a new Brightloop customer needed 14 days on average to get their first team scheduled.
We timed it across 112 new accounts. Most of the delay wasn't the software. It was waiting: for the customer's admin to find an export file, for IT to approve a calendar connection, for us to reply to questions.
We changed three things and got the average down to 5 days. I wrote up what they were, and what surprised us.
Full article: [ARTICLE LINK]
Post 2: X thread, Fri 9 Oct
1/ Our onboarding took 14 days on average. We cut it to 5. Here are the three changes we made ?
2/ Change 1: we replaced the 40-page setup guide with a 6-step checklist inside the product.
3/ Change 2: we started the calendar connection request on day zero, before the kickoff call, so IT approval ran in parallel.
4/ Change 3: we moved from "reply within one business day" to a shared channel with the customer. Questions got answered the same day.
5/ Result after six months: 14 days → 5. First-month support tickets fell by about a third. Which change mattered most surprised us. Full write-up: [ARTICLE LINK]
Post 3: Instagram carousel, Sun 11 Oct
- Slide 1: "Where did our 14 days of onboarding go?"
- Slide 2: "We timed 112 new accounts. Average: 14 days to first team scheduled."
- Slide 3: "Wait #1: the customer's admin finding an export file."
- Slide 4: "Wait #2: IT approving a calendar connection."
- Slide 5: "Wait #3: us replying to questions."
- Slide 6: "Most of the delay wasn't the software. It was waiting."
- Slide 7: "How we got to 5 days: full article, link in bio."
- Caption: "We timed our onboarding and found the delay wasn't where we expected. The full story is in the article (link in bio)."
Post 4: X single post, Mon 12 Oct
We assumed the new in-product checklist would be our big onboarding win.
It saved only about two days.
Most of our drop from 14 days to 5 came from starting IT approval in parallel and answering questions the same day.
[ARTICLE LINK]
Post 5: Instagram quote card, Tue 13 Oct
- Card text: "The fastest way to onboard a customer is to stop making them wait on us." Marcus Lee, CEO, Brightloop
- Caption: "That was Marcus's summary in our onboarding retro. We went from 14 days to 5. The article covers what we changed (link in bio)."
Post 6: LinkedIn recap, Wed 14 Oct
If you want to shorten customer onboarding, start by timing where the waiting happens.
For us, that meant 112 accounts and an average of 14 days. The product wasn't the bottleneck. Waiting on customer admins, IT approvals and our own replies was.
Three changes later, the average is 5 days, and first-month support tickets are down by about a third.
A caveat: we didn't measure revenue impact. Churn also fell that quarter, but for other reasons too, so we can't say faster onboarding caused it.
The full breakdown, including which change mattered most: [ARTICLE LINK]
Source map
| # | Platform / format | Source location | Notes |
|---|---|---|---|
| 1 | LinkedIn text | Para 1 (14 days, 112 accounts, three waits), para 2 (14→5) | Doesn't name the three fixes, to leave the thread something new |
| 2 | X thread | Para 2 (three changes), para 3 (results) | "Surprised us" teases para 4 without giving it away |
| 3 | Instagram carousel | Para 1 (the three waits), para 3 (5 days) | Slide 7 points to the article for the fixes |
| 4 | X single | Para 4 (checklist about two days; parallel IT and same-day replies) | "We assumed" is a framing inference, see check below |
| 5 | Instagram quote card | Para 4 (Marcus Lee quote) | Quote is word for word |
| 6 | LinkedIn recap | Para 5 (advice), para 3 (results and caveat) | Caveat kept in full |
4. Schedule
| Day | Platform | Post |
|---|---|---|
| Thu 8 Oct | 1: the finding | |
| Fri 9 Oct | X | 2: three changes thread |
| Sun 11 Oct | 3: carousel on the waits | |
| Mon 12 Oct | X | 4: the checklist surprise |
| Tue 13 Oct | 5: quote card | |
| Wed 14 Oct | 6: recap with caveat |
Every post links to the article. Put the link in the first reply or comment if you'd rather keep it out of the main text. Keep it in your Instagram bio all week. No two platforms post the same text on the same day.
5. Fidelity check
- Numbers and names: 14 days, 5 days, 112 accounts, 40-page guide, 6-step checklist, "about a third", "about two days", six months, Marcus Lee (CEO). All match the source.
- Quote: Post 5 matches word for word, with the speaker credited. The "as our CEO Marcus Lee put it in the retro" lead-in is paraphrased in the caption.
- Caveat: Post 6 keeps the no-revenue and no-causation points. Posts 1–5 make no churn or revenue claims.
- Fixed: I removed "support tickets fell by a third" as an unqualified claim. Posts 2 and 6 now say "first-month" and "about".
- Needs your confirmation:
- Post 4's "We assumed the checklist would be our big win" isn't in the source. The article only says the checklist saved about two days and the surprise was where most of the gain came from. Either approve the line with Dana or change it to "The checklist alone saved only about two days."
- Post 3 says "most of the delay wasn't the software". The source says "Most of the delay was not the software." That's the same meaning, but check you're happy with the shorter phrasing.
- The article link, the account voice, and rights to reuse.
The article is fictional and was written by the reviewer. I compared all six drafts with it: the numbers, names and the CEO quote are exact. Post 4 says We assumed the checklist would be our big win, which is not in the article; the model caught this itself in its fidelity check and asked for confirmation before use.
What it does
Gets many good posts out of one strong piece and keeps them true to it. It reads the whole source first, then lists the thesis, five to eight key points, exact quotes, numbers and stories with where they appear, the objections it answers, and what the source does not say. It maps those to formats that suit each platform, drafts the posts with a source map that shows which part each one comes from, spaces them over the period, and runs a fidelity check: no claims the source does not make, numbers and names exactly as written, quotes word for word and credited, nothing taken out of context.
Output
A source summary, a format plan per platform, drafts with the source map, a schedule and the fidelity check with anything you must confirm.
Good for
Writers, marketers and teams who want to get more out of one article, newsletter, podcast or talk without misquoting it.
Low risk: pure instructions with no scripts, no network access and no file writes. It writes drafts and a plan and does not publish or connect to any account. Only repurpose content you own or have permission to reuse, and quote others exactly with credit. Test-run once with a fictional article; I compared all six drafts with it, and one line the article does not contain was flagged by the model itself and needs your confirmation. Nothing was posted. Original by AIBars (MIT).