In this guide
A completed publishing queue is an operational result. Audience growth and enquiries require another loop: record comparable results, find the weak step and test a specific change. Taisly keeps distribution organised while you evaluate those outcomes in platform and website analytics.
Write a testable content question
Start with a question such as “Will a demonstration bring more relevant enquiries than a list of tips?” Keep the audience and offer consistent. Create several variations over a manageable period rather than declaring a winner from one viral post.
Use your weekly publishing plan to record the format, opening and next action before release. A comparison across platforms is not a controlled A/B test: audiences, distribution and metric definitions differ.
Record three different kinds of evidence
| Layer | Where to check | What to record |
|---|---|---|
| Delivery | Taisly History and the actual post | Account, time, status and post URL |
| Audience response | Each platform's analytics | Views, watch metrics where available, saves and profile actions |
| Business outcome | Website analytics and your enquiry log | Tagged visits, suitable enquiries or purchases |
Take readings after the same elapsed time, for example 72 hours, and compare within the same platform. Note missing data instead of inventing a value. View counts alone do not identify customers.
Work through the funnel, not just the largest number
These are invented figures to demonstrate the calculation, not Taisly customer results.
| Video | Views | Profile visits | Website visits | Enquiries |
|---|---|---|---|---|
| Tips list | 10,000 | 80 | 20 | 1 |
| Exercise demonstration | 4,000 | 120 | 48 | 6 |
The tips list has 0.8% profile visits per view: 80 ÷ 10,000 × 100. The demonstration has 3%. Website-to-enquiry rates are 5% and 12.5% respectively. The smaller-view video generated more enquiries in this example, but a small sample does not prove causation or a lasting advantage.
Give each video a relevant next step
A useful caption connects the lesson to the offer: “Want feedback on your squat? Trial-session details in the profile.” The profile should explain who the session is for; the linked page should offer that session directly.
Where a platform supports a clickable profile link, tag it by platform, for example ?utm_source=instagram&utm_medium=organic_social&utm_campaign=squat_series. A shared profile link cannot reliably attribute every visit to an individual video. Use per-video attribution only when the placement and tracking actually support it.
Choose one change and run another batch
| Observed weak step | Next test |
|---|---|
| Viewers leave before the demonstration | Show the useful moment earlier; shorten the introduction |
| Views but few profile visits | Make the relevance and next action clearer |
| Profile visits but few website visits | Align the bio, offer and destination |
| Website visits but few suitable enquiries | Review the landing page, form and audience fit |
These are hypotheses, not algorithm rules. Use native watch data where available and repeat the comparison. Keep the file, accounts and offer stable enough to learn from a changed opening.
Schedule the next approved variations in Taisly. Automate distribution after the workflow is reliable, while keeping content decisions tied to evidence rather than posting volume.
Your checklist
0/3 completePut it into practice
Upload & publish
Choose a finished video and the connected accounts that should receive this publication.