Resource · Frameworks
Cold DM Message Frameworks
A message framework is a skeleton you drop your audience and offer into, so you are not staring at a blank box wondering how to start. This guide covers the structures that work in DMs, AIDA, PAS, BAB, and Hook-Offer-Proof, each with a worked example and guidance on when to use it. Frameworks speed up writing and keep messages consistent at volume, which is exactly when consistency protects reply rate. Pick one, fill it, and test it against a plain personalized example.
Hook-Offer-Proof (default)
The simplest and most reliable structure for short DMs. Open on a true observation about the recipient, state the outcome in one line, then add one proof point. It respects the reader's time and leads with relevance, which is what earns the reply.
Hook-Offer-Proof
Best for: Best for most cold audiences and short DM windows.
AIDA
Attention, Interest, Desire, Action. Use AIDA when you have room for a slightly longer message, such as LinkedIn. It builds from a grab to a reason to want the outcome to a clear next step.
| Step | What to write | Example |
|---|---|---|
| Attention | Observation that stops the scroll | Your recent post on churn |
| Interest | The problem you address | Onboarding drops are costly |
| Desire | The outcome you deliver | We cut that 20 percent |
| Action | A tiny ask | Open to a 10-min chat? |
PAS (Problem-Agitate-Solve)
Problem, Agitate, Solve. Use it when the pain is sharp and the prospect may not yet feel its cost. The agitate step names the cost of inaction, which creates urgency without pressure.
PAS
Best for: Strong for pain-led offers; avoid over-dramatizing.
Agitate the cost, not the person. Fear-based framing backfires in DMs.
BAB (Before-After-Bridge)
Before, After, Bridge. Use it for transformation offers where the gap between current and desired state is the selling point. The bridge is how you get them there.
Before
Name the current frustrating state briefly.
After
Paint the better state they want.
Bridge
State how you get them there in one line.
Choosing and testing
Pick one framework per test and compare reply rates with the A/B testing guide. The first-message templates and personalized examples show filled versions. Switching frameworks mid-test scrambles your data, so change one variable at a time.
- Use Hook-Offer-Proof for short, fast DMs.
- Use AIDA or BAB when you have more room.
- Use PAS when the pain is clear but underfelt.
- Test one framework at a time against your baseline.
Filling each framework with real inputs
A framework is only as good as the specifics you drop into it. The structures below show the slot for a real observation; leaving it generic is the most common failure. Write the observation from the recipient's own profile or post so the framework reads as a person, not a template. The personalized examples page shows filled versions side by side.
| Framework | Observation slot | Offer slot |
|---|---|---|
| Hook-Offer-Proof | Saw you [specific detail] | I help [audience] [outcome] |
| AIDA | Your recent post on [topic] | We cut that [metric] [amount] |
| PAS | [Audience] lose [cost] to [problem] | We fixed it for [example] |
| BAB | Before: [current state] | Bridge: [how you get them there] |
Common framework mistakes
Frameworks fail in predictable ways: the observation is fabricated, the offer is two paragraphs, or the ask is a meeting instead of a reply. Each mistake costs reply rate more than the framework gains. The write-better-hooks guide fixes the observation; the first-message templates keep the offer tight.
- Inventing a detail the recipient cannot recognize.
- Burying the offer under explanation.
- Asking for a meeting in the first message.
- Switching frameworks mid-test and blaming the wrong one.
The framework did not fail; the empty observation slot did. Fill it every time.
Adapting frameworks across platforms
The same framework flexes by platform. Hook-Offer-Proof stays three lines on Instagram but can expand with an AIDA flavor on LinkedIn where length is tolerated. The breakup message works as a final PAS or Hook-Offer-Proof close depending on the tone that fit the thread. Match length and tone to the platform while keeping the underlying structure intact.
Pick the base framework
Start with Hook-Offer-Proof unless pain is sharp.
Match length to platform
Short on IG and X, longer on LinkedIn.
Keep the observation first
Whatever the platform, relevance leads.
Hook library by framework
Having a few ready hooks per framework removes the blank-page pause. The examples below are starter observations you adapt to your recipient; the personalized examples page shows filled versions. Keep a running library and add winners as you find them, because the hook is the single highest-leverage line you write.
| Framework | Starter hook |
|---|---|
| Hook-Offer-Proof | Saw you [detail] and thought of [outcome]. |
| AIDA | Your post on [topic] caught me, because [interest]. |
| PAS | Most [audience] lose [cost] to [problem]. |
| BAB | Right now [before]; imagine [after]. |
How to A/B two frameworks correctly
To compare frameworks, you must hold everything else constant: same audience, same offer, same send window. Only the structure changes. The A/B testing guide details the math; the key discipline is a fixed batch size so the result is meaningful, not a coin flip on 20 messages.
Pick one audience and offer
Hold them identical across variants.
Split the batch evenly
Across the two frameworks.
Send in the same window
So timing does not confound.
Compare reply rate
Keep the winner as the baseline.
If you change the list while changing the framework, you learn nothing. Hold the variable.
When a framework stops working
No framework works forever; audiences tire of a structure, or your list shifts, and the rate falls. Recognize the decline early and rotate rather than forcing the same skeleton on a changed audience. The campaign scorecard records the decline so you do not rationalize it away.
Watch reply rate per framework
Weekly, not yearly.
Flag two bad batches
Below your baseline, not the band.
Rotate the structure
Try a different framework.
Test against the old baseline
Confirm the new one is better.
Length and formatting by platform
Beyond structure, length and formatting change by platform, and getting them wrong gets the message skimmed or flagged. The table summarizes the safe envelope per platform so the framework fits the container it lands in.
| Platform | Length | Formatting |
|---|---|---|
| 3 to 5 sentences | Paragraph, no emoji spam | |
| 2 to 3 sentences | Casual, one line break | |
| X | 1 to 2 sentences | Tight, no fluff |
| Value first | No pitch, soft sign-off |
Framework and the breakup message
Every framework should end with a graceful exit, the breakup message, so non-responders are closed politely rather than abandoned. A final Hook-Offer-Proof or PAS close often prompts a reply from the silent, and it protects deliverability by not leaving threads open. The breakup examples show the phrasing; keep it short and pressure-free.
The breakup is not a last pitch; it is a polite close that sometimes converts.
Suggested image brief
| Placement | Purpose | Filename and alt text |
|---|---|---|
| After the direct answer | Create an original AI-generated workflow graphic that summarizes the decision, metric, and next action for this topic without third-party logos. | cold-dm-message-frameworks-workflow.webp - Cold DM Message Frameworks workflow diagram |
Quick checklist
- One framework chosen as the test baseline.
- Hook references a true, checkable detail.
- Offer stated in one outcome-led sentence.
- Proof point included where the framework allows.
- Message kept to three to five sentences.
- Variant compared with A/B testing guidance.
Related: Personalized Examples · First Message Templates · Breakup Examples · Write Better Hooks · Best Templates Tool
Frequently asked questions
Which framework is best?
Hook-Offer-Proof for most short DMs; the others when you have room or a specific angle.
Can I combine frameworks?
You can blend elements, but test one primary structure at a time to read the result.
Do frameworks feel robotic?
Only if you skip the specific detail; the observation is what makes any framework human.
How long should the message be?
Three to five sentences for most platforms; longer only where the platform invites it.
Where do I get fill-in versions?
The templates tool and first-message templates provide ready structures.
How do I know which won?
Compare reply rate per framework with the A/B testing guide over a fixed batch.
Forecast your next cold DM campaign.
Run the free calculator — no signup required.
Forecasts are estimates based on user-provided assumptions. Results are not guaranteed.
Benchmarks, templates, and examples on this page are illustrative planning references, not guarantees of performance. Adjust your outreach to comply with platform terms and applicable regulations.