Use the AI Research Brief for Deeper Personalization

Some outreach personalizes the wording.

getsignals can personalize the reasoning behind the wording.

That is what the AI Research Brief is for.

Instead of asking the model to generate copy directly from a handful of variables, getsignals can first compile a structured internal research layer and then ask the writer to act on that research.

This is one of the clearest examples of how getsignals is built differently from standard AI outreach tools.

What the AI Research Brief is

The AI Research Brief is a structured JSON research layer generated automatically inside the AI writing process.

You trigger it by using:

{{prospect_analysis_brief_json}}

inside an [[AI:...]] block.

When the system sees that tag inside the AI instruction, it does not treat it like a normal merge variable.

Instead, it runs a two-step process:

1. compile the research brief

2. write from that brief

That means the model gets a deeper, more organized understanding of the lead before it writes anything.

Why this matters

Most AI outreach tools try to write immediately.

They take a few fields, pass them into a prompt, and generate copy. That can work for light personalization, but it often leads to shallow output because the model is reacting to fragments rather than an organized picture.

The AI Research Brief changes that.

It gives the model a chance to reason before it writes.

That helps it answer questions like:

• What problem seems most likely here?

• What is this person actually signaling?

• What do their posts and their own comments on other posts suggest they care about?

• Which angle best fits the sender, the offer, and the Signal?

• What should be emphasized, and what should be ignored?

That is why the output can feel more strategic and more human.

How the two-step process works

When you use {{prospect_analysis_brief_json}} inside an AI block, the system works like this:

Step 1: Research compilation

A server-side research prompt assembles a structured brief from the available context.

Step 2: Writer action

The AI writer then uses that research brief inside your [[AI:...]] instruction to generate the final copy.

So the system is not just writing from raw inputs.

It is writing from analyzed inputs.

That is a major difference.

What goes into the research brief

Depending on the lead and the campaign context, the research brief can combine:

• the Signal that surfaced the lead

• the signal post

• the lead's recent posts

• the posts the lead has commented on

• what the lead said in those comments

• the lead's LI profile context

• the lead's company context

• sender context

• workspace firmographics

This is why the AI Research Brief is so powerful in getsignals.

It does not just know who the lead is. It can reason about:

• what they are saying publicly

• what they seem to care about

• what kind of problem-awareness or buyer intent is emerging

• which angle is most likely to resonate

Why this fits a Signal-first methodology

The AI Research Brief is especially valuable in getsignals because campaigns are supposed to start from a Signal.

That means the model is not inventing the reason for outreach. The reason already exists.

The brief helps the system preserve and interpret that reason all the way through.

In a Signal-first workflow:

• the Signal identifies why this lead matters now

• the research brief compiles the full context around that Signal

• the writer turns that into the next action

That is how you avoid the common problem where the context dies the moment the lead enters a campaign.

When to use the AI Research Brief

You do not need the research brief for every step.

Use it when the message needs deeper reasoning, not just surface personalization.

Good use cases

Use the research brief when you want the model to:

• identify the strongest pain point

• choose the best outreach angle

• synthesize post history and comment history

• connect the Signal to the sender's value proposition

• write from full buyer context instead of a narrow prompt

Lower-need use cases

You may not need it when:

• the step is very short and constrained

• a simple merge field plus light AI is enough

• the message only needs one obvious Signal reference

• you are optimizing for speed over depth

Best surfaces for the research brief

Messages

Messages are one of the best places to use the research brief.

They give the model enough room to:

• interpret the Signal

• connect it to the lead's recent activity

• infer the most relevant problem or angle

• write a message that feels like a continuation, not a cold opener

This is usually the strongest default use case.

InMail body

InMail body is also a strong fit.

The subject line stays simple, but the body can carry richer reasoning.

That makes the research brief useful when you want the body to reflect:

• the strongest pain point

• a relevant point of view

• a better-chosen ask

Connection notes

Connection notes can use the research brief, but carefully.

Because notes are constrained by the invite limit, the benefit is less about including more context and more about selecting the cleanest, sharpest single angle from the available context.

Use it only when the extra reasoning is worth the tighter space.

The strongest variable combinations

The research brief becomes more valuable when combined with the right context tags.

For deeper message personalization

A strong combo can include:

• {{signal_post_snippet}}

• {{target_post_history}}

• {{target_comment_history}}

• {{company_post_history}}

• {{prospect_analysis_brief_json}}

That gives the system:

• the original Signal

• the lead's recent voice and themes

• company-level context

• a compiled reasoning layer before final writing

For note personalization

A lighter combo can include:

• {{signal_post_snippet}}

• {{headline}}

• {{company}}

• {{prospect_analysis_brief_json}}

That works best when the writer's job is to choose the single strongest angle, not include lots of detail.

What the brief helps the writer decide

The AI Research Brief is valuable because it helps the writer answer harder questions before generating copy.

For example:

• What is the real pain under the surface?

• Is this person problem-aware or actively in-market?

• Which idea from the Signal matters most?

• What language should the writer echo from the prospect's world?

• What should the sender focus on first?

• What kind of message belongs here: perspective, question, challenge, or soft CTA?

That is not basic personalization. That is strategic message selection.

What this is not

The AI Research Brief is not:

• a visible merge field for end readers

• a normal variable you drop in as-is

• a replacement for strong Signals

• a fix for weak campaign strategy

It is a reasoning layer.

If the Signal is weak, the brief cannot invent real timing. If the campaign flow is poor, the brief cannot fix the whole system on its own.

It works best when:

• the Signal is strong

• the context is rich

• the step is designed well

• the writer prompt is clear about what it should do with the brief

Best practices

Use it when depth actually matters

Do not add the research brief by default to every step. Use it where deeper reasoning improves the result.

Pair it with real context

The brief becomes strongest when it is supported by the right source material:

• signal post

• target post history

• target comment history

• company context

Give the writer a job, not just data

Do not only provide the brief. Tell the AI what to do with it.

For example:

• identify the strongest angle

• choose the most relevant pain

• write one pointed follow-up

• connect the prospect's stated problem to the sender's value

Keep the final output aligned with the step

The brief may be deep. The final copy should still fit the step:

• concise for notes

• richer for messages

• body-focused for InMail

Common mistakes to avoid

Using it without a strong Signal

If the reason for outreach is weak, the brief will not create real relevance by itself.

Treating it like a normal merge field

It is not a display variable. It is a trigger for a deeper two-step AI flow.

Feeding the brief into a vague prompt

The research layer is powerful, but the writer still needs a clear job to do.

Using it where a simpler prompt would work

Not every step needs deep reasoning. Sometimes a direct prompt and a few strong context variables are enough.

Final advice

Think of the AI Research Brief as the layer that turns context into judgment.

Signals tell you why this lead matters now.
Merge variables bring in the raw facts.
The research brief organizes the picture.
The writer turns that picture into the next best action.

That is how getsignals produces outreach that feels more aware, more timely, and more grounded in what the buyer is actually signaling.