Why We Overthink Text Messages — And How AI Can Help Us See the Bigger Pattern

Have you ever reread the same text message so many times that it started to mean something completely different? That is the problem behind What Brandon Thinks, an AI-powered tool designed to analyze WhatsApp and iMessage conversations for patterns in interest, effort, consistency, mixed signals, and follow-through.

Most of us have probably experienced some version of this.

Someone sends:

“Maybe this weekend.”

And suddenly your brain starts working overtime.

Are they interested?

Are they avoiding making plans?

Are they actually busy?

Did I say something wrong?

Should I reply now?

Should I wait?

The strange thing is that the message itself may not contain enough information to answer any of those questions.

But the conversation around it might.

The Problem With Analyzing One Message at a Time

Text messages feel incredibly important when emotions are involved.

A short reply can feel cold.

A delayed response can feel intentional.

A missing emoji can suddenly seem meaningful.

And when we care about the person on the other side of the conversation, it becomes surprisingly easy to turn tiny details into evidence.

But individual messages are noisy.

Someone may reply late because they're working.

They may sound brief because they're tired.

They may cancel plans because something genuinely came up.

One event rarely tells us very much.

Patterns are different.

One slow response probably means nothing.

Twenty conversations where one person consistently takes hours to respond while the other replies immediately may tell us something about the dynamic.

One cancelled plan happens.

Repeatedly suggesting plans without ever making them happen is different.

That's where looking at the whole conversation becomes more useful than obsessing over a single message.

Words Are Only Part of the Conversation

When people try to understand relationships through text messages, they usually focus on what was said.

But conversations contain other signals too.

Who initiates?

Who asks questions?

Who keeps the conversation alive?

Who remembers details?

Who suggests meeting?

Who actually follows through?

Does the effort feel balanced?

Does someone's tone change over time?

Those behaviors can sometimes tell us more than the words themselves.

Imagine someone regularly says:

“We should definitely meet soon.”

That sounds positive.

But if six weeks pass and every attempt to choose an actual date becomes vague, the words and the behavior aren't telling exactly the same story.

That's a mixed signal.

And mixed signals are usually confusing precisely because both sides contain something that feels real.

The words may show interest.

The behavior may show inconsistency.

Both can exist at the same time.

This Is Where AI Gets Interesting

AI doesn't need to become a digital relationship fortune teller to be useful.

In fact, it probably shouldn't.

No AI can prove exactly what another person feels simply by reading their messages.

But AI is very good at looking across large amounts of information and identifying recurring patterns.

That makes conversation analysis an interesting use case.

Instead of asking:

“Does this person like me?”

you can ask better questions.

Has their effort changed?

Who usually starts conversations?

Are they more engaged at certain times?

Do they regularly make plans and then avoid following through?

Has their tone become noticeably warmer or colder?

Are their words consistent with their actions?

Those questions don't require mind-reading.

They require observation.

Observation vs. Interpretation

This distinction is important.

Consider these two statements:

“They don't care about you.”

and:

“You initiated eight of the last ten conversations.”

The first is an interpretation.

The second is an observation.

Or compare:

“They're playing games.”

with:

“They suggested meeting three times, but none of the plans became specific.”

Again, one tries to explain someone's intentions.

The other describes what actually happened.

The second type of information is often much more useful.

You can decide what it means.

You can talk about it.

You can compare it with the rest of the relationship.

But you're starting from something concrete instead of guessing what is happening inside another person's head.

Why We Struggle to See These Patterns Ourselves

The funny part is that the information is usually already sitting on our phones.

We have the messages.

We were there for the conversations.

So why do we sometimes need another perspective?

Because emotion changes attention.

We remember the message that made us smile.

We remember the compliment.

We remember the night someone stayed awake talking to us until 2 a.m.

We remember the heart emoji.

We remember the promise.

But we may not notice that we're always the one restarting the conversation.

We may overlook how often plans get cancelled.

We may ignore that someone's tone changes whenever the relationship becomes more serious.

We naturally pay more attention to moments that confirm what we hope is happening.

That's human.

Sometimes seeing the entire conversation summarized as patterns makes those dynamics easier to recognize.

Interest Is Not the Same as Effort

This is another useful distinction.

Someone can genuinely enjoy talking to you while still putting very little effort into the relationship.

Someone can be affectionate but inconsistent.

Someone can show strong interest in messages while rarely following through in real life.

These things aren't mutually exclusive.

That's why reducing a conversation to a simple “they like you” or “they don't like you” answer is often not very helpful.

Relationships are rarely that binary.

A better analysis might look at several dimensions separately:

• Interest
• Effort
• Consistency
• Tone
• Mixed signals
• Follow-through

Once those pieces are separated, the situation can become much easier to understand.

AI Should Help You Think, Not Think for You

There is an important boundary here.

Technology shouldn't make major relationship decisions for us.

An AI shouldn't decide whether you should date someone, end a relationship, confront someone, or trust them.

But it can help organize information.

It can surface patterns.

It can give you another perspective.

And sometimes that's exactly what people need when they've been staring at the same messages for too long.

The goal isn't:

“Tell me exactly what this person is thinking.”

A healthier goal is:

“Help me understand what has actually been happening.”

That is a much more realistic use of AI.

Looking at the Bigger Picture

This is what makes tools like What Brandon Thinks interesting.

Instead of treating one message as the answer, the idea is to zoom out and examine the broader conversation.

Maybe the person is interested.

Maybe their effort is inconsistent.

Maybe their tone has changed.

Maybe they say the right things but rarely follow through.

Maybe everything is actually much more balanced than you thought.

Whatever the result, the useful part is seeing the pattern more clearly.

Because sometimes the message you've been rereading for twenty minutes isn't the important part at all.

The answer may already be visible across the hundreds of messages that came before it.

We just need to stop staring at one sentence long enough to see it.