Am I Chasing Them? A Data-Based Check for One-Sided Effort

You may be chasing when you repeatedly create the contact, reopen stalled conversations and supply most of the curiosity while the other person mainly responds. One uneven week is weak evidence. A repeated imbalance across several signals—and a pattern that stays uneven after busy periods pass—is more meaningful. Chat analytics can make that pattern visible, but they cannot tell you why it exists or decide what the relationship means.

What ‘chasing’ looks like in a chat

Chasing is best treated as a working description of behavior, not a diagnosis. It can mean initiating again after silence, asking new questions when an exchange is fading, sending several messages before receiving a reply, or continually finding the next subject. A one-sided relationship is broader: emotional support, plans, affection, repair after conflict and offline responsibility may also be uneven. A WhatsApp export observes only the part that happened in that chat.

That distinction prevents a common mistake. Someone can write fewer messages yet contribute strongly through calls, in-person plans or practical care. Another person can send many messages but offer little responsiveness. Research describes perceived partner responsiveness as feeling understood, cared for and appreciated; a message count does not directly measure any of those experiences.

Four signals to read together

MyChatWrap’s Reciprocity card brings four additive behaviors into one view. Each bar shows how the observed total is divided between two participants. These are clues about how the conversation is maintained, and each clue answers a different question.

The example is more informative than a single 79/21 number because three related behaviors point in the same direction. The nearly even media split complicates the story in a useful way: Person B is not absent. They participate differently. The next question is whether that difference fits the relationship or leaves Person A feeling chronically unsupported.

1. Conversation starts show who creates opportunities

A start is most useful when it means contact after a meaningful break, rather than every first message of a calendar day. A persistent 68/32 split says A usually reopens contact. It does not say B rejected those openings. Check whether B responds substantially, initiates through another channel, or has less opportunity because A always messages first.

2. Questions show explicit curiosity

Questions can invite detail, check understanding and keep an exchange moving. A 74/26 split may show that A carries more of this visible curiosity. Yet question detection misses curiosity expressed as statements—‘tell me more’—and cannot judge whether a question is warm, routine or confrontational. Read representative messages around the count before interpreting it.

3. Double texts show follow-up pressure, not desperation

A double text is simply another message sent before the other participant replies. Several short bubbles can form one natural thought, and schedules can delay an answer. The 79/21 split matters here because it aligns with initiation and questions. It still should not be converted into a universal ‘too much’ threshold; whether to send another message depends on context, expectations and boundaries.

4. Shared media can reveal another way of contributing

Photos, voice notes, links, stickers and documents often create topics or share daily life. The 55/45 split is close. It shows why effort should remain multidimensional: B may contribute fewer openings and questions while still bringing material into the relationship. Media count cannot tell whether an item was intimate, logistical or merely forwarded.

Normalize the numbers before calling it one-sided

Raw shares are easiest to misread when the two people have different opportunities to act. Use a period long enough to represent normal life, then check the circumstances around it. Useful adjustments include:

• Message volume: a person who sends more messages has more chances to ask questions or double text.
• Schedule and time zones: work shifts, caregiving, travel and sleep can determine who is available to start.
• Channel mix: calls, meetings, Instagram, SMS and in-person contact are invisible in a WhatsApp-only analysis.
• Communication style: one person may use many short bubbles; another may send one complete reply.
• Relationship stage: initiation naturally looks different in early dating, long-term partnership, friendship and reconnection.
• Agreements and boundaries: an explicitly preferred rhythm matters more than an internet rule about how often people should text.

Compare periods, not just people

A long-history total can hide change. Split the chat into comparable windows—such as the previous eight weeks and the eight weeks before that—and ask whether the shares moved. A stable 65/35 rhythm that both people like is different from a shift from 50/50 to 80/20 after a particular event. The change supplies a better question; it still does not supply the cause.

Choose windows that include similar circumstances. Comparing a holiday with an exam month, illness or a work deadline creates an obvious confound. Also inspect absolute activity: a share can rise because A did more, because B did less, or both. Percentages alone cannot distinguish those paths.

What chat data can—and cannot—tell you

The data can describe recorded behavior: who started, asked, followed up and shared. It can help you replace a vague feeling with a specific observation and notice whether the pattern is stable or changing. It cannot read private motives or the quality of everything outside the export.

• It cannot prove love, attraction, commitment, boredom or manipulation.
• It cannot decide whether an unequal split is fair or acceptable to either person.
• It cannot measure unrecorded calls, dates, practical help, emotional care or household labor.
• It cannot diagnose attachment style, abuse or a mental-health condition.
• Automated counts can be imperfect; use message examples and your lived context to audit them.

A five-step effort check

Use analytics to prepare a clearer conversation, not to prosecute a case. This sequence keeps the evidence proportional to what it can show.

1. Pick a representative period. Exclude a crisis, holiday or unusual deadline unless that event is what you want to examine.
2. Read all four reciprocity bars. Note which ones lean the same way and which ones contradict the story.
3. Open a few message examples around starts and follow-ups. Check whether the exchange became mutual after A opened it.
4. Compare with an earlier equal-length period and with contact outside WhatsApp.
5. Translate the pattern into an ‘I’ statement and a concrete request: describe what you observe, how it feels, and what rhythm would work better.

Turn the chart into a better conversation

A useful discussion stays curious. Avoid ‘the data proves you do not care.’ Name the limited observation, invite missing context and ask for a workable change. You are testing whether you can understand and respond to each other—not whether both people can manufacture identical percentages.

• ‘I noticed I have been starting most of our chats lately, and I feel unsure about whether you want this much contact. How has texting felt to you?’
• ‘Our routines may be different. What kind of contact feels realistic and caring for both of us?’
• ‘Would you be willing to initiate sometimes, or should we rely more on calls and plans?’

Research behind the interpretation

Relationship research gives two guardrails. Perceived responsiveness—feeling understood, appreciated and cared for—is associated with well-being, but a log cannot directly measure that perception. Equity research finds that feeling underbenefited can accompany lower satisfaction and commitment, while communal-relationship theory explains why healthy close relationships are not simple tit-for-tat exchanges. Studies of remote communication also show that the meaning of texting depends on context and perceived responsiveness. These findings support asking better questions; they do not validate a universal reciprocity cutoff.