Carlo Bellati's audit covers 3 public posts, including 3 with both reaction and comment counts. The measured median is 18 interactions. AI & building products is the most represented detected theme (1/3; themes can overlap). Start with a concrete follow-up to an existing subject, then test the opening and reply invitation across a consistent four-week sample.
3 of 3 posts have both public counts. Their median is 18 interactions. Compare future experiments with your own sample rather than an unrelated peer score.
“Some days as a founder are just bad. No matter what you do.” has 27 visible interactions. That is 1.5× the sample median. Its topic and opening are candidates for a follow-up experiment; this does not identify why it performed.
1 of 3 posts contain a question. Measured posts generated 13 comments per 100 reactions. Counts can include the author's replies and do not measure lead quality or sentiment.
The median gap between sampled dates is 1.9 days; the longest is 3 days (2 intervals). Missing posts can exaggerate gaps, so this is not a complete publishing calendar.
Profile & visual identityImage preview, section grades, missing data and specific fixesExplore
Photo unavailable
SOURCE PROFILE PREVIEW
Carlo Bellati
Headline not supplied by the source
These are source images, which may be resized. An unavailable image does not mean it is missing on LinkedIn.
PROFILE CHECKLIST
Your first impression, section by section.
Open any section to see the evidence and a practical fix. Unknown data is not a failed check.
Profile picture
UnknownNeeds source data
0 / 15 points assessable · partial coverage
UnknownProfile image supplied—
What we observed
No usable image URL was returned; absence on LinkedIn is unconfirmed.
What to do
Choose a current, recognizable photo. Preview the circular crop at comment size and leave space around your head.
UnknownImage decodes and fits its display shape—
What we observed
The source image could not be inspected. This is not evidence that the profile has no image.
What to do
Use a square original, check sharpness at full size, and preview the circular crop. A small provider thumbnail does not prove your original upload is low resolution.
Cover & visual identity
A+10/10 observed points
10 / 10 points assessable
PassCover image supplied5/5
What we observed
The source supplied an image URL.
Keep it strong
Use a simple cover that says who you help and what you do. Keep important text clear of the profile-photo overlay.
PassImage decodes and fits its display shape5/5
What we observed
Returned image: 800 × 199px. Dimensions and decoding were measured on the returned image, which may be a resized LinkedIn thumbnail. Original resolution, face framing, lighting and professional suitability were not assessed automatically.
Keep it strong
Preview your cover on desktop and mobile. Start with a 1584 × 396 canvas and keep the message readable without tiny text.
Name & identity
A+10/10 observed points
10 / 10 points assessable
PassReadable display name5/5
What we observed
Source text: “Carlo Bellati”
Keep it strong
Use the name people know you by, consistently across your profile and other professional channels. Single names and all writing systems are supported.
PassName field stays focused on identity5/5
What we observed
Source text: “Carlo Bellati”
Keep it strong
Move URLs, contact details and sales calls to action into About or Contact info. Keep your actual name unchanged.
Headline & positioning
UnknownNeeds source data
0 / 15 points assessable · partial coverage
UnknownHeadline supplied—
What we observed
Not returned in readable form by the source.
What to do
State your role or specialty in the headline, using terms your intended reader understands.
UnknownEnough context to explain your work—
What we observed
Not returned in readable form by the source.
What to do
Use this structure: [role or specialty] · Helping [audience] achieve [outcome]. Replace every bracket with a truthful detail.
UnknownHeadline is concise enough to scan—
What we observed
Not returned in readable form by the source.
What to do
Keep the headline focused on one main promise. Move your longer story into About.
About & descriptions
Partial5/5 observed points
5 / 20 points assessable · partial coverage
PassReadable introduction5/5
What we observed
Truncated source excerpt: “Founder/CRO of Zipchat.ai Zipchat.ai: AI-Powered Chat That Turns Ecommerce…”
Keep it strong
Lead with who you help, the problem you solve and the work you do today.
UnknownAn introduction with supporting detail—
What we observed
Truncated source excerpt: “Founder/CRO of Zipchat.ai Zipchat.ai: AI-Powered Chat That Turns Ecommerce…”
What to do
Add a specific example of your work, your approach, and the kinds of opportunities you welcome. Length is a structure check, not a quality guarantee.
UnknownEasy-to-scan paragraph structure—
What we observed
Truncated source excerpt: “Founder/CRO of Zipchat.ai Zipchat.ai: AI-Powered Chat That Turns Ecommerce…”
What to do
Break About into short paragraphs: current focus, relevant proof, then how to connect.
UnknownA clear next step—
What we observed
Truncated source excerpt: “Founder/CRO of Zipchat.ai Zipchat.ai: AI-Powered Chat That Turns Ecommerce…”
What to do
End with one relevant next step, such as who should message you and what to include. This automated check recognizes English phrases and links; review other languages manually.
Experience & proof
UnknownNeeds source data
0 / 15 points assessable · partial coverage
UnknownProfessional experience identified—
What we observed
No readable entries were returned. We cannot conclude this section is empty.
What to do
List relevant roles with an accurate title and organization. Connect them to the work you want to do next.
UnknownExperience includes useful descriptions—
What we observed
Not returned in readable form by the source.
What to do
For each relevant role, explain your responsibility, a concrete contribution, and an outcome you can substantiate. Do not invent numbers. Masked source descriptions need manual review.
Education & learning
UnknownNeeds source data
0 / 10 points assessable · partial coverage
UnknownEducation or learning entry identified—
What we observed
No readable entries were returned. We cannot conclude this section is empty.
What to do
Add relevant education or structured learning you want to share. Name the institution or course accurately; formal degrees are not required for a strong professional story.
UnknownLearning context is described—
What we observed
Not returned in readable form by the source.
What to do
Describe your subject, course or qualification and one relevant project or skill. Use the full qualification name where helpful. We assess clarity, never institution prestige or degree level.
Location & context
A+5/5 observed points
5 / 5 points assessable
PassReadable location5/5
What we observed
Source text: “Genoa”
Keep it strong
Use an accurate city or region if you want it visible. Do not add a home address. Omitted provider data is not penalized.
Read all available profile text
Headline
Not returned in readable form by the source.
About
Founder/CRO of Zipchat.ai Zipchat.ai: AI-Powered Chat That Turns Ecommerce…
This is a truncated source excerpt.
Experience
No readable entries returned.
Education
Name unavailable
Grenoble Ecole de Management
Hidden entries are omitted from this readable preview. Source text is not independently verified.
VISUAL REVIEW
Check how your profile looks.
The automatic image grade checks file readability and shape. Use this checklist to review the visual choices it cannot judge.
Is your face recognizable in the small circular crop?
Is the image sharp, evenly lit, and free of distracting clutter?
Does the banner communicate your work without overlapping your photo?
Are colors and typography consistent with your professional identity?
A transparent presentation checklist, not a LinkedIn ranking, talent assessment or prediction of reach. Grade thresholds: A+ 90, A 80, B 65, C 50, D 35, F below 35.
Score = earned points ÷ assessed points × 100. Coverage is assessed weight out of 100. Unavailable and masked data are excluded. Scores with different coverage are not directly comparable.
Truncated text earns only checks already demonstrated in the visible excerpt; the unseen remainder is not marked wrong. Text rules measure structure, not factual accuracy, writing quality or every language equally.
Photo and banner grades cover image availability, decoding and shape only. Face framing, lighting, expression, visual style and authenticity require a human review. No appearance, identity or personality judgments are made.
An education entry is evidence of profile context, not verified credentials. School prestige, age, degree level and background do not affect the score.
Section weights: photo 15, cover 10, name 10, headline 15, About 20, experience 15, education 10, location 5. Passed checks earn their full weight; improvements earn zero; unknown checks are excluded.
Unknown means the source did not verify the format. Small or unequal groups do not establish a winning format.
Hooks and writing structure
Read writing structure for 3 posts
Text rules applied to every sampled post
Opening
Hook type
Words
Paragraphs
Final CTA
Tags / mentions
Il 17 Settembre a Focus Fashion & Beauty (Netcomm Forum , Milano) non farò il solito panel sull'Agen
Number-led
52
3
Detected
0 / 0
I couldn't wait the monthly update for this milestone. 🎉
Personal opening
22
1
Not detected
0 / 0
Some days as a founder are just bad. No matter what you do.
Statement
109
7
Not detected
0 / 0
Hook type uses the first nonempty line. CTA detection checks the final paragraph for a question or action phrase. Paragraphs use blank-line breaks. These rules can miss meaning and linked mentions.
Hashtags in the sample
No data available for this grouping.
Tags travel with the topic, author and format. These associations cannot isolate a hashtag's effect.
Recommended next stepsPrioritized actions, effort and what to measureExplore
PRIORITIZED RECOMMENDATIONS
Less guessing. Clear next steps.
Ordered from building a useful baseline to testing individual writing choices. Potential impact describes the purpose of an experiment, not a promised uplift.
01Develop a specific follow-up
Medium effort
OBSERVATION
The sample includes “Some days as a founder are just bad. No matter what you do.”. Its visible interaction total is 27.
DO THIS
Build a follow-up to “Some days as a founder are just bad. No matter what you do.”. Add a concrete example, a decision you made and one lesson. Keep the format similar so topic development is the main experiment.
Potential impact: Test repeatable interest in an existing subject
Metric to watch: Reactions and comments after the same seven-day window; compare with the baseline median
Median text length is 52 words with 3 paragraphs. 1 of 3 openings are classified as statements.
DO THIS
For the next post related to “Some days as a founder are just bad. No matter what you do.”, open with a specific problem and a concrete outcome you can substantiate. Move context into the second paragraph. Avoid invented results or curiosity gaps that the post does not resolve.
Potential impact: Improve clarity; any engagement effect remains to be tested
Metric to watch: Median interactions across at least three comparable posts per opening approach
1 of 3 final content lines contain a detected question or action phrase; 1 posts contain a question anywhere. These are text rules, not semantic judgments.
DO THIS
End one upcoming post with a focused trade-off question connected to ai & building products. Ask readers to share a decision and why. Read replies before deciding whether the discussion was useful.
Potential impact: Test whether a clearer invitation supports useful discussion
Metric to watch: Public comments per post and manual review of reply relevance
3 dated posts span 3.8 days. Observed cadence is 3.7 posts per week.
DO THIS
Choose two slots per week for the four-week experiment below, or reduce the schedule if your capacity is lower. Keep the slots consistent and log every published post. Two slots are a planning choice, not a LinkedIn benchmark.
Potential impact: Create comparable records and a manageable learning loop
Metric to watch: Planned versus published posts, time spent, and median interactions at seven days
0 of 3 posts contain hashtags and 0 contain visible @mentions. LinkedIn may return linked mentions as plain text, so detection can undercount.
DO THIS
Use relevant topic labels only when they clarify the subject. Mention people only when they contributed or are directly relevant. Test a consistent tagging approach on comparable posts; do not attribute performance changes to tags alone.
Potential impact: Keep context and attribution useful
Metric to watch: Median interactions for tagged/untagged samples, with sample sizes and topic differences
Turn one idea related to “Some days as a founder are just bad. No matter what you do.” into a concise checklist or document post. Keep a similar topic and ask the same type of question. Label this a new format experiment, not a proven winning format.
Potential impact: Learn whether a different presentation helps explain the same idea
Metric to watch: Public interactions at the same post age; at least three posts in each format before comparison
Content ideas & basic outlineThree starting ideas and a simple structure for your next postExplore
INCLUDED IN YOUR FREE REPORT
A starting point for your next post.
Simple ideas from this profile’s available evidence. Choose one, add your own experience, and review the result.
IDEA 1
The next chapter
Follow up on “Some days as a founder are just bad. No matter what you do.”: what changed since publishing, what stayed difficult, and the next step. Only include outcomes you can verify.
Break the work behind “Some days as a founder are just bad. No matter what you do.” into three decisions. Add an example or screenshot you have permission to share.
Some days as a founder are just bad. No matter what you do.
You hit a wall you didn't see coming.
You feel like you don't have what it takes to get to the next level.
You look back at the day and feel like you wasted it.
Nothing worked. Nothing moved.
Then one small win shows up. A customer reply. A number that ticked up. A "this actually works" from someone on the team.
And suddenly the whole day flips.
Not because the problem went away. Because you remembered why you're doing this.
The bad days aren't a bug in the journey. They're part of it.
Keep pushing.
Il 17 Settembre a Focus Fashion & Beauty (Netcomm Forum , Milano) non farò il solito panel sull'Agentic Commerce.
Intervisterò dal vivo Lanfranco Tartarelli che racconterá come Piquadro, Lancel e The Bridge usano Zipchat ogni giorno sul loro ecommerce e di come sta ridisegnando l'esperienza d'acquisto nel Fashion con agenti AI.
Ci vediamo li? 👇
16reactions
0comments
A RECORD YOU CAN RETURN TO
Saved audit history.
Reading a saved report does not start a new collection. A fresh snapshot requires confirmation and uses your current review allowance. Refresh availability and shared spending limits are checked before it starts.
A comparison appears after another audit is saved. A single snapshot cannot establish follower growth.
Help improve this report
Independent analysis of publicly available information. Not affiliated with LinkedIn or the profile owner. Impressions, profile views, private analytics, and follower growth are not available in this snapshot.
