Rhea

The two 'don't recognise this charge' ones are the actionable half and nobody's mentioned them. That's almost always a descriptor problem - whatever appears on the statement is probably your legal name or some abbreviation of your domain that doesn't match the brand they bought from.

Set the statement descriptor to something recognisable, put your support email on the receipt, and send the receipt immediately at purchase. I halved these by changing a string in a settings page.

Half agree. The energy cost is the real damage and you are right about that. But there is one asymmetric move worth ten minutes: put your customers' logos or names or a couple of quotes on your page, if you have permission. A clone can copy your words. It cannot copy the six people who agreed to be named next to them.

Careful with required fields, because they will make your metrics look wonderful for the wrong reason. A form with six mandatory boxes reduces vague tickets by making a chunk of people close the tab, and those people do not become happy customers who figured it out, they become quiet customers who churn at renewal. If you build a form, make everything optional except one free text box, and put the auto-captured technical detail in behind the scenes where it costs the user nothing.

On the data half, ask three concrete questions rather than trying to parse the policy prose.

One: can I export my full history in a usable file, today, without emailing support. Two: does deleting my account delete the data or just my access to it. Three: is there a setting to keep it local or is the cloud mandatory for the features I use.

Whichever product answers those well is the one I would buy next. The policies are all written to be broad. The export button either exists or it does not, and that is the honest signal.

Wet brushing with a paddle brush, by a distance, and it's the one you described as unavoidable.

Hair swells when it's wet and the internal structure is at its weakest, so a brush that drags a whole sheet of hair from the roots down is applying tension exactly when the fibre can least take it. A paddle brush is the worst tool for it because the pins are set in a flat bed and can't flex around a knot - it pulls the knot instead of passing through it.

I switched to a flexible detangling brush, sections of about four, starting at the ends and working up, always with conditioner still in. The breakage band grew out over about eight months and hasn't come back. I still blow dry.

Housekeeping before you scale this past 120: send from a separate domain, not your main one. Get SPF, DKIM and DMARC aligned on that domain and check alignment with a real test rather than assuming your provider did it. Plain text, no tracking pixel, no images, no unsubscribe footer that makes a one-to-one email look like a newsletter, because that is what pushes a personal message into a promotions tab.

Daily volume is not your constraint at this size - a paid mailbox is good for a couple of thousand messages a day, a trial account far less - but that ceiling is irrelevant when you are sending forty. Complaints are the constraint. One person marking you as spam out of forty is a far worse signal than one out of two thousand.

Politely: 120 contacts and seven replies is not enough data to tune a cadence on. You are optimising the gap between messages when the two variables that matter are who is on the list and what the first line says.

The other thing nobody mentions at this size - your sending setup is doing more work than your copy. If those 120 went from the same domain your customers get invoices from, and a few of them marked it as spam, you have quietly damaged something that matters much more than this campaign.

I did the use-the-dead-week-for-admin thing for a year and it made everything worse. Admin expands to fill whatever you give it, and then the good week opens with a backlog of half finished small things instead of the one thing that mattered.

Now the on-call week is genuinely nothing. I do not open the repo. One fifteen minute email sweep on Wednesday, that is it. It feels wasteful and my output went up.

Going to push back on mastering being the answer, because it is the thing people reach for when a mix feels unfinished. Mastering is not a finishing process for a mix that has problems, it is a final polish and a translation step for a mix that is already right. If your mixes never sound finished, sending them to mastering gets you the same unfinished mix, slightly louder and slightly brighter. Spend on the mix, or on your ability to hear the mix.

I use that experiment whenever someone insists it is the gyroscopes. It does not tell you what the answer is, but it settles what the answer is not, which is most of the battle with this question.

Worth separating the two things sold under that label. Physical scrubs are abrasive particles in a cleanser, they remove product buildup and loose flakes, and that's genuinely all. Chemical ones use an acid, usually salicylic, which is oil-soluble so it gets into the follicle rather than sitting on top.

If your issue is buildup from dry shampoo and styling product, either works and the scrub is fine. If your issue is flaking that comes back within days regardless, neither is treating the cause, and you'll be buying the £22 tub monthly forever.

A 0.5% reply rate is a targeting problem wearing a deliverability costume. 400 identical emails to a scraped directory is a list, not a segment.

Forty emails you actually researched will beat those 400 every time. The structure that works for me:

• one line proving you looked at their thing specifically, and it has to be something you could not have merge-fielded
• one line about a problem they have, in their words not yours
• one ask that costs them under two minutes

No links in the first email. Links hurt both deliverability and reply rate, and the first email does not need one.

You sent cold mail from the domain that carries your transactional mail. That is the mistake, and everything else is downstream of it.

How it should be set up:

• cold outreach goes out from a separate lookalike domain you do not care about (getyourtool.com alongside yourtool.com), with its own SPF, DKIM and DMARC
• that domain gets warmed for four to six weeks before it sends anything real, ramping from 10 a day
• transactional mail lives on its own subdomain with its own DKIM key, so its reputation is scored separately from marketing

Recovery for what you have: stop all cold sending today, not next week. Keep sending only mail people asked for, receipts, resets, digests to engaged users. Reputation is a rolling window and it climbs back over three or four weeks if the recent signal is clean. Sign up for Google Postmaster Tools so you can see the curve instead of guessing at it.

Put DMARC on p=none first and read the aggregate reports for a fortnight before you tighten it. Half the people who go straight to p=reject break their own invoicing tool and then blame the mailbox provider.

This was exactly my problem, with one addition: I put the command in /etc/rc.local and it silently stopped working after a distribution upgrade. The systemd unit version has survived two upgrades since.

Second most likely thing given the time of day: temperature. Coin cells and AAA cells lose a chunk of usable voltage when they get cold, and an unheated garage at 3am in the cold half of the year is the coldest point of the day. The sensor browns out, misses its check ins, and gets aged out.

If the failures started when the weather turned, that is your answer. Lithium AA or AAA cells hold voltage far better in the cold than alkaline.

Counterintuitive numbers from my side. The month after a near clone of my thing launched, my signups went up about twelve percent and stayed there.

Two of them told me they had been comparing us, which means the clone did the category education I could not afford to do. That only works if you are the one with the reviews, the docs and the two years of changelog. If you are both nine days old, competition is just noise.

So the practical question is not whether they copied you. It is whether a buyer comparing the two pages side by side has a reason to pick yours. If they do not, the roadmap was never the problem.

Asked for the click data. Median was 210 clicks over the last five slots, not the number I had in my head. Still might do it but at a very different expected value.

This is the most useful thing in the thread, thank you. 620 clicks off a paid slot puts the 210 median I was quoted in perspective too.

Do not aim at zero yet. Aim at knowing what the two hours are made of, because right now you are describing it as one lump and it is not.

Time yourself across the next five clients with a stopwatch and a list of tasks. Every time I have done this the answer was six or seven distinct tasks where two of them ate most of the minutes. In my case it was the import fixing and one specific configuration question, and the rest was small.

Then automate only the two. Not the flow, not the whole onboarding, the two. My equivalent went from two hours to about thirty five minutes with one importer improvement and one better default, and I had been planning a full self serve rebuild that would have taken a season.

The remaining thirty five minutes is a call, and the call is where I learn what to build next, so I am in no hurry to remove it.

Both experiences are common and I suspect it depends on whether the underlying issue is a skill gap or anxiety. Drilling helps the first and feeds the second.

Placement matters more than count. Two repeaters in sensible places beat five clustered around the office where the coordinator already has good coverage.

Three habits, in rough order of usefulness.

First, follow actual cosmetic chemists rather than stylists or brands. There are a handful writing publicly - the ones with a chemistry doctorate who cite papers and will happily say "this ingredient does very little" about a product category their audience loves. That last part is the filter.

Second, read ingredient lists directly. There are free sites that decode INCI names and tell you what a given ingredient is for; ten minutes there kills more marketing claims than any blog post.

Third, when a claim has a study attached, check whether it was done on hair tresses in a lab or on people's heads. Almost all of them are the former, and it matters enormously.

Cadence lock has a name. Somehow that makes me feel better about the fortnight I spent wondering if my heart was broken.

One practical warning: whatever you pick, write down the date you switched. In six months you'll look at a chart with a step in it and spend an evening investigating a change you made yourself.