The real lesson from a fitness coach that lives in iMessage
Most social media operators I know are drowning in dashboards, not lacking them. We have schedulers, analytics suites, link-in-bio tools, AI caption generators, and a graveyard of half-used apps we opened twice and forgot. What actually changes behavior isn’t another tab — it’s a tool that lives where we already are. That’s why Kilo, an agentic fitness coach you text in iMessage, caught my attention even though it has nothing to do with social media. The distribution model is the story. If a coaching product can win by meeting users inside their default messaging app, the same logic should reshape how we think about publishing, community, and client work across Instagram, TikTok, YouTube, X, LinkedIn, Facebook, Threads, and Pinterest. The lesson isn’t the fitness plan. It’s the interface.
What Kilo actually is, stripped of the launch-page gloss
The maker, Zac, describes Kilo as an agentic fitness coach you text in iMessage. It builds your program, sends your workouts, and uses your training history and check-ins to recommend adjustments. When your schedule or recovery changes, you talk it through with a coach that has your training context. The motivation was personal: Zac says he finished a marathon and wanted to keep lifting while preserving muscle, but couldn’t find a plan that fit both goals. The current state is a free private beta, with lifters onboarded in small groups. Pricing beyond the beta, user counts, and retention data are not disclosed.
That’s the entire factual footprint. No revenue figures, no “10x” claims, no growth numbers. Which is refreshing, and also means the interesting analysis has to come from pattern-matching against the rest of the tooling landscape rather than from the launch copy itself.
Why the iMessage interface is the actual product
Every social media manager I’ve worked with has the same complaint about their stack: the tools are fine, but the workflow is fragmented. You plan in Notion, design in Canva or CapCut, schedule in Buffer or Later, check performance in Metricool or the native Instagram and TikTok dashboards, and coordinate approvals over email or Slack. Each handoff leaks context. Each additional app is another login, another notification, another reason to procrastinate.
Kilo’s bet is that the coaching relationship — the back-and-forth, the “I’m sore today, what should I do” moment — belongs in the same thread as your texts to your partner about dinner. That’s a distribution insight, not a feature. And it’s the same insight behind the quiet success of tools that push notifications into WhatsApp, Slack, or Discord instead of asking users to open yet another app.
For creators, the parallel is obvious. Your audience already lives in DMs. Instagram’s DM tab, TikTok’s inbox, X’s direct messages — these are where the highest-intent conversations happen. The creators who monetize well in 2025 aren’t the ones with the prettiest link-in-bio page; they’re the ones who treat DMs as a first-class channel. Kilo is a reminder that the interface with the least friction usually wins, even if it’s technically less capable than a dedicated app.
Where the math breaks
Here’s the part the launch page doesn’t address, and where I’d push back if I were advising Zac. Conversational interfaces are great for engagement and terrible for structured data entry. A training log needs consistent fields — sets, reps, weight, RPE, sleep, soreness. Typing that into a text thread every day is tedious, and LLM-based parsing of freeform text is good but not perfect. If Kilo misreads “felt heavy” as “went heavy,” the program adjusts wrong, and trust erodes fast.
The same problem shows up in social media tooling. I’ve tested AI caption generators that live inside chat interfaces, and the output is fine for one-off posts but breaks down when you need brand-consistent voice across 30 scheduled posts for five different clients. Conversational UX shines for ideation and quick edits. It struggles for batch operations, approval workflows, and anything that requires a spreadsheet view.
So my take: Kilo’s model works best for a specific user — the solo lifter who wants adaptive programming without spreadsheet discipline — and worse for the power user who already tracks everything in Hevy or Strong. The launch page doesn’t say which segment Kilo is optimizing for. That’s an open question, and it’s the one I’d want answered before recommending it to anyone.
What social media operators should steal from this launch
Three things, and none of them require you to care about fitness.
First, the “small group onboarding” pattern. Zac says he’s onboarding lifters in small groups for the free private beta. That’s not just capacity management — it’s a retention strategy. Small cohorts create social proof, surface bugs faster, and let the maker build personal relationships with early users. If you’re launching a creator product, a course, or a paid community, steal this. Don’t open the floodgates. Onboard in waves, and use each wave’s feedback to refine the next.
Second, the “coach with context” framing. Kilo’s differentiator isn’t that it’s an AI chatbot — it’s that the coach has your training history. Context is the moat. In social media, the equivalent is a tool that remembers your brand voice, your top-performing hooks, your audience’s timezone, and your past campaign results. Most AI content tools today have no memory. They regenerate from scratch every time, which is why the output feels generic. The winners in 2025 will be the tools that accumulate context and use it.
Third, the channel choice. iMessage, not a standalone app. For creators, the equivalent move is to meet your audience where they already are — not by building a new community platform, but by showing up consistently in the DMs, comments, and group chats they already use. I’ve seen creators 3x their email list by adding a simple “DM me the word GUIDE” call-to-action on Threads and Instagram Reels. The mechanic is unglamorous. It works.
Where my judgment says Kilo falls short — and where it might not
I haven’t used Kilo. It’s in private beta, and I’m not a marathon runner balancing a lifting program. So treat everything below as inference from the launch page plus my experience testing adjacent tools, not as a review.
The biggest risk is the one every conversational AI product faces: hallucination in a domain where wrong advice has physical consequences. A fitness coach that suggests a deload when you needed a PR week is annoying. A fitness coach that ignores a warning sign of injury is dangerous. The launch page doesn’t mention safety guardrails, medical disclaimers, or how Kilo handles pain reports. That’s a gap. Not disclosed.
The second risk is retention. Texting a coach is novel for two weeks. By week six, the novelty fades, and the user has to actively want to log their training. Kilo’s check-ins help, but check-ins are also easy to ignore — the same way push notifications from Hootsuite get swiped away. The product’s real test is whether the coaching relationship feels valuable enough to sustain daily engagement past the honeymoon.
The third risk is competition. Whoop, Oura, and Fitbit all have coaching features now, and they have hardware lock-in. Future has human coaches. Fitbod has algorithmic programming. Kilo’s edge is the interface and the conversational flexibility. Whether that’s enough to carve out a durable niche is genuinely unclear.
Why TikTok creators should care more than LinkedIn ones
This is where the platform dynamics matter, and it’s the part of the story most social media operators will miss if they only read the launch page.
TikTok’s algorithm rewards watch time and completion rate above almost everything else. That means the format that wins is the one that holds attention for the full 15 to 60 seconds — usually a hook, a payoff, and a reason to rewatch. A conversational coaching product is a natural fit for TikTok content because the “text your coach” format is inherently visual and inherently narrative. You can film the screen, react to the response, and build a story arc in under a minute. I’d bet Kilo’s best organic growth channel, if they lean into it, is TikTok screen recordings, not paid ads.
LinkedIn, by contrast, rewards dwell time and comment depth. The same product demo that crushes on TikTok would flop on LinkedIn because the audience there wants frameworks, not vibes. If Kilo’s team tries to run one content strategy across both platforms, they’ll waste budget. This is the mistake I see most founder-led brands make: they treat “social media” as one channel instead of eight, each with its own algorithm, format, and audience expectation. The operators who win are the ones who localize the message per platform — same core idea, different packaging.
The repurposing workflow Kilo’s launch accidentally demonstrates
One more thing worth stealing. The launch page is essentially a single piece of content — a maker’s story — that could be repurposed across every platform with minimal effort:
- Instagram and Threads: a carousel of “5 lessons from building a fitness coach in iMessage,” with the maker’s marathon story as the hook.
- TikTok and YouTube Shorts: a 45-second screen recording of the onboarding flow, with a voiceover about why iMessage was the right channel.
- X: a thread breaking down the distribution thesis — why conversational interfaces beat standalone apps for habit-forming products.
- LinkedIn: a longer post about the small-group beta onboarding strategy and what it teaches about community-led growth.
- Pinterest: an infographic comparing conversational coaching to traditional fitness apps, optimized for the “fitness planning” search vertical.
- YouTube: a 10-minute build-in-public vlog about the technical decisions behind Kilo, including how the agent handles context and check-ins.
That’s six platforms, one source asset, maybe three hours of production work. This is the repurposing workflow every creator should have in their back pocket, and it’s exactly what tools like Repurpose.io and Opus Clip are trying to automate. The automation helps, but the strategic mapping — which idea goes where, in what format — is still a human job. That’s your job.
What I’d watch / test next
If you’re a creator or social media operator reading this, here’s what I’d actually do this week.
First, audit your own tool stack for the “another app” problem. Count how many tabs you open to publish a single post. If it’s more than three, you have a fragmentation problem, and the fix isn’t another tool — it’s consolidating the workflow into one place, even if that place is less powerful.
Second, test a DM-first call-to-action on one platform. Pick the platform where your audience is most active, and add a “DM me [keyword]” prompt to your next three posts. Track the response rate with a UTM-tagged link or a simple spreadsheet. In my experience, this outperforms link-in-bio CTAs for warm audiences, but your mileage will vary — measure it.
Third, watch Kilo’s beta rollout. Not because you need a fitness coach, but because the product is a live experiment in conversational distribution and small-group onboarding. If it works, the pattern will spread to other verticals — including social media tooling. The first mover in “AI social media coach that lives in your DMs” will have a real advantage, and I’d bet we see that launch within 12 months.
Fourth, build a repurposing map for your own content. Take your best-performing long-form asset from the last quarter and sketch how it would look on each of the six platforms above. You don’t have to produce all six versions. But the exercise will show you where your current strategy has blind spots — usually Pinterest and YouTube, in my experience.
Finally, be skeptical of any tool that promises to replace your judgment. Kilo’s maker is honest that this is a beta, that onboarding is limited, and that the product is still finding its shape. That honesty is a trust signal, and it’s the standard I’d hold every AI tool to. If a launch page can’t tell you what the product doesn’t do, it’s not ready for your workflow.
