What works in marketing: a decade of tactics, honestly scored

Marketing advice has a reliability problem, and after a decade of running the tactics myself, I can tell you where it comes from.

The people selling a tactic lead with their own results and leave out their starting advantages: the capital they could burn, the audience they already had, the best-case students who arrived well funded. The tactic gets credit for the head start. Then thousands of practitioners try it from a standing start and quietly get a fraction of the promise.

So here is my scorecard, from ten years of running these tactics for clients and for myself. Every score comes with its context, because that is the part the sales page omits.

How I score, and how to read anyone's advice

My standing rule is simple. Take advice that is fairly well established, then go read how implementation is going for ordinary practitioners in the forums and groups. Score the tactic on the whole picture, never on the teacher's highlight reel.

Hold my scores to the same rule. They come from my contexts: expert firms, professional services, high-trust sales. Your context can move any score, and anyone whose scores never move with context is selling.

What held up

Publishing genuine expertise held up best of anything I have run. Years of publishing grew a client's site a hundredfold and won first-place national rankings, and the mechanism was never clever: quality content that people found and cited on their own. It compounds, it keeps working after you stop paying for it, and search engines and AI assistants now reward the firsthand version more than ever. (What survived the AI shift and what did not is its own story: SEO in the AI era.)

Email held up beside it. A newsletter list is the one audience no algorithm can take away from you, and every platform change of the last decade made that more valuable, never less.

And the oldest tactic of all held up: being genuinely good, visibly. Referrals run on it. Recommendations run on it. Every marketing asset I build for a client is a multiplier on it, and a multiplier on zero is zero.

What worked, but less than advertised

Traffic worked, and taught me its limits. That hundredfold site win came with a lesson the case studies skip: converting the traffic into appointments stayed hard the whole way. Visitors are not buyers, and rankings are not revenue. I score traffic as fuel, never as the engine.

Social platforms worked the same way. Owners who chased audience report followers, press, even magazine covers, and no engagements to show for them. An audience multiplies trust with people already headed toward you; it rarely manufactures buyers from strangers.

Cold outreach lands here too. Done with research and restraint, it opens real conversations, and it carries advantages nobody mentions. Done from a template at volume, it burns the sender's name. The score depends entirely on which version you mean.

What I score low

Hacks, as a category. Any tactic sold on a loophole dies the year everyone copies it, and the buyers of the course are usually the last ones in.

Volume content. Publishing thin material at scale was a weak play before AI summaries began answering thin questions for free, and now it is a dead one.

And paid ads for high-ticket expert services, scored low with a caveat: the math can work with money, patience, and a presence that already converts, and I have rarely watched it work without all three. Most expert firms reach for ads before the conversion layer exists, which is paying to send strangers somewhere that loses them.

What the scorecard will not tell you

Now the honest part, because you deserve the whole picture.

No scorecard replaces fit. The right tactic for your firm depends on your buyers, your strengths, and your calendar, which is why strategy has to fit the business like a glove rather than follow a list, including mine.

And every score here is provisional. Tactics decay, platforms turn, and the AI shift is repricing everything as I write. The scoring rule outlasts every score: established advice, checked against practitioners, tempered before you bet on it.

The pattern under the scores

Look back across the scorecard and one pattern holds it together. Everything that held up compounds, and everything I score low expires. Quality content compounds. An email list compounds. A reputation compounds. Hacks, volume, and rented attention expire on someone else's schedule.

So when the next tactic lands in your inbox, ask it one question before any other. Does this compound, or does it expire?