Twelve names. The same twelve for at least three years. Eleven US-listed, one Canadian. Mix of consumer staples, healthcare, industrials, financials, one tech. None I bought for the yield; all I bought for some combination of dividend growth and what I believed about the underlying business at the time of purchase.
This is not a brag piece. The yields are unremarkable, the portfolio is not large, and the total dividend income is not life-changing money. What is interesting is how the year of logging actually felt versus how the marketing of dividend investing describes it.
What I held and why
Without naming specific tickers — this site does not recommend individual securities — the rough mix:
- Three consumer-staples names (food, household products, beverages)
- Two healthcare (one large-cap pharma, one medical-devices)
- Two industrials (one diversified, one defense)
- Two financials (one large-bank, one insurance)
- One tech (mature, large-cap, not the obvious one)
- One utility
- One Canadian railway
Total portfolio cost basis: $58,400. Total market value at year-end: $74,210. Total 2025 dividends received: $2,108. Forward dividend yield on cost: 3.6%; on current market value, 2.84%.
Surprise one: it was lumpier than I expected
Marketing copy for dividend investing implies a steady monthly stream. Reality, with twelve US-listed names paying quarterly, is two months of light cash and one month of heavy cash, repeating. June, September, and December were each above $300; July, August, October, and November were each below $100. The Canadian name paid in February and August on a different schedule. The utility paid monthly, smoothing things slightly.
If I were optimizing for monthly cash flow — which I am not — I would care about this. For my actual use case, where the dividends reinvest automatically into a single account, the lumpiness is a non-event. But anyone planning to live off the dividends in retirement should be aware that the cash arrives in waves, not a stream.
Surprise two: growth beat yield, even short-term
The single highest-yielding name in my book — the utility, which started 2025 with a 4.8% yield — produced $268 of cash. Decent. But the position with the lowest yield at the start of the year, the mature tech name at 1.2%, produced $186 of cash and grew its dividend 11% mid-year. Compounded forward, the tech name's cash will pass the utility's within four years if both keep doing what they did last year.
The case for dividend growth investing — slow yield today, faster yield tomorrow — has been made in finance writing for decades. I had read it. Watching it happen at small scale in a portfolio I owned was more convincing than reading it.
Yield is a snapshot. Dividend growth is the movie. Most beginner portfolios over-index on the snapshot.
Surprise three: the tax bill
The $2,108 of dividends was held in a taxable brokerage. Roughly $1,950 of it was qualified, $158 was non-qualified (the Canadian railway, partially, and one bond-like distribution from the utility). At my marginal rate, the qualified portion was taxed at 15% — $293 — and the non-qualified at 24% — $38. Total federal tax: $331. State: $74.
Effective tax drag on the year's dividend income: about 19%. The dividends themselves grew the account by $2,108; the after-tax cash that actually ended up mine was $1,703. Not bad, but a 19% haircut. In a Roth IRA or HSA, all $2,108 would have remained in the account.
This is why, over the next two years, I am rotating as much of the dividend exposure as I can into Roth space and reducing it in taxable. Inside a Roth, the same portfolio is meaningfully more efficient.
What I would change
Three things I am acting on for 2026.
- Shift more of the dividend book into the Roth. Subject to contribution limits, but every dollar that moves from taxable to Roth saves 19% of its annual cash flow from tax indefinitely.
- Stop reinvesting in the taxable account, automatically. DRIP made sense when I was building. With twelve mature positions, automatic reinvestment can quietly grow concentration. I would rather take the cash quarterly and decide where it goes manually.
- Trim two names whose dividend growth has stalled. Both have not raised in two years. The market is signaling something; I have not, until now, been willing to act on the signal.
A year of logging cost me ten minutes a quarter. It changed two material decisions and one strategic one. Whatever you hold, write it down. The aggregate numbers your broker shows you are useful but smoothed; the per-position story is where the actual learning lives.
