Why Isn't My Sourdough Starter Rising? A Troubleshooting Guide

The Day Four Graveyard

More sourdough starters are thrown away on day four than on any other day, and almost none of them deserved it.

The story is always the same. You mix flour and water. Day one, nothing. Day two, it erupts — doubles, triples, foams to the rim, and you think this is easy. Day three it's still going. Day four, it's flat. Day five, flatter. By day six there's a sad grey puddle in the jar that smells faintly of feet, and you conclude that you have killed it, that your kitchen is wrong, that some people just can't do this.

You have killed nothing. What you watched on day two was never your bread yeast in the first place, and what you're watching now is the handover. It is the single most predictable event in the whole process, and it has been measured.

This page works through every diagnosable reason a starter isn't rising, in rough order of how often it's the actual culprit. We researched it against university extension services and published starter research rather than baking forums, because sourdough advice online is unusually confident and unusually unsourced — and at least one of the most-repeated rules turns out not to hold up at all.

Cause 1: It's Too Cold — By a Wide Margin the Most Common

If you read nothing else, read this section. Temperature is not a minor variable in fermentation. It is the throttle on the entire process, and a kitchen that feels merely "a bit cool" to you can halve the speed of everything happening in that jar.

Colorado State University Extension puts the target in human terms: the fermenting microorganisms are more viable at temperatures that feel comfortable for you, a warm room temperature of around 70°F, and fermentation will slow at colder temperatures — or run too rapidly, or stop, when too hot.

The research puts a number on why. In the classic modelling study of sourdough organisms, Gänzle, Ehmann and Hammes found that the sourdough yeast Candida milleri has an optimum growth temperature of 27°C — about 81°F — and that the sourdough lactobacilli optimum sits higher still. Read that against a typical autumn flat at 18°C (64°F) and the picture is obvious: your starter isn't broken, it is running a long way below the temperature at which its organisms do their best work.

This is why the same recipe "works for everyone else". Everyone else's kitchen is warmer than yours.

What to do about it

• Find the warm spot and use it consistently. The top of the fridge, a high shelf, an airing cupboard, near (never on) a radiator, or inside the oven with only the light on. Check it with a thermometer once rather than guessing — kitchens are colder than they feel, especially at floor and counter height overnight.
• Use warm water, not cold, at every feed. CSU Extension's starter care guidance specifies warm or room temperature water. Straight from a cold tap in winter you may be dropping the whole culture several degrees at the exact moment you're asking it to get going.
• Give it a gentle heat source if your home is genuinely cold. A low-wattage fermentation or seedling mat under a folded towel holds a jar in the mid-70s without cooking it. We cover the same hardware for tea cultures in our kombucha heating mat guide — a starter jar has exactly the same requirement, and the mats are interchangeable.
• Don't chase heat. Warmer is faster only up to a point, and an oven that clicks on to 100°F+ will cook a starter outright. Steady beats hot.
• Adjust your expectations, not just your thermostat. A cold-kitchen starter that peaks in ten hours instead of five isn't failing. Feed it on its clock — when it peaks and begins to fall — rather than on the clock a recipe written in a warmer climate gave you.

Cause 2: The Day 2 False Rise — And the Death That Follows It Isn't Death

This is the one that costs the most starters, because the disappointment is so sharp. Something worked, and then it stopped working, and that feels categorically worse than nothing happening at all.

Here is what actually happened. A 2023 study in PeerJ tracked the microbial communities and the physical rise of new starters across several flour types, day by day. It verified the pattern bakers describe informally — a dramatic increase in height on day 2 or 3, followed by a plateau or decrease in activity until roughly day 7 — and it identified who was responsible.

The organisms dominating those first three days are not bread yeast and not the lactobacilli you're trying to cultivate. They are opportunists that arrived on the flour: the study found young starters of three days or less dominated by Enterobacteriaceae, with Pantoea at around 24% mean abundance and unidentified Enterobacteriaceae at around 27%, alongside Klebsiella, Pseudomonas and Erwinia. Leuconostoc and Weissella follow as an intermediate wave, more abundant after day 1 but before day 10, and neither dominates a mature starter.

Those early bacteria produce carbon dioxide enthusiastically. That's your day 2 rise. But they have no acid tolerance, and they are busy building the exact environment that will finish them: the study recorded pH dropping on day 2, continuing to fall on day 3, and eventually settling around pH 3.6. As acidity climbs, the opportunists collapse. The jar goes quiet.

And then, over the following few days, the organisms you actually wanted — acid-tolerant Lactobacillus, Pediococcus, and wild yeasts — move into an environment now sour enough to exclude their competition and slowly take it over. Mature starters at days 10 to 14 in the study were dominated by Lactobacillus at around 48% mean abundance.

The flat spell isn't a failure state. It is the changeover, and it is load-bearing. A starter that never went through it would be a starter that never acidified — and that would genuinely be a problem.

How to tell a false rise from the real thing

• Timing. Day 2 to 3 from a standing start is too early for an established yeast population. Day 7 onwards is credible.
• Smell. A false rise often smells sweetish, vegetal, cabbagey or plain odd. The real thing smells sour, tangy, yeasty and faintly alcoholic.
• Texture. The early bacterial foam tends to be loose, runny and large-bubbled, and it deflates completely the moment you stir it. A developed starter is thick and batter-like, webbed with fine even bubbles, and holds its structure.
• Repeatability. This is the real test. The false rise happens once and never returns. A working starter does the same thing after every feed, on a predictable clock.

So: do not discard, do not start over, and do not panic-feed it four times a day. Hold your schedule through the quiet stretch. The jar is busier during the slump than it looks.

Cause 3: You Haven't Given It Enough Time

Sourdough content online is built around a promise of speed, and that promise is the source of an enormous amount of unnecessary failure. If your reference point is "bubbly starter in 5 days, bread by the weekend," you will conclude you have failed somewhere around day six of a completely normal process.

The published timelines are considerably longer and considerably less dramatic.

• Illinois Extension states that it can take 14 days to a month to establish a starter, fed daily, and twice daily starting on day 4.
• The PeerJ study concluded that starters were functionally mature and stable — that is, suitable for baking — by day 10 across all flour types tested, supporting the traditional bakers' "10 day rule".
• CSU Extension simply says that creating a starter will take several days of regular feeding, and defines readiness by behaviour rather than by date: it bubbles and rises, and develops a pleasantly sour smell.

Note what all three have in common: readiness is a behaviour, not a date. Nobody credible says "ready on day 7." They say ready when it does the thing, reliably, more than once.

And if you are still in week one: nothing is wrong. Two weeks is normal. A month is not unheard of, particularly in a cold flat or with a low-extraction white flour. The people whose starters came alive in four days mostly had warm kitchens.

Cause 4: Flour and Water — Where One Rule Holds and One Doesn't

Flour: wholemeal and rye genuinely do kickstart a starter

This one is real, it has a clear mechanism, and it is the cheapest fix on this page.

White flour is, from a microbe's point of view, a stripped-down food. The bran and germ — the parts milled out of it — are where the flour's own microbial population lives, along with most of its minerals and enzymes. Whole grain flour brings all of that to the jar. As King Arthur Baking's bakers put it, whole grains contain exponentially more sourdough-friendly microorganisms than white flour, and the minerals and other food present in whole grains give the culture a big feed that encourages growth.

Rye is the strongest of the lot. It carries more free sugars that are immediately accessible to the yeast and bacteria, and a higher content of starch-degrading enzymes — which between them speed fermentation up measurably. Illinois Extension's own starter guidance reflects this in practice, reporting better results from building a starter with at least 50% rye alongside all-purpose or bread flour, then moving to all all-purpose once the starter is ready and active.

The PeerJ study found something consistent with that too: rye starters retained an unusually high diversity of bacterial taxa even at maturity, which the authors attribute to rye's easily degraded starches and whole-grain nutrients opening up more ecological niches.

Browse whole rye flour for starters on Amazon →

Two smaller flour points. Bleached flour is worth avoiding for starter-building — extension starter guidance consistently specifies unbleached. And flour that has sat open for a year in a warm cupboard is a weaker inoculum than a fresh bag; if everything else has failed, a new bag of flour is a reasonable thing to try.

Water: the chlorine rule mostly doesn't hold up

Now the folklore. "Never use tap water, the chlorine kills your starter" is possibly the most repeated instruction in beginner sourdough. We went looking for the evidence behind it, and it is much weaker than its confidence suggests.

King Arthur Baking lists it outright as a myth. Its position is that the effect of chemicals like chlorine is minimal, and that many bakers — including those in its own test kitchen — simply use tap water. Unless the water from your faucet is treated heavily enough that you can actually smell the chemicals, there's no need for bottled water.

Oregon State University Extension, answering through Ask Extension, lands in the same place with a usefully practical test: chloramine levels in domestic water fluctuate and the dose matters, and if you cannot smell or taste the "chlorine" in your tap water, it is likely to work just fine in your starter.

For scale: the US EPA sets a maximum residual disinfectant level of 4.0 mg/L for both chlorine and chloramine in public drinking water. That is a concentration set to be safe to drink continuously — it is not a sterilising dose, and a starter is a dense, buffered slurry of flour, not a thin solution.

Our read: the chlorinated-water rule is not baseless, but it has been inflated far beyond what the sources support. Heavily treated water that you can smell is worth working around, especially in the first fragile week. Ordinary municipal tap water that tastes of nothing is very unlikely to be why your starter isn't rising — and we'd rather tell you that than let you spend a fortnight buying bottled water while the actual problem, a 64°F kitchen, goes unaddressed.

Cause 5: Feeding Ratio, Hydration and Consistency

Once temperature and time are ruled out, feeding is the next place to look — and the failure is almost always inconsistency rather than any particular ratio being wrong.

Under-feeding

The commonest version. A starter kept in a warm room and fed once every 24 hours will frequently exhaust its flour in eight or ten hours, then sit for fourteen more in a sour, alcoholic, increasingly hostile environment of its own making. It's not dead when you next look at it — it's starving, and its rise is getting weaker each cycle because you keep catching it after the crash.

The tells: hooch forming daily, a sharp acetone or nail-polish smell, a thin and watery texture, and a rise that peaks and collapses long before you come back to it. The fix is to feed more often, feed a bigger proportion of fresh flour, or move the jar somewhere cooler to slow the clock down.

Over-feeding

The rarer and more confusing one, because it looks like nothing is happening. If you feed a small quantity of starter into a large quantity of fresh flour and water, you have diluted the culture so heavily that it takes a long time to build a population big enough to visibly lift the mixture. Nothing is wrong — you just can't see it. If you are feeding a young starter at a high ratio because a recipe told you to, pull the ratio back while it establishes.

Ratios, without the mystique

A feeding ratio is written starter : flour : water. 1:1:1 means equal weights of each and is a good default for a young or sluggish starter — it keeps the population concentrated and turns over fast. 1:5:5 or wider suits a mature, vigorous starter that would otherwise peak and crash before you got back to it. If your starter is struggling, tighten the ratio; if it's crashing too fast, widen it.

Extension feeding guidance is deliberately unfussy about the exact numbers — CSU Extension's maintenance routine, for instance, works in cups rather than precise ratios and simply asks that you feed and use the starter weekly, warming it at room temperature for an hour or two after taking it out of the fridge. The point is that no single ratio is correct. What matters is that you use the same one long enough to read the result.

Hydration

Hydration is water as a percentage of flour weight. Most starters run at 100% hydration — equal weights of flour and water — which gives a thick batter. It matters here for one reason that catches people out: a stiffer starter rises more slowly but holds its shape, and a wetter one rises faster but collapses sooner and more dramatically. If you've drifted towards a loose, soupy starter by eyeballing your water, you may be getting a real rise that falls back before you ever see it. Weigh flour and water on a scale rather than measuring by volume, and the variable disappears.

Our sourdough starter kit guide covers the practical side of this — scales, jars, the feeding rhythm and what a maintenance routine actually looks like once the starter is established.

Cause 6: You Can't See the Rise (The Jar Problem)

This one is worth taking seriously, because it produces a fake problem that no amount of feeding will fix: a starter that is rising perfectly well, in a vessel that makes it impossible to notice.

The geometry is simple. Doubling the volume of a batter in a tall, narrow jar produces a dramatic, obvious climb up the glass. Doubling exactly the same volume in a wide mixing bowl, a large crock or a broad-based jar raises the surface by a centimetre or two, which you will absolutely fail to register — especially at day nine when you have already half-decided it's dead.

Worse, people tend to use large vessels for small amounts of starter, on the reasonable-sounding theory of leaving room for it to grow. That's exactly backwards while you're trying to observe it.

How to actually measure activity

• Use a straight-sided, reasonably narrow jar, and keep the quantity of starter modest relative to the jar. You want a tall column, not a shallow pool.
• Mark the level immediately after every feed. A rubber band round the glass, a dry-erase mark, or a strip of tape. This is the single most useful habit in starter troubleshooting, and it costs nothing. Without a mark you are comparing the jar in front of you against your memory of the jar, which is worthless.
• Note the time, not just the height. What you're looking for is peak height and how long it took. "Doubled in 6 hours" is a data point. "It looks a bit bubbly" is not.
• Watch for the fall. The peak is the moment to use it or feed it. If you only ever check it once every 24 hours, you may be consistently arriving after the peak and concluding nothing happened.

Browse straight-sided starter jars with marker bands on Amazon →

The float test, and why we don't lean on it

The float test — drop a spoonful of starter in water, and if it floats it's ready — is the standard beginner readiness check. It is also, on the evidence, not reliable enough to trust.

King Arthur Baking tested it directly and found false positives in both directions of the process: yeast dough that had risen for only 30 minutes floated, and starter only four hours into a fermentation with several hours still to go floated too. Its conclusion is that both partially risen dough and growing-but-not-yet-ripe starter will float, which makes a float a very weak signal. Its published verdict is that the float test is simply too unreliable a method for determining whether a starter is healthy and active enough to use.

The asymmetry is the useful part. A sink is informative — it tells you the starter definitely isn't ready. A float tells you very little. A stiff young starter can trap enough air to float while being nowhere near strong enough to raise a loaf.

Use the marked band and the clock instead. Very bubbly, reliably doubling within 6 to 8 hours of a feed, three feeds running — that is a real readiness test, and it is the one worth building the habit around. Once you get there, our bread proofing basket guide covers the next stage, and the Dutch oven guide covers the bake itself, where a genuinely active starter finally shows what it can do.

Cause 7: Hooch — The Dark Liquid

A layer of thin, dark liquid on top of a starter alarms almost everyone the first time, and it is one of the easiest signs on this page to read correctly.

It's called hooch, and it means hungry. Colorado State University Extension describes it plainly as a by-product of the fermenting yeast, which can either be poured off or stirred in. It is largely alcohol and acids produced once the culture ran out of accessible food and kept metabolising anyway.

Its colour ranges from clear through straw to grey to almost black, and the colour of the liquid layer itself is not a spoilage signal — darker usually just means longer since the last feed. What matters is whether there is growth on the surface, which is a different question entirely and covered below.

What to do

• Pour it off for a milder, less acidic starter, or stir it in for a tangier one. Both are explicitly fine. Pouring off is generally the better choice if your starter has been neglected, since it removes accumulated acid and alcohol that are suppressing the culture.
• Then feed it — hooch is a request, not a state to leave it in.
• Treat repeated hooch as a schedule problem. If it appears every single day, your starter is outrunning your feeding. Feed more often, feed a larger proportion of fresh flour, or move it somewhere cooler.

Persistent hooch plus a sharp acetone or nail-varnish smell is the classic signature of chronic under-feeding. It is completely recoverable, and it is not the same thing as spoilage.

Cause 8: Genuinely Dead, or Just Sluggish? (And How to Revive It)

Start from the correct prior: truly dead starters are rare. A sourdough culture is a robust, self-defending, acid-loving ecosystem that survives conditions that would finish most things in your kitchen. Neglect makes it weak far more often than it makes it dead.

The short list of genuine write-offs

• Fuzzy, raised or coloured mould on the surface. CSU Extension's line on this is unambiguous: any sign of mould, coloured and/or fuzzy, means it should not be used.
• A genuinely putrid smell — rotten, cheesy, faecal, rancid. Not sour, not boozy, not sharp: those are all normal. Wrong-family smells mean wrong-family organisms.
• Anything pink, orange or black growing through the starter rather than sitting in the liquid layer.

That's the list. Everything else — flat, thin, smelly, hooch-covered, forgotten for six months behind the milk — falls under "sluggish", and sluggish is a recoverable condition.

The revival routine

1. Pour off all the hooch and any discoloured top layer.
2. Discard almost everything. Keep a tablespoon or two from the middle of the jar and bin the rest. This is counterintuitive and it is the most important step: you are diluting out accumulated acid and alcohol so the surviving organisms get a clean environment. Trying to rescue the whole jar is what keeps a starter suppressed.
3. Move to a clean jar. A crust of dried old starter round the rim is exactly where mould gets a foothold.
4. Feed at 1:1:1 with warm water and flour that includes a meaningful proportion of wholemeal or rye, for the reasons in the flour section above.
5. Keep it warm — around 70°F, consistently. This is where cold-kitchen revivals fail.
6. Feed every 12 hours and mark the jar every time.
7. Give it a full week before you make any judgement. Neglected starters often show nothing for three or four days and then come back strongly.

King Arthur Baking's observation is worth holding onto through this: mature starters maintained on unbleached all-purpose flour generally revive well even after extended neglect in the fridge. The whole grain helps, but it isn't a requirement — patience and a tight feeding schedule do most of the work.

Cause 9: Smell and Colour — Normal Weirdness vs Genuine Spoilage

Starters smell strange. That is not a defect, it's the point — you are cultivating an acid-producing microbial community in an open jar on your counter, and it is going to announce itself.

Normal, in all its variety

• Sour, tangy, yoghurty. The baseline. This is lactic acid and it is the target.
• Sharp, vinegary. Acetic acid, favoured by cooler temperatures and longer gaps between feeds. Normal, and often desirable for flavour.
• Boozy, beery, acetone or nail polish. Alcohol from a hungry starter. A feeding signal, not a spoilage signal — feed it more often and it settles.
• Sweetish, cabbagey or oddly vegetal in the first few days. The early bacterial phase described above. It passes as acidity rises.
• Yeasty and bready. A well-fed, well-established starter near its peak. This is what you're aiming for.
• Grey or grey-brown liquid on top. Hooch. Colour in the liquid layer is normal.

The line that means stop

The rule is simple and it is the same one we apply across every ferment on this site: fuzzy, raised or coloured growth means discard. Flat discolouration in the liquid is not the same thing as something growing up off the surface. Height and fuzziness are the tells, not colour — and you judge them by looking at the jar from the side, at eye level, rather than from above.

We won't restate the whole framework here, because we've already written it properly. Our guide to telling good fermentation from spoilage works through the full kahm-yeast-versus-mould distinction, why you discard the entire container rather than scraping the visible patch, and how the same logic applies across vegetable ferments, kombucha and starters — all traced to extension services, the FDA and the CDC. If you're standing over a jar right now trying to decide, that's the page to read.

Two starter-specific notes that page covers and that are worth repeating here. CSU Extension notes that refrigerated starters can develop whitish clumps on the surface of the liquid layer that are safe yeasts, not mould — flat, soft, in the liquid. And the disposal rule is the same as everywhere else in preserving: if it's mould, the whole jar goes, the container gets washed in hot soapy water, and you begin again. Starters are cheap to rebuild. That's the same standard we hold in our canning starter kit guide, and it doesn't relax just because a starter took you two weeks to grow.

The Diagnostic Table

The fast lookup. Find the row that matches what's in your jar.

Frequently Asked Questions

Sources

Every technical claim on this page is drawn from the following university extension publications, peer-reviewed research and recognised baking-science references. We researched and compared these sources; we do not bake, taste or test products ourselves.

1. Colorado State University Extension — Sourdough Starter Best Practices (a warm room temperature of around 70°F; fermentation slows at colder temperatures and may stop when too hot; creating a starter takes several days of regular feeding; readiness judged by bubbling, rising and a pleasantly sour smell; hooch is a by-product of the fermenting yeast that can be poured off or stirred in; whitish clumps on the liquid layer are safe yeasts, not mould; any sign of coloured or fuzzy mould means it should not be used)
2. Colorado State University Extension — Sourdough Basics: Caring for Your Starter and Making Bread (feed and use the starter weekly; warm at room temperature for 1–2 hours after refrigeration; warm or room temperature water; unbleached high-protein flour; readiness when it has doubled and then collapsed back)
3. University of Illinois Extension — Sourdough Bread: 101 (it can take 14 days to a month to establish a starter, fed daily and twice daily starting on day 4)
4. University of Illinois Extension — Sourdough Bread: From Starter to Loaf — Webinar FAQs (better results building a starter with at least 50% rye alongside all-purpose or bread flour, moving to all-purpose once active)
5. Landis et al., PeerJ 11:e16163 (2023) — Sourdough starters exhibit similar succession patterns but develop flour-specific climax communities (verified dramatic increase in height on day 2 or 3 followed by a plateau or decrease until roughly day 7; young starters of 3 days or less dominated by Enterobacteriaceae, Pantoea, Klebsiella, Pseudomonas and Erwinia; Leuconostoc and Weissella higher after day 1 but before day 10; mature day 10–14 communities dominated by Lactobacillus, Pediococcus and Gluconobacter; pH falling from day 2 and stabilising near 3.6; starters functionally mature and suitable for baking by day 10; rye retaining higher bacterial diversity at maturity)
6. Oregon State University Extension, via Ask Extension — Chloramine in water and yeast breads (chloramine levels fluctuate and dose matters; if you cannot smell or taste the chlorine it is likely to work fine in your starter; chloramine does not off-gas by standing overnight; bottled water for the first week, or carbon filtration, reverse osmosis or extended boiling, if treatment is genuinely needed)
7. Gänzle, Ehmann & Hammes, Applied and Environmental Microbiology 64(7):2616–2623 (1998) — Modeling of Growth of Lactobacillus sanfranciscensis and Candida milleri in Response to Process Parameters of Sourdough Fermentation (Candida milleri optimum growth temperature of 27°C, lower than that of the sourdough lactobacilli; growth across a pH range of 3.9 to 6.7)
8. King Arthur Baking — Debunking the biggest sourdough myths (filtered or distilled water is a myth given the minimal effect of chemicals like chlorine, and test kitchen bakers simply use tap water; the float test is too unreliable a method; twice-daily feeding is not mandatory for a maintained starter)
9. King Arthur Baking — The float test for yeast dough and sourdough starter (both partially risen yeast dough and growing but not yet ripe starter will float; dough risen only 30 minutes and starter only 4 hours in both floated; use very bubbly and reliably doubling within 6 to 8 hours instead)
10. King Arthur Baking — Sluggish starter? Add a little whole grain flour (whole grains contain exponentially more sourdough-friendly microorganisms than white flour; minerals and other food in whole grains encourage growth; rye contains more free sugars and more starch-degrading enzymes, speeding fermentation; 10% to 20% whole grain suggested; neglected mature starters generally revive well on all-purpose flour)
11. US Environmental Protection Agency — National Primary Drinking Water Regulations (maximum residual disinfectant level of 4.0 mg/L for both chlorine and chloramines in public drinking water)

Disclaimer: This is general information compiled from published university extension, peer-reviewed and baking-industry sources. It is not food-safety certification and not a substitute for guidance from your county extension office. We hold no food-safety qualification. For any question about whether a ferment is safe to eat, follow the guidance of your local extension service over anything written here.