diff --git a/.gitignore b/.gitignore
index a9fe1acf..b56b10b7 100644
--- a/.gitignore
+++ b/.gitignore
@@ -3,4 +3,4 @@ Manifest.toml
 *.scss
 *.css
 *style.jl
-docs/src/examples.md
\ No newline at end of file
+docs/src/examples/*.md
\ No newline at end of file
diff --git a/docs/make.jl b/docs/make.jl
index 292a48e6..a87ecf99 100644
--- a/docs/make.jl
+++ b/docs/make.jl
@@ -4,13 +4,20 @@ using TransitionsInTimeseries, Statistics, StatsBase
 
 using Literate
 
-Literate.markdown("src/examples.jl", "src"; credit = false)
+# process examples and add then in a sidebar column
+example_files = readdir(joinpath(@__DIR__, "src", "examples"))
+example_pages = String[]
+for file in example_files
+    mkdownname = splitext(file)[1]*".md"
+    Literate.markdown("src/examples/$(file)", "src/examples"; credit = false)
+    push!(example_pages, "examples/$(mkdownname)")
+end
 
 pages = [
     "index.md",
     "tutorial.md",
     "api.md",
-    "examples.md",
+    "Examples" => example_pages,
 ]
 
 import Downloads
diff --git a/docs/src/examples.jl b/docs/src/examples/logistic.jl
similarity index 74%
rename from docs/src/examples.jl
rename to docs/src/examples/logistic.jl
index 83b54cef..bb629cc1 100644
--- a/docs/src/examples.jl
+++ b/docs/src/examples/logistic.jl
@@ -1,6 +1,4 @@
-# # Examples for TransitionsInTimeseries.jl
-
-# ## Permutation entropy for dynamic regime changes
+# # Permutation entropy for dynamic regime changes
 
 # Permutation entropy is used frequently to detect a transition between
 # one dynamic regime to another. It is useful when the mean and std. of the
@@ -12,7 +10,7 @@
 # are arguably better suitable in such an application than the [Tutorial](@ref)'s default
 # of significance via random Fourier surrogates.
 
-# ### Logistic map timeseries
+# ## Logistic map timeseries
 
 # A simple example of this is transitions from periodic to weakly chaotic to chaotic
 # motion in the logistic map. First, let's generate a timeseries of the logistic map
@@ -20,20 +18,20 @@
 using DynamicalSystemsBase
 using CairoMakie
 
-logistic_rule(u, p, t) = @inbounds SVector(p[1]*u[1]*(1 - u[1]))
-ds = DeterministicIteratedMap(logistic_rule, [0.5], [1.0])
-
+## time-dependent logistic map, so that the `r` parameter increases with time
 r1 = 3.83
 r2 = 3.86
 N = 2000
 rs = range(r1, r2; length = N)
-x = zeros(N)
-for (i, r) in enumerate(rs)
-    set_parameter!(ds, 1, r)
-    step!(ds)
-    x[i] = current_state(ds)[1]
+
+function logistic_drifting_rule(u, rs, n)
+    r = rs[n+1] # time is `n`, starting from 0
+    return SVector(r*u[1]*(1 - u[1]))
 end
 
+ds = DeterministicIteratedMap(logistic_drifting_rule, [0.5], rs)
+x = trajectory(ds, N-1)[1][:, 1]
+
 # Plot it, using as time the parameter value (they coincide)
 fig, ax = lines(rs, x; linewidth = 0.5)
 ax.xlabel = "r (time)"
@@ -45,7 +43,7 @@ fig
 # to weak chaos at r ≈ 3.847. This transition is barely visible in the
 # timeseries, and in fact many of the timeseries statistical properties remain identical.
 
-# ### Using a simpler change metric
+# ## Using a simpler change metric
 
 # Now, let's compute and various indicators and their changes,
 # focusing on the fourth indicator, the permutation entropy. We use
@@ -102,7 +100,7 @@ fig = plot_change_metrics()
 # Due to its construction, permutation entropy will have a spike for periodic data at the
 # start of the timeseries, so we can safely ignore the spike at r ≈ 3.83.
 
-# ### Significance via random Fourier surrogates
+# ## Significance via random Fourier surrogates
 
 # One way to test for significance would be via the standard way as in the [Tutorial](@ref),
 # utilizing surrogate timeseries and [`SurrogatesSignificance`](@ref).
@@ -117,10 +115,11 @@ fig = plot_change_metrics()
 surromethod = RandomFourier()
 
 ## Define a function because we will re-use later
+using Random: Xoshiro
 function overplot_surrogate_significance!(fig, surromethod, color = "black")
 
     signif = SurrogatesSignificance(;
-        n = 1000, tail = [:both, :both, :right], surromethod
+        n = 2000, tail = [:both, :both, :right], surromethod, rng = Xoshiro(42),
     )
     flags = significant_transitions(results, signif)
 
@@ -148,39 +147,44 @@ overplot_surrogate_significance!(fig, surromethod)
 
 fig
 
-# ### More appropriate surrogates
+# ## Different surrogates
 # %% #src
-# Using random Fourier surrogates does not make much sense in our application.
-# Those surrogates perserve the power spectrum of the timeseries, but the power spectrum
+# Random Fourier surrogates perserve the power spectrum of the timeseries, but the power spectrum
 # is a property integrated over the whole timeseries. It doesn't contain any information
-# regarding a sharp transition at some point in the timeseries.
+# highlighting the _local_ dynamics or information that preserves the local
+# changes of dynamical behavior.
+
+# A surrogate type that does a better job in preserving local sharp
+# changes in the timeseries (and hence provides **stricter**
+# surrogate-based significance) is for example `RelativePartialRandomization`.
+
 # A much better alternative is to use block-shuffled surrogates, which
 # preserve the short term local temporal correlation in the timeseries and hence
 # also preserve local short term sharp changes in the dynamic behavior.
 
-surromethod = BlockShuffle(15)
+surromethod = RelativePartialRandomization(0.25)
 fig = plot_change_metrics()
-overplot_surrogate_significance!(fig, surromethod, "red")
+overplot_surrogate_significance!(fig, surromethod, "gray")
 fig
 
-# The results are better for the variance and AR1 indicators.
-# For the permutation entropy the results do not change because it already is an
-# exceptionally well
-# suited indicator for this application scenario. But in other cases where things
-# are not as clear, or data are contaminated with noise, or we have shorter data,
-# choosing a more suitable
-# surrogate generator may make the difference between a false positive or not.
+# Our results have improved. In the permutation entropy, we see only two
+# transitions detected as significant, which is correct: only two
+# real dynamical transitions exist in the data.
+# In the other two indicators we also see fewer transitions, but as we
+# have already discussed, no results with the other indicators should
+# be taken into meaningful consideration, as these indicators
+# are simply inappropriate for what we are looking for here.
 
-# ### Simpler Significance
+# ## Simpler Significance
 # %% #src
 # Arguably, exactly because we are using the [`difference_of_means`](@ref) as a
-# change metric, we may want to be much less strict with our tests for significance.
+# change metric, we may want to be less strict and more simple with our tests for significance.
 # Instead of using [`SurrogatesSignificance`](@ref) we may use the simpler and much faster
-# [`QuantileSignificance`](@ref), which simply claims significant time points
-# whenever a change metric exceeds some pre-defined quantile of its timeseries.
+# [`SigmaSignificance`](@ref), which simply claims significant time points
+# whenever a change metric exceeds some pre-defined factor of its timeseries standard deviation.
 
 fig = plot_change_metrics()
-flags = significant_transitions(results, QuantileSignificance())
+flags = significant_transitions(results, SigmaSignificance(factor = 5.0))
 
 ## Plot the flags
 for (i, indicator) in enumerate(indicators)
@@ -188,5 +192,5 @@ for (i, indicator) in enumerate(indicators)
         color = Cycled(3), linestyle = :dash, linewidth = 3
     )
 end
-content(fig[1, 1]).title = "significance from quantile"
+content(fig[1, 1]).title = "significance from std"
 fig
diff --git a/src/TransitionsInTimeseries.jl b/src/TransitionsInTimeseries.jl
index 3b7ad7a9..ac2a4afe 100644
--- a/src/TransitionsInTimeseries.jl
+++ b/src/TransitionsInTimeseries.jl
@@ -54,7 +54,7 @@ export difference_of_means
 # analysis
 export WindowedIndicatorConfig, estimate_indicator_changes, WindowedIndicatorResults
 export TransitionsSignificance, significant_transitions
-export QuantileSignificance, SurrogatesSignificance
+export QuantileSignificance, SigmaSignificance, SurrogatesSignificance
 
 # timeseries
 export isequispaced, equispaced_step
diff --git a/src/analysis/quantile_significance.jl b/src/analysis/quantile_significance.jl
index 0d7c927c..c571f343 100644
--- a/src/analysis/quantile_significance.jl
+++ b/src/analysis/quantile_significance.jl
@@ -1,17 +1,21 @@
 """
-    QuantileSignificance(; p = 0.95, tail = :right) <: TransitionsSignificance
+    QuantileSignificance(; p = 0.95, tail = :both) <: TransitionsSignificance
 
 A configuration struct for significance testing [`significant_transitions`](@ref).
-When used with [`WindowedIndicatorResults`](@ref), significance is estimated as
+When used with [`WindowedIndicatorResults`](@ref), significance is estimated
 by comparing the value of each change metric with its `p`-quantile.
 Values that exceed the `p`-quantile (if `tail = :right`)
 or subseed the `1-p`-quantile (if `tail = :left`)
 are deemed significant.
 If `tail = :both` then either condition is checked.
+
+`QuantileSignficance` guarantees that some values will be significant by
+the very definition of what a quantile is.
+See also [`SigmaSignificance`](@ref) that is similar but does not have this guarantee.
 """
 Base.@kwdef struct QuantileSignificance{P<:Real}
     p::P = 0.95
-    tail::Symbol = :right
+    tail::Symbol = :both
 end
 
 using Statistics: quantile
@@ -34,3 +38,44 @@ function significant_transitions(res::WindowedIndicatorResults, signif::Quantile
     return flags
 end
 
+"""
+    SigmaSignificance(; factor = 3.0, tail = :both) <: TransitionsSignificance
+
+A configuration struct for significance testing [`significant_transitions`](@ref).
+When used with [`WindowedIndicatorResults`](@ref), significance is estimated
+by comparing how many standard deviations (`σ`) the value exceeds the mean value (`μ`).
+Values that exceed (if `tail = :right`) `μ + factor*σ`, or subseed (if `tail = :left`) `μ - factor*σ`
+are deemed significant.
+If `tail = :both` then either condition is checked.
+
+`factor` can also be a vector of values,
+in which case a different value is used for each change metric.
+
+See also [`QuantileSignificance`](@ref).
+"""
+Base.@kwdef struct SigmaSignificance{P}
+    factor::P = 0.95
+    tail::Symbol = :both
+end
+
+using Statistics: std, mean
+
+function significant_transitions(res::WindowedIndicatorResults, signif::SigmaSignificance)
+    flags = similar(res.x_change, Bool)
+    for (i, x) in enumerate(eachcol(res.x_change))
+        μ = mean(x)
+        σ = std(x; mean = μ)
+        factor = signif.factor isa AbstractVector ? signif.factor[i] : signif.factor
+        flag = view(flags, :, i)
+        if signif.tail == :right
+            @. flag = x > μ + factor*σ
+        elseif signif.tail == :left
+            @. flag = x < μ - factor*σ
+        elseif signif.tail == :both
+            @. flag = (x < μ - factor*σ) | (x > μ + factor*σ)
+        else
+            error("`tail` can be only `:left, :right, :both`. Got $(tail).")
+        end
+    end
+    return flags
+end