diff --git a/qiskit_experiments/base_experiment.py b/qiskit_experiments/base_experiment.py
index 2afd94abbf..a045f334f9 100644
--- a/qiskit_experiments/base_experiment.py
+++ b/qiskit_experiments/base_experiment.py
@@ -126,6 +126,7 @@ def run(
 
         # Generate and transpile circuits
         circuits = transpile(self.circuits(backend), backend, **self.transpile_options.__dict__)
+        self._postprocess_transpiled_circuits(circuits, backend, **run_options)
 
         if isinstance(backend, LegacyBackend):
             qobj = assemble(circuits, backend=backend, **run_opts)
@@ -308,6 +309,10 @@ def set_analysis_options(self, **fields):
         """
         self._analysis_options.update_options(**fields)
 
+    def _postprocess_transpiled_circuits(self, circuits, backend, **run_options):
+        """Additional post-processing of transpiled circuits before running on backend"""
+        pass
+
     def _metadata(self) -> Dict[str, any]:
         """Return experiment metadata for ExperimentData.
 
diff --git a/qiskit_experiments/randomized_benchmarking/__init__.py b/qiskit_experiments/randomized_benchmarking/__init__.py
index 171dab52db..15e1c6c02f 100644
--- a/qiskit_experiments/randomized_benchmarking/__init__.py
+++ b/qiskit_experiments/randomized_benchmarking/__init__.py
@@ -39,3 +39,4 @@
 from .interleaved_rb_experiment import InterleavedRBExperiment
 from .rb_analysis import RBAnalysis
 from .interleaved_rb_analysis import InterleavedRBAnalysis
+from .rb_utils import RBUtils
diff --git a/qiskit_experiments/randomized_benchmarking/rb_analysis.py b/qiskit_experiments/randomized_benchmarking/rb_analysis.py
index f42f7359da..2c6ed3e18a 100644
--- a/qiskit_experiments/randomized_benchmarking/rb_analysis.py
+++ b/qiskit_experiments/randomized_benchmarking/rb_analysis.py
@@ -26,7 +26,9 @@
     get_opt_value,
     get_opt_error,
 )
+
 from qiskit_experiments.analysis.data_processing import multi_mean_xy_data
+from .rb_utils import RBUtils
 
 
 class RBAnalysis(CurveAnalysis):
@@ -85,6 +87,9 @@ def _default_options(cls):
         default_options.xlabel = "Clifford Length"
         default_options.ylabel = "P(0)"
         default_options.fit_reports = {"alpha": "\u03B1", "EPC": "EPC"}
+        default_options.error_dict = None
+        default_options.epg_1_qubit = None
+        default_options.gate_error_ratio = None
 
         return default_options
 
@@ -148,6 +153,16 @@ def _format_data(self, data: CurveData) -> CurveData:
             data_index=mean_data_index,
         )
 
+    def _run_analysis(self, experiment_data, **options):
+        """Run analysis on circuit data."""
+        error_dict = options["error_dict"]
+        qubits = experiment_data.metadata()["physical_qubits"]
+        if not error_dict:
+            options["error_dict"] = RBUtils.get_error_dict_from_backend(
+                experiment_data.backend, qubits
+            )
+        return super()._run_analysis(experiment_data, **options)
+
     def _post_analysis(self, analysis_result: CurveAnalysisResult) -> CurveAnalysisResult:
         """Calculate EPC."""
         alpha = get_opt_value(analysis_result, "alpha")
@@ -157,4 +172,33 @@ def _post_analysis(self, analysis_result: CurveAnalysisResult) -> CurveAnalysisR
         analysis_result["EPC"] = scale * (1 - alpha)
         analysis_result["EPC_err"] = scale * alpha_err / alpha
 
+        # Add EPG data
+        count_ops = []
+        for meta in self._data(label="raw_data").metadata:
+            count_ops += meta["count_ops"]
+        gates_per_clifford = RBUtils.gates_per_clifford(count_ops)
+
+        num_qubits = len(self._physical_qubits)
+        gate_error_ratio = self._get_option("gate_error_ratio")
+        if gate_error_ratio is None:
+            # we attempt to get the ratio from the backend properties
+            gate_error_ratio = self._get_option("error_dict")
+        if num_qubits in [1, 2]:
+            if num_qubits == 1:
+                epg = RBUtils.calculate_1q_epg(
+                    analysis_result["EPC"],
+                    self._physical_qubits,
+                    gate_error_ratio,
+                    gates_per_clifford,
+                )
+            elif self._num_qubits == 2:
+                epg_1_qubit = self._get_option("epg_1_qubit")
+                epg = RBUtils.calculate_2q_epg(
+                    analysis_result["EPC"],
+                    self._physical_qubits,
+                    gate_error_ratio,
+                    gates_per_clifford,
+                    epg_1_qubit=epg_1_qubit,
+                )
+            analysis_result["EPG"] = epg
         return analysis_result
diff --git a/qiskit_experiments/randomized_benchmarking/rb_experiment.py b/qiskit_experiments/randomized_benchmarking/rb_experiment.py
index b253705103..6398bacfab 100644
--- a/qiskit_experiments/randomized_benchmarking/rb_experiment.py
+++ b/qiskit_experiments/randomized_benchmarking/rb_experiment.py
@@ -27,6 +27,7 @@
 from qiskit_experiments.analysis.data_processing import probability
 from .rb_analysis import RBAnalysis
 from .clifford_utils import CliffordUtils
+from .rb_utils import RBUtils
 
 
 class RBExperiment(BaseExperiment):
@@ -46,7 +47,7 @@ def __init__(
         lengths: Iterable[int],
         num_samples: int = 1,
         seed: Optional[Union[int, Generator]] = None,
-        full_sampling: bool = False,
+        full_sampling: Optional[bool] = False,
     ):
         """Standard randomized benchmarking experiment.
 
@@ -162,3 +163,13 @@ def _generate_circuit(
                 rb_circ.measure_all()
                 circuits.append(rb_circ)
         return circuits
+
+    def _postprocess_transpiled_circuits(self, circuits, backend, **run_options):
+        """Additional post-processing of transpiled circuits before running on backend"""
+        for c in circuits:
+            c_count_ops = RBUtils.count_ops(c, self.physical_qubits)
+            circuit_length = c.metadata["xval"]
+            average_count_ops = [
+                (key, value / circuit_length) for key, value in c_count_ops.items()
+            ]
+            c.metadata.update({"count_ops": average_count_ops})
diff --git a/qiskit_experiments/randomized_benchmarking/rb_utils.py b/qiskit_experiments/randomized_benchmarking/rb_utils.py
new file mode 100644
index 0000000000..d4248376f3
--- /dev/null
+++ b/qiskit_experiments/randomized_benchmarking/rb_utils.py
@@ -0,0 +1,264 @@
+# This code is part of Qiskit.
+#
+# (C) Copyright IBM 2019-2021.
+#
+# This code is licensed under the Apache License, Version 2.0. You may
+# obtain a copy of this license in the LICENSE.txt file in the root directory
+# of this source tree or at http://www.apache.org/licenses/LICENSE-2.0.
+#
+# Any modifications or derivative works of this code must retain this
+# copyright notice, and modified files need to carry a notice indicating
+# that they have been altered from the originals.
+
+"""
+RB Helper functions
+"""
+
+from typing import Tuple, Dict, Optional, Iterable, List
+import numpy as np
+from qiskit import QiskitError, QuantumCircuit
+from qiskit.providers.backend import Backend
+
+
+class RBUtils:
+    """A collection of utility functions for computing additional data
+    from randomized benchmarking experiments"""
+
+    @staticmethod
+    def get_error_dict_from_backend(
+        backend: Backend, qubits: Iterable[int]
+    ) -> Dict[Tuple[Iterable[int], str], float]:
+        """Attempts to extract error estimates for gates from the backend
+        properties.
+        Those estimates are used to assign weights for different gate types
+        when computing error per gate.
+
+        Args:
+            backend: The backend from which the properties are taken
+            qubits: The qubits participating in the experiment, used
+            to filter irrelevant gates from the result.
+
+        Returns:
+            A dictionary of the form (qubits, gate) -> value that for each
+            gate on the given qubits gives its recorded error estimate.
+        """
+        error_dict = {}
+        try:
+            for backend_gate in backend.properties().gates:
+                backend_gate = backend_gate.to_dict()
+                gate_qubits = tuple(backend_gate["qubits"])
+                if all(gate_qubit in qubits for gate_qubit in gate_qubits):
+                    for p in backend_gate["parameters"]:
+                        if p["name"] == "gate_error":
+                            error_dict[(gate_qubits, backend_gate["gate"])] = p["value"]
+        except AttributeError:
+            # might happen if the backend has no properties (e.g. qasm simulator)
+            return None
+        return error_dict
+
+    @staticmethod
+    def count_ops(
+        circuit: QuantumCircuit, qubits: Optional[Iterable[int]] = None
+    ) -> Dict[Tuple[Iterable[int], str], int]:
+        """Counts occurances of each gate in the given circuit
+
+        Args:
+            circuit: The quantum circuit whose gates are counted
+            qubits: A list of qubits to filter irrelevant gates
+
+        Returns:
+            A dictionary of the form (qubits, gate) -> value where value
+            is the number of occurrences of the gate on the given qubits
+        """
+        if qubits is None:
+            qubits = range(len(circuit.qubits))
+        count_ops_result = {}
+        for instr, qargs, _ in circuit._data:
+            instr_qubits = []
+            skip_instr = False
+            for qubit in qargs:
+                qubit_index = circuit.qubits.index(qubit)
+                if qubit_index not in qubits:
+                    skip_instr = True
+                instr_qubits.append(qubit_index)
+            if not skip_instr:
+                instr_qubits = tuple(instr_qubits)
+                count_ops_result[(instr_qubits, instr.name)] = (
+                    count_ops_result.get((instr_qubits, instr.name), 0) + 1
+                )
+        return count_ops_result
+
+    @staticmethod
+    def gates_per_clifford(
+        ops_count: List,
+    ) -> Dict[Tuple[Iterable[int], str], float]:
+        """
+        Computes the average number of gates per clifford for each gate type
+        in the input from raw count data coming from multiple circuits.
+        Args:
+            ops_count: A List of [key, value] pairs where
+            key is [qubits, gate_name] and value is the average
+            number of gates per clifford of the type for the given key
+
+        Returns:
+            A dictionary with the mean value of values corresponding
+            to key for each key.
+
+        """
+        result = {}
+        for ((qubits, gate_name), value) in ops_count:
+            qubits = tuple(qubits)  # so we can hash
+            if (qubits, gate_name) not in result:
+                result[(qubits, gate_name)] = []
+            result[(qubits, gate_name)].append(value)
+        return {key: np.mean(value) for (key, value) in result.items()}
+
+    @staticmethod
+    def coherence_limit(nQ=2, T1_list=None, T2_list=None, gatelen=0.1):
+        """
+        The error per gate (1-average_gate_fidelity) given by the T1,T2 limit.
+        Args:
+            nQ (int): number of qubits (1 and 2 supported).
+            T1_list (list): list of T1's (Q1,...,Qn).
+            T2_list (list): list of T2's (as measured, not Tphi).
+                If not given assume T2=2*T1 .
+            gatelen (float): length of the gate.
+        Returns:
+            float: coherence limited error per gate.
+        Raises:
+            ValueError: if there are invalid inputs
+        """
+        # pylint: disable = invalid-name
+
+        T1 = np.array(T1_list)
+
+        if T2_list is None:
+            T2 = 2 * T1
+        else:
+            T2 = np.array(T2_list)
+
+        if len(T1) != nQ or len(T2) != nQ:
+            raise ValueError("T1 and/or T2 not the right length")
+
+        coherence_limit_err = 0
+
+        if nQ == 1:
+
+            coherence_limit_err = 0.5 * (
+                1.0 - 2.0 / 3.0 * np.exp(-gatelen / T2[0]) - 1.0 / 3.0 * np.exp(-gatelen / T1[0])
+            )
+
+        elif nQ == 2:
+
+            T1factor = 0
+            T2factor = 0
+
+            for i in range(2):
+                T1factor += 1.0 / 15.0 * np.exp(-gatelen / T1[i])
+                T2factor += (
+                    2.0
+                    / 15.0
+                    * (
+                        np.exp(-gatelen / T2[i])
+                        + np.exp(-gatelen * (1.0 / T2[i] + 1.0 / T1[1 - i]))
+                    )
+                )
+
+            T1factor += 1.0 / 15.0 * np.exp(-gatelen * np.sum(1 / T1))
+            T2factor += 4.0 / 15.0 * np.exp(-gatelen * np.sum(1 / T2))
+
+            coherence_limit_err = 0.75 * (1.0 - T1factor - T2factor)
+
+        else:
+            raise ValueError("Not a valid number of qubits")
+
+        return coherence_limit_err
+
+    @staticmethod
+    def calculate_1q_epg(
+        epc_1_qubit: float,
+        qubits: Iterable[int],
+        gate_error_ratio: Dict[str, float],
+        gates_per_clifford: Dict[Tuple[Iterable[int], str], float],
+    ) -> Dict[int, Dict[str, float]]:
+        r"""
+        Convert error per Clifford (EPC) into error per gates (EPGs) of single qubit basis gates.
+        Args:
+            epc_1_qubit: The error per clifford rate obtained via experiment
+            qubits: The qubits for which to compute epg
+            gate_error_ratio: Estiamte for the ratios between errors on different gates
+            gates_per_clifford: The computed gates per clifford data
+        Returns:
+            A dictionary of the form (qubits, gate) -> value where value
+            is the epg for the given gate on the specified qubits
+        """
+        epg = {qubit: {} for qubit in qubits}
+        for qubit in qubits:
+            error_sum = 0
+            found_gates = []
+            for (key, value) in gate_error_ratio.items():
+                qubits, gate = key
+                if len(qubits) == 1 and qubits[0] == qubit and key in gates_per_clifford:
+                    found_gates.append(gate)
+                    error_sum += gates_per_clifford[key] * value
+            for gate in found_gates:
+                epg[qubit][gate] = (gate_error_ratio[((qubit,), gate)] * epc_1_qubit) / error_sum
+        return epg
+
+    @staticmethod
+    def calculate_2q_epg(
+        epc_2_qubit: float,
+        qubits: Iterable[int],
+        gate_error_ratio: Dict[str, float],
+        gates_per_clifford: Dict[Tuple[Iterable[int], str], float],
+        epg_1_qubit: Optional[Dict[int, Dict[str, float]]] = None,
+        gate_2_qubit_type: Optional[str] = "cx",
+    ) -> Dict[int, Dict[str, float]]:
+        r"""
+        Convert error per Clifford (EPC) into error per gates (EPGs) of two-qubit basis gates.
+        Assumes a single two-qubit gate type is used in transpilation
+        Args:
+            epc_2_qubit: The error per clifford rate obtained via experiment
+            qubits: The qubits for which to compute epg
+            gate_error_ratio: Estiamte for the ratios between errors on different gates
+            gates_per_clifford: The computed gates per clifford data
+            epg_1_qubit: epg data for the 1-qubits gate involved, assumed to
+            have been obtained from previous experiments
+            gate_2_qubit_type: The name of the 2-qubit gate to be analyzed
+        Returns:
+            The epg value for the specified gate on the specified qubits
+            given in a dictionary form as in calculate_1q_epg
+        Raises:
+            QiskitError: if a non 2-qubit gate was given
+        """
+        epg_2_qubit = {}
+        qubit_pairs = []
+        for key in gate_error_ratio:
+            qubits, gate = key
+            if gate == gate_2_qubit_type and key in gates_per_clifford:
+                if len(qubits) != 2:
+                    raise QiskitError(
+                        "The gate {} is a {}-qubit gate (should be 2-qubit)".format(
+                            gate, len(qubits)
+                        )
+                    )
+                qubit_pair = tuple(sorted(qubits))
+                if qubit_pair not in qubit_pairs:
+                    qubit_pairs.append(qubit_pair)
+        for qubit_pair in qubit_pairs:
+            alpha_1q = [1.0, 1.0]
+            if epg_1_qubit is not None:
+                list_epgs_1q = [epg_1_qubit[qubit_pair[i]] for i in range(2)]
+                for ind, (qubit, epg_1q) in enumerate(zip(qubit_pair, list_epgs_1q)):
+                    for gate_name, epg in epg_1q.items():
+                        n_gate = gates_per_clifford.get(((qubit,), gate_name), 0)
+                        alpha_1q[ind] *= (1 - 2 * epg) ** n_gate
+            alpha_c_1q = 1 / 5 * (alpha_1q[0] + alpha_1q[1] + 3 * alpha_1q[0] * alpha_1q[1])
+            alpha_c_2q = (1 - 4 / 3 * epc_2_qubit) / alpha_c_1q
+            inverse_qubit_pair = (qubit_pair[1], qubit_pair[0])
+            n_gate_2q = gates_per_clifford.get(
+                (qubit_pair, gate_2_qubit_type), 0
+            ) + gates_per_clifford.get((inverse_qubit_pair, gate_2_qubit_type), 0)
+            epg = 3 / 4 * (1 - alpha_c_2q) / n_gate_2q
+            epg_2_qubit[qubit_pair] = {gate_2_qubit_type: epg}
+        return epg_2_qubit
diff --git a/test/test_rb.py b/test/test_rb.py
index df8cb9764c..d0dbb33cdc 100644
--- a/test/test_rb.py
+++ b/test/test_rb.py
@@ -22,7 +22,20 @@
 from qiskit.quantum_info import Clifford
 from qiskit.test import QiskitTestCase
 from qiskit.test.mock import FakeParis
-from qiskit.circuit.library import XGate, CXGate
+from qiskit import QuantumCircuit
+from qiskit.circuit.library import (
+    IGate,
+    XGate,
+    YGate,
+    ZGate,
+    HGate,
+    SGate,
+    SdgGate,
+    CXGate,
+    CZGate,
+    SwapGate,
+)
+from qiskit.providers.aer import AerSimulator
 import qiskit_experiments as qe
 
 
@@ -40,7 +53,7 @@ def test_rb_experiment(self, qubits: list):
         Args:
             qubits (list): A list containing qubit indices for the experiment
         """
-        backend = FakeParis()
+        backend = AerSimulator.from_backend(FakeParis())
         exp_attributes = {
             "qubits": qubits,
             "lengths": [1, 4, 6, 9, 13, 16],
@@ -135,7 +148,7 @@ def test_interleaved_rb_experiment(self, interleaved_element: "Gate", qubits: li
             interleaved_element: The Clifford element to interleave
             qubits (list): A list containing qubit indices for the experiment
         """
-        backend = FakeParis()
+        backend = AerSimulator.from_backend(FakeParis())
         exp_attributes = {
             "interleaved_element": interleaved_element,
             "qubits": qubits,
@@ -157,3 +170,69 @@ def test_interleaved_rb_experiment(self, interleaved_element: "Gate", qubits: li
         self.validate_metadata(exp_circuits, exp_attributes)
         self.validate_circuit_data(exp_data, exp_attributes)
         self.is_identity(exp_circuits)
+
+
+@ddt
+class TestRBUtilities(QiskitTestCase):
+    """
+    A test class for additional functionality provided by the RBExperiment
+    class.
+    """
+
+    instructions = {
+        "i": IGate(),
+        "x": XGate(),
+        "y": YGate(),
+        "z": ZGate(),
+        "h": HGate(),
+        "s": SGate(),
+        "sdg": SdgGate(),
+        "cx": CXGate(),
+        "cz": CZGate(),
+        "swap": SwapGate(),
+    }
+    seed = 42
+
+    @data(
+        [1, {((0,), "x"): 3, ((0,), "y"): 2, ((0,), "h"): 1}],
+        [5, {((1,), "x"): 3, ((4,), "y"): 2, ((1,), "h"): 1, ((1, 4), "cx"): 7}],
+    )
+    @unpack
+    def test_count_ops(self, num_qubits, expected_counts):
+        """Testing the count_ops utility function
+        this function receives a circuit and counts the number of gates
+        in it, counting gates for different qubits separately"""
+        circuit = QuantumCircuit(num_qubits)
+        gates_to_add = []
+        for gate, count in expected_counts.items():
+            gates_to_add += [gate for _ in range(count)]
+        rng = np.random.default_rng(self.seed)
+        rng.shuffle(gates_to_add)
+        for qubits, gate in gates_to_add:
+            circuit.append(self.instructions[gate], qubits)
+        counts = qe.randomized_benchmarking.RBUtils.count_ops(circuit)
+        self.assertDictEqual(expected_counts, counts)
+
+    def test_calculate_1q_epg(self):
+        """Testing the calculation of 1 qubit error per gate
+        The EPG is computed based on the error per clifford determined
+        in the RB experiment, the gate counts, and an estimate about the
+        relations between the errors of different gate types
+        """
+        epc_1_qubit = 0.0037
+        qubits = [0]
+        gate_error_ratio = {((0,), "id"): 1, ((0,), "rz"): 0, ((0,), "sx"): 1, ((0,), "x"): 1}
+        gates_per_clifford = {((0,), "rz"): 10.5, ((0,), "sx"): 8.15, ((0,), "x"): 0.25}
+        epg = qe.randomized_benchmarking.RBUtils.calculate_1q_epg(
+            epc_1_qubit, qubits, gate_error_ratio, gates_per_clifford
+        )
+        error_dict = {
+            ((0,), "rz"): 0,
+            ((0,), "sx"): 0.0004432101747785104,
+            ((0,), "x"): 0.0004432101747785104,
+        }
+
+        for gate in ["x", "sx", "rz"]:
+            expected_epg = error_dict[((0,), gate)]
+            actual_epg = epg[0][gate]
+            self.assertTrue(np.allclose(expected_epg, actual_epg, rtol=1.0e-2))