diff --git a/README.md b/README.md
index 1a7044e97bc5cbab2340d78e9d5997012a5bdddf..d2d9dba06b498eea11b7438d8d39b658fa0166e6 100644
--- a/README.md
+++ b/README.md
@@ -1,6 +1,26 @@
 # kmatch
-- on the scicore ubuntu system, you do `source ubuntu_compile` to execute the compile command therein
-- on other systems: good luck
-- usage examples are in the two Python scripts. Make sure to read the comments therein before executing
-- before you can execute the scripts with Python, you need to load the exact same Python module as for compiling
 
+Simple k-mer counting for ultrafast screening of large sequence databases in Python. You're at the wrong place if you look for highly sensitive sequence searches. Usage examples:
+
+* create_db.py - Builds search index for user defined k-mer size k.
+* search_db.py - Search previously built index. The index is optionally loaded in memory to speed up repeated searches on the same index.
+
+Setup
+-----
+Dependencies:
+* A recent c++ compiler supporting C++11
+* Python 3
+* pybind11 (https://github.com/pybind/pybind11)
+
+Compile:
+```
+g++ -O3 -Wall -shared -std=c++11 -fPIC $(python3-config --includes) -I<path_to_pybind11>/include kmat
+ch.cpp -o kmatch$(python3-config --extension-suffix)
+```
+
+Tell Python where to find the library:
+```
+export PYTHONPATH=<path_to_compile_dir>:$PYTHONPATH
+```
+
+Test it with the provided example scripts!
\ No newline at end of file