A follow-on to Peter's comment is that the markers tested for CODIS/NDIS compatibility are autosomal STRs, simple tandem repeats, not the SNPs (single nucleotide polymorphisms) that are tested by all other companies offering genealogy/ancestry services. And the very reason that autosomal STRs are used for forensic identification is why they are, for all intents and purposes, useless for genealogy: these allele repetitions are highly volatile and often change within a single generation. Two full siblings are unlikely to report identically on even these 13 CODIS marker; two 1st cousins theoretically never. Very helpful in distinguishing between two individuals in a court of law, but of no value at all for genealogy.
HomeDNA claims to, in addition to the STRs, also test "100 DNA markers" for their starter ancestry test, and 80,000 markers for their advanced test. What these markers are, precisely, they never define. Presumably--hopefully--these would be population-specific SNPs. The other genealogy companies test 600,000 to 650,000 SNPs for 45% of the HomeDNA $149 advanced test price.
The company in its various iterations was discussed last year in this G2G thread (one of the rare times I guess I wasn't logged-in). The founder and the person upon whose science these tests are based is Eran Elhaik. You'd be hard-pressed to find many in the genetics community who have respect for his work.
I hate it when people pay for a service with a reasonable expectation, but receive something that, while not falsely advertised in a strict sense, turns out not to be fit for the expected purpose. My best advice for DNA testing is to thoroughly research the company/service; if in doubt ask someone whom you believe knowledgeable in the area; and if in doubt stick to the largest companies who have demonstrably sizable databases of customers.