- A quick summary of how what I do is different to building websites
- 2010-10-07: This is an outline of what I offer that I sent to a web developer who is looking for collaborees, so I would do the Internet marketing bit and he would do the website building bit. After I'd written it, I thought it summarised quite neatly the difference between what I do and what a 'normal' web developer does.
- You'll see that I top and tail the web development process. What I do is crucial right up front when you're deciding what you want your website to be like. And after your website is done, you need to market it .. that's me. So here's what I sent (I've changed the names and subjects):
- OK Geoff, you'll need a comfortable seat and a coffee for this, here's my take
on The Granite Place, Matlock.
Initial design for SEO
- I guess the bottom line is we get to do what the client wants. What
follows is what I would do in an ideal situation.
- The first thing, before you even start to think what you are going to do
with your website is to get the marketing right, and that starts with
demand, so the first thing I would want to do is to look at the number of
people who search on a wide range of keyphrases related to this company's
business, assess competition and (alongside the client) relevance to their
business, then group the keyphrases. I would then want to write a page for
each of those keyphrase groups. That, for me, forms the start of the
structure of a website.
- (The problem with most websites is that they are organised according to
the structure of the business, not according to customer needs. So for
instance, let's imagine you go to a car website and it gives you links to
the Careera, the Poncho, the Cashew and the WhitePurple .. all cars they
manufacture. They assume you know which is the hatchback and which the
4x4. Or, they assume you know which you want. Or you go to a mobile
company and they say do you want PAYG or a contract? Basically, you don't
know until you learn more. Those are organised according to the owner's
deep understanding of their business, but that's not how the customers see
things. They may have just started thinking they need a new car and may
not have chosen yet whether they want a diesel or petrol, or a saloon v a
hatchback. You need a new phone, but whether you go for a contract or PAYG
surely depends on the phone you like and the relative costs.
- Often you can start with lifestyle, or develop a short flowchart to help
guide people to the product that would suit them, so for fireplaces you
might, from gas fires, ask .. modern or traditional styling .. it has to
fit in a room so that's already decided.
- You can end up with menu options that mean nothing to the customer. The shop owner will have to help me with those phrases and topic, and if you confuse your
customers, you're dead, so on your example website I would have something to explain each phrase to help the customer know what they've got and what they need. If we can guide people through their needs to our solutions and turn a list of 50 choices into a choice of three or four, that's more like it.
- And that explanation place becomes something the search engine picks up
whenever someone wants help with that, and if you're helpful, you might
get the business.)
- So, in principle, we are writing pages about the relevant things that
clients are searching for.
- Let's assume for a minute that that's enough to get you a first page
position and some traffic from natural search.
Improving your conversion rate (sell more with the same traffic)
- The next thing you want is for your website visitor to buy something from
you, the percentage who do is your conversion rate.
- So, how do we improve conversion?
- (Everyone wants more traffic, very few are looking at conversion .. it's not right. Conversion begets traffic. If you convert, you're doing a great job. People will beat a path to your door. If you're not converting, more traffic just disappoints more people).
- Usability is important, so a usability test is important using real
users. It's fairly cheap to do and usually brings out a load of things
that could be better that you and I and your client would never have
noticed. I include accessibility and internationalisation in this.
- Psychology: we should try to appeal to our visitors' subconscious as
well as their conscious intellect. So we should provide pictures of food
and/or attractive and happy people and probably include a small amount of
facts, figures, comparison charts and so on for our intellect.
There are a number of persuasion possibilities too, I'll pick three at
random: social proof: testimonials, Facebook/Twitter pages with good
comments and company responses and a fair number of fans; commitment: get
someone to give you their contact details and they are implying that they
trust you enough. If they've done that, you are halfway to selling to them, so give them small ways in .. download a PDF guide to all the
different product types, for instance, or encourage them to follow or
friend you on one of the social media networks. Scarcity: if you mark
something up as your most popular model and show there's only one left,
that will encourage sales.
- We should, of course, cover the basics: a competitive price, delivery,
guarantee, and add-on services such as fitting, maintenance and support.
- Copy: Text is persuasive. You need good copy because that's what
persuades people to buy and because that's partly what Google uses when it
decides when and where to display you in the search results. Combining the
two is a skill. You need someone good on that just as much as you need
professional, not amateur photographs (unless amateur is part of your branding).
- Video. It's still the next big thing. Check out this and this if you think you need an ultra professional five-figure setup before you make a start.
- Interact: This isn't about setting up a shop and just taking people's
money. Their questions need to be answered (and then organised into a real
FAQ). There's a basic Internet principle that says we should provide our
content in the way the user wants it. So if they like Twitter, we should
be on Twitter. If they want email, we'll write back to them by email. If
they want a live video interaction on Skype .. hell, why not? Social media
is partly like the phone and fax used to be. But social media is also
public. Answer a question on Twitter and you don't just impress the person
you're tweeting with, you impress the crowd of twenty gathered around,
- Information architecture: basically, how information is gathered, used
and presented to the benefit of everyone. In ecommerce, it's about
gathering information about your business and using it to help future
customers. Which of these products is the most popular? Which do young people
like? What's new? Which is most fuel efficient? Which do people add to
their wishlist the most? Can I share my wishlist on Facebook and get my
friends' inputs? Can I organise these products by price and unselect them
by category until I've a shortlist of two or three?
- Blogging: While we are at it, we should work out what the news and
information opportunities are, decide whether anyone in the business is
going to have the will, the time, and the skills to blog, Tweet and so on,
or will they need someone to do it, or are we not going to bother (bad
- All of this, ideally, is to be considered right at the start.
Build your website
- Having answered all that, you get to build the site.
After you've built your website ..
- Then post-build, we are into:
- coaching/training in social media/blogging and systems setup: there's a
guy in Bridlington who runs a B&B and moans about his clients. There's a
right way and many wrong ways to do this and there's a lot to it, so
clients need coaching and that's ongoing until it's right.
- usability testing, watching Analytics, running tests and continual
improvement: once you see the real traffic flowing through your site it
should inform improvements. Someone who knows what to look for needs to
watch that and make suggestions. If, for instance, searches on the phrase
'granite worktop' convert better than 'marble worktop' then maybe we should blog
more on granite worktops. Or check the prices against competitors. And maybe
check our product ranges. And our online copy.
You can (and should) set up tests. Pages convert at a particular rate. If
your granite worktop page converts at, say, 2%, wouldn't it be nice if it
converted at 4%? By testing different headlines, layouts, photographs and
so on, you can iterate your way to better conversion for the whole
website. Testing is ongoing and forever.
- Link building: The website that comes top in your industry for your main keyphrase has 608 pages of its own and the home page has 80
inbound links. It's been around since 2004 so let's take an initial guess
and say that it's currently adding 2 pages a week and one or two inbound
links a month. An active blog with two entries per week and an equally
active programme of fleshing out website content wherever traffic shows
there is opportunity and need would allow us to catch up with them over
the medium term and gain us the search position we want.
- in the meantime, you can buy traffic using pay per click (PPC)
advertising, usually Adwords or Facebook but other places too. You
probably know how that works, but basically it's not simple. The price you
pay in Adwords is partly to do with what you bid, but it's also your
quality score, and that is partly to do with the closeness of match
between the search term, the ad text, and the landing page text. So
basically you need a specific landing page and ad for each search term.
Also, you need a fast site and a low bounce rate. Once you are over all
that, you can buy specific traffic and watch, say, 200 go to a specific
page on your site and then decide whether it's converting at a good enough
rate for you. If so, fine, press on with SEO stuff (which takes time,
effort and money), if not, change something and go again. Basically, use
PPC to check the efficacy of your website's conversion mechanisms for
particular search terms.
- if you like, you can think about competitions and ways to engage people
and go for viral marketing effects through social media. That may not be
totally appropriate for this client, but I mention it generally.
- OK, so what this boils down to is there should be a lump of money for
marketing input to the initial concepts and requirements of the website.
Then you build the website with a little bit of input from Internet
marketing, more if I'm writing the copy. Then there's ongoing work on all
the things above .. blog coaching, PPC, link building, testing, traffic
monitoring and so on.
- It's a lot. It may put off some clients. But here's the thing. Actually
it's one of my favourite life rules: everything is as complex as it can
humanly be. It's the natural order of things. Kicking a ball into the back
of a net is a lifetime's work once you want to make it your profession.
There are ten natural search slots on page one of Google for any search
term. If it's a moneymaking term, you bet there's competition. 42% of
traffic goes to the person in the top position. That person will defend
their livelihood with all the passion they can bring to it, and they'll
stop you taking the food from their children's mouths if they can. It's a fight. If
you want to make money online, you need to throw some resources at it to
dislodge an incumbent or two. Once you're there, it's easier to defend.
How much resource is needed? That depends on the profit from the search
- So if you are working for small companies, then we can start with small
traffic, specific long-tail search terms and build up a flow of small-beer
positions which gradually add up to something worthwhile.
- If you are working for larger companies, then the budget should reflect
that. I've seen companies spending £10m a month on Internet marketing.
Most often the bulk of that, wrongly IMHO, is on PPC.
- I offer everything I've spoken about here.
- I haven't spoken about email marketing .. I haven't done much of that,
it's not really my thing, but I suppose if you've got someone on board who
knows about that, fine, just keep the costs in proportion with all the
other things mentioned here and remember that ads and promotions like this
are not long term, whereas SEO, link building and social media connections
- Let me know what you think :-)
- All the best
By John Allsopp
Previous: Here's how not to blog (IMHO) -
Back to blog list