First all - love these tips. Many are in line with what I do. I want to add my own variation on this. I'll continue the numeration that the last post did.
As has been pointed out, it is most important to get the top three rows. Also as pointed out there are sometimes point advantages not to focus on one column first and then do the next column.
6) Now I start the game assuming I will fill in all of the first three rows. For those rows, when I pick tiles and place them I don't care if they are beside each other. That means initially my top half of the board looks crazy as I place tiles randomly, and I have low points. But then I do the fillers and my points shoot right up. So yes - try to place the top half so that you are grouping together. But don't be afraid to grab a 3 pile and place in the third row even if it touches nothing. Ideally by round 5 all three rows are filled and you have fully caught up (or surpassed) in points. Typically I only manage to get two rows, but three is not uncommon.
7) Now for rows 4/5 and going for complete columns or complete colours. For the column you only need to consider row 4/5. Just assume you will get 1/2/3 by the end. So when you choose what column to go for only look at 4/5. For me the "choice" of the column is usually opportunistic. Are there a low of factories with 2 tiles of the same colour - good chance you can get row 4 with two pickups. Is there a factory with 3 tiles and you cannot use it for you row 3 - put it in row 5. Can you seed the middle with a color nobody else can take a large pile of, then let it grow and you might be able to do row 4 or 5 in one pickup (rare but very satisfying).
8) Once you have one column you have a choice - do you go for a second column or for a color. If you get the column right beside, you have a chance of getting both. But remember - getting rows 1-3 is more important. Only on the last round I might skip a row 1-3 if it means I can complete a colour or column.
8a) So many people do something like at the beginning of the game they pick two yellows, put them in row 2. Having put something in the third column they then feel like they have committed themselves to taking 4 whites and 5 blacks for rows 4/5. They will skip better choices for row 4/5 because they feel like they committed to a specific column on their first pickup. Instead you should make your choice for row 4/5 completely independent of what row 1-3 spaces you happened to have filled in during the first round.
Other tips:
9) Ideally you need to do two row 4 and two row 5 in five rounds, or to rephrase you should do each of those rows every two rounds. Don't rush even if there is a good opportunity. Lets say you have two red on row 5, you start the round with first pick and there is factory with three red. Pick up those three red and put them in row 3. It is so tempting to complete the hard-to-do row 5. But row 3 is more important. Look at it this way: Normally you will either need to do two pickups to do row 3 or two pickups to complete row 5. If there is an opportunity to do something in one pickup, do row 3 first. Because you NEED to complete row 3 this round and you don't need to complete row 5 this round (barring last round).
10) Sometimes there will be something you need, lets say 2 blues. It is available. But look around at the other player's tables. Are any of them likely to pick up those two blues? There are times that you can take a justified risk and pick something else this round, knowing that the two blues will still be there next time it is your turn. Often when I see someone else make a non-obvious pickup I realize it is because they expect to the thing they skipped to still be there next round (and there is an opportunity for you to thwart them).
11) Sometimes there is value in taking a risk. Lets say you need 2 blues. There is one in the middle and one in a factory. The safe thing is to to pick the one in the middle this round, and the one in the factory next round. A riskier play is to pick a different colour from that factory that you need, thereby throwing the second blue in the middle. If you are reasonable sure no one else will take it, then you can get them both in one pickup. Seeding the middle to pick up what you want is a risk but it can pay off. Why spend two moves to pick up a total of two tiles, when you can spend those two moves to pick up three tiles.
12) Remember that picking up more tiles is generally better than fewer tiles. Ideally row 1 and 2 is always a single pickup. If you can keep First Token you can do row 3 in a single pickup at least once or twice in the game, at most two. Row 4 is ideally 2 pickups, Row 5 is 2-3.
13) A three player game has 7 factories, 28 tiles. Evenly distributed that is 9.3 tiles each. Two player games have 10 tiles each. Four player games have 9 each. So during a round if you pick up more than 9 tiles you are ahead, as generally more tiles picked up means more tiles placed.
13a) There are 15 places (empty spots) to put tiles that you pick up (left side of the board). Be aware, lets say you start a round and there are only 8 empty spaces on the left side of your board. You are in danger of having a full board and getting too many tiles.
14) Broadly speaking during a round you want as many pickups as you can. By picking up from the middle you do not combine pile, which can extended the total number of pickups and allow you to pick up more times. By picking from a factory, you are combining multiple piles of a certain colour which reduced the total number of pickups and also makes the pile richer for other players. While there are many specific reasons not do to this, if everything else is equal then pick from the center rather than a factory. Later in the round you can calculate the most likely number of pickups left and you can see if picking from a factory or the center will result in you getting another pickup or not. Even more, calculating this way is key to figuring out who is going to get a pile of 7 tile in the center that they have to throw on the floor - and whether you can influence who gets it.
As has been pointed out, it is most important to get the top three rows. Also as pointed out there are sometimes point advantages not to focus on one column first and then do the next column.
6) Now I start the game assuming I will fill in all of the first three rows. For those rows, when I pick tiles and place them I don't care if they are beside each other. That means initially my top half of the board looks crazy as I place tiles randomly, and I have low points. But then I do the fillers and my points shoot right up. So yes - try to place the top half so that you are grouping together. But don't be afraid to grab a 3 pile and place in the third row even if it touches nothing. Ideally by round 5 all three rows are filled and you have fully caught up (or surpassed) in points. Typically I only manage to get two rows, but three is not uncommon.
7) Now for rows 4/5 and going for complete columns or complete colours. For the column you only need to consider row 4/5. Just assume you will get 1/2/3 by the end. So when you choose what column to go for only look at 4/5. For me the "choice" of the column is usually opportunistic. Are there a low of factories with 2 tiles of the same colour - good chance you can get row 4 with two pickups. Is there a factory with 3 tiles and you cannot use it for you row 3 - put it in row 5. Can you seed the middle with a color nobody else can take a large pile of, then let it grow and you might be able to do row 4 or 5 in one pickup (rare but very satisfying).
8) Once you have one column you have a choice - do you go for a second column or for a color. If you get the column right beside, you have a chance of getting both. But remember - getting rows 1-3 is more important. Only on the last round I might skip a row 1-3 if it means I can complete a colour or column.
8a) So many people do something like at the beginning of the game they pick two yellows, put them in row 2. Having put something in the third column they then feel like they have committed themselves to taking 4 whites and 5 blacks for rows 4/5. They will skip better choices for row 4/5 because they feel like they committed to a specific column on their first pickup. Instead you should make your choice for row 4/5 completely independent of what row 1-3 spaces you happened to have filled in during the first round.
Other tips:
9) Ideally you need to do two row 4 and two row 5 in five rounds, or to rephrase you should do each of those rows every two rounds. Don't rush even if there is a good opportunity. Lets say you have two red on row 5, you start the round with first pick and there is factory with three red. Pick up those three red and put them in row 3. It is so tempting to complete the hard-to-do row 5. But row 3 is more important. Look at it this way: Normally you will either need to do two pickups to do row 3 or two pickups to complete row 5. If there is an opportunity to do something in one pickup, do row 3 first. Because you NEED to complete row 3 this round and you don't need to complete row 5 this round (barring last round).
10) Sometimes there will be something you need, lets say 2 blues. It is available. But look around at the other player's tables. Are any of them likely to pick up those two blues? There are times that you can take a justified risk and pick something else this round, knowing that the two blues will still be there next time it is your turn. Often when I see someone else make a non-obvious pickup I realize it is because they expect to the thing they skipped to still be there next round (and there is an opportunity for you to thwart them).
11) Sometimes there is value in taking a risk. Lets say you need 2 blues. There is one in the middle and one in a factory. The safe thing is to to pick the one in the middle this round, and the one in the factory next round. A riskier play is to pick a different colour from that factory that you need, thereby throwing the second blue in the middle. If you are reasonable sure no one else will take it, then you can get them both in one pickup. Seeding the middle to pick up what you want is a risk but it can pay off. Why spend two moves to pick up a total of two tiles, when you can spend those two moves to pick up three tiles.
12) Remember that picking up more tiles is generally better than fewer tiles. Ideally row 1 and 2 is always a single pickup. If you can keep First Token you can do row 3 in a single pickup at least once or twice in the game, at most two. Row 4 is ideally 2 pickups, Row 5 is 2-3.
13) A three player game has 7 factories, 28 tiles. Evenly distributed that is 9.3 tiles each. Two player games have 10 tiles each. Four player games have 9 each. So during a round if you pick up more than 9 tiles you are ahead, as generally more tiles picked up means more tiles placed.
13a) There are 15 places (empty spots) to put tiles that you pick up (left side of the board). Be aware, lets say you start a round and there are only 8 empty spaces on the left side of your board. You are in danger of having a full board and getting too many tiles.
14) Broadly speaking during a round you want as many pickups as you can. By picking up from the middle you do not combine pile, which can extended the total number of pickups and allow you to pick up more times. By picking from a factory, you are combining multiple piles of a certain colour which reduced the total number of pickups and also makes the pile richer for other players. While there are many specific reasons not do to this, if everything else is equal then pick from the center rather than a factory. Later in the round you can calculate the most likely number of pickups left and you can see if picking from a factory or the center will result in you getting another pickup or not. Even more, calculating this way is key to figuring out who is going to get a pile of 7 tile in the center that they have to throw on the floor - and whether you can influence who gets it.
